UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Thomas Henry Huxley's agnostic philosophy of science Byun, Jiwon 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2017_september_byun_jiwon.pdf [ 1.72MB ]
Metadata
JSON: 24-1.0354412.json
JSON-LD: 24-1.0354412-ld.json
RDF/XML (Pretty): 24-1.0354412-rdf.xml
RDF/JSON: 24-1.0354412-rdf.json
Turtle: 24-1.0354412-turtle.txt
N-Triples: 24-1.0354412-rdf-ntriples.txt
Original Record: 24-1.0354412-source.json
Full Text
24-1.0354412-fulltext.txt
Citation
24-1.0354412.ris

Full Text

  Thomas Henry Huxley’s Agnostic Philosophy of Science  by  Jiwon Byun M.A., The University of Chicago, 2009 B.A., The University of Wisconsin, Madison, 2007    A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE DEGREE OF Doctor of Philosophy in The Faculty of Graduate and Postdoctoral Studies (Philosophy)   THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2017  © Jiwon Byun, 2017  ii  ▌ABSTRACT    This dissertation examines Thomas Henry Huxley’s notion of agnosticism and its bearings on his conception of science. Although agnosticism is commonly regarded as a position that recognizes the limits of human knowledge, Huxley – who coined the term “agnostic” – characterized it as more than a theory of ignorance or limits. I argue that Huxley intended his agnosticism to be a guide to knowledge that can work regardless of our ignorance or limits. To this end, I draw attention to Huxley’s less famous philosophical works. I examine his discussions of Descartes to show that he had an epistemological project and to clarify the structure of agnosticism; I analyze his Hume to illuminate the reasoning behind his claim that verification is the only justificatory method and to highlight his reasons for situating agnosticism within what he called “modern critical philosophy”; I investigate his other essays to argue that his agnosticism concerns a claim to knowledge and should not be understood as ethics of belief.  Based on his epistemological inquiry, Huxley offered a quick guide to knowledge, consisting of an account of legitimate evidence and an ethics of knowing: agnosticism. It can be summarized as follows. Propositions concerning anything beyond phenomena lack evidential value; verified propositions have evidential value; if one wishes to make a claim about the knowledge status of a proposition, one should evaluate the evidence and be honest about the result without further pretension. Huxley discussed the realm of ignorance to show its lack of justificatory value. The signature remark of Huxleyan agnostics is “Show me evidence,” rather than “I don’t know.”  This interpretation undermines the widely accepted view that Huxley’s endorsement of agnosticism poses philosophical obstacles to his larger project of promoting science in Victorian society. His intention behind agnosticism was to establish and maintain epistemic merit of science without any unknowable, metaphysical or theological, apparatus. Science is the practice of agnosticism, and for this reason, our best way to knowledge. Our understandings of his life-long project and of the growth of science’s autonomy during the 19th century would remain incomplete without due appreciation of this notion of agnosticism.         iii  ▌LAY SUMMARY    Although agnosticism is commonly regarded as a religious position on the existence of God, the coiner of the term, Thomas Henry Huxley (1825-1895), had more to say. This dissertation presents a more comprehensive understanding of agnosticism and its bearings on the conception of science by examining Huxley’s underappreciated philosophical works. Agnosticism was a guide to knowledge that tells us what to count as evidence and what to do with evidence; his discussed the realm of ignorance to show its lack of evidential value; he understood science as the practice of agnosticism, and for this reason, as our best way to knowledge. This dissertation shows that Huxley’s meta-scientific views deserve philosophical attention: science is metaphysic-agnostic, not metaphysics-free; science needs faith in the sense of trusting and risking. It also leads us to re-visit our preconceptions: agnosticism is not a passive, noncommittal or wishy-washy attitude, but a constructive, firm and confident position.     iv  ▌PREFACE    This dissertation is original, unpublished, independent work by the author, Jiwon Byun.    v  ▌TABLE OF CONTENTS    Abstract ............................................................................................................................................ ii  Lay Summary .................................................................................................................................. iii  Preface............................................................................................................................................. iv  Table of Contents ............................................................................................................................. v  Acknowledgements ........................................................................................................................ vii  Dedication ..................................................................................................................................... viii  Chapter 1.  Introduction .................................................................................................................. 1  1.1.  A question for agnostics ........................................................................................................ 1 1.2.  Huxley and Victorian scientific community .................................................................... 5 1.3.  Possible reasons for not paying philosophical attention to Huxley ........................ 11 1.4.  Structure of the dissertation ................................................................................................ 17  Chapter 2.  The Reception of Huxley’s Agnosticism ................................................................... 20  2.1.  A science defender’s nescience: a puzzling position ................................................. 20 2.2.  Partial approach to Huxley’s agnosticism: the common approach ........................ 23 2.2.1.  Type one: an incidentally conjoined position .............................................. 24   2.2.2.  Type two: an application of scientific method ............................................ 28  2.2.3.  Type three: an intellectual code ..................................................................... 30 2.3.  Huxley’s promotion of agnosticism ................................................................................ 33 2.4.  Inclusive approach to Huxley’s agnosticism ................................................................ 39  Chapter 3.  Huxley’s Cartesian Project ......................................................................................... 43  3.1.  Dockrill’s account of Huxley’s agnosticism ................................................................ 44 3.2.  The importance of Huxley’s discussions of Descartes .............................................. 48  3.3.  Two paths that Descartes opened .................................................................................... 51 3.3.1. The first path: legitimate Idealism ................................................................ 51   3.3.2. The second path: legitimate Materialism ..................................................... 55 3.4.  Huxley’s reshaping of Descartes’s project ................................................................... 60  vi  Chapter 4.  Huxley’s Epistemology: Memory, Verification, and Physical Thinking ................... 74   4.1.  How to identify knowledge ................................................................................................. 75 4.1.1. Huxley’s Hume ................................................................................................ 76 4.1.2. Anatomy of mind ............................................................................................. 79 4.1.3. Memory ............................................................................................................. 86 4.1.4. Memory as a model of knowledge ................................................................ 90 4.2.  Huxley’s defense of the method of verification ............................................................ 96 4.2.1.  Metaphysicians’ appeal to necessity ............................................................. 96 4.2.2.  Religious thinkers’ appeal to faith............................................................... 108 4.3.  Physical thinking as a constituent of a guide to knowledge .................................... 113  Chapter 5.  Agnosticism as a Guide to Knowledge .................................................................... 120  5.1.   The two paths in agnosticism: an account of legitimate evidence ....................... 120 5.2.   Another component of agnosticism: the agnostic principle ................................... 129 5.3.   Huxley’s agnosticism and Clifford’s ethics of belief ............................................... 135 5.4.   How to follow the agnostic principle ............................................................................ 156 5.5.   Agnosticism in the tradition of modern critical philosophy ................................... 173 5.6.   Chapter summary ................................................................................................................ 185  Chapter 6.  Science and Faith ..................................................................................................... 186  6.1.  Huxley’s incapability, insistence, and inconsistence ................................................ 186  6.2.  Huxley’s conception of science ....................................................................................... 191 6.2.1.  Relying on principles and relying on the principles of science ............... 191 6.2.2.  Anti-realistic conception of science ............................................................ 204 6.2.3.  Science as practice of agnosticism .............................................................. 219 6.3.  Huxley’s conception of faith ............................................................................................ 226 6.3.1.  An agnostic’s faith ......................................................................................... 227 6.3.2.  A critique of blind faith................................................................................. 231 6.4.  Chapter summary ................................................................................................................. 235  Chapter 7.  Conclusion ................................................................................................................ 237  References ................................................................................................................................... 249   vii  ▌ACKNOWLEDGEMENTS     When I wondered to myself about doing a PhD, I did not know that huge support, motivation and inspiration from others awaited me. It is hard to express gratitude that I feel. Here is my try.              I am most grateful to my supervisor, John Beatty. Although I have countless things to say, I note here only one. Thanks to John, I was able to have a variety of experiences which will continue to have influences on my way of thinking and living. I do not simply mean that John gave me valuable opportunities for doing new things; John seems to have some mysterious power of turning a single, ordinary experience into multiple, strange experiences. This dissertation is a small fraction of the experiences made possible and richer by John.  Everyone knows that my research committee is amazing: Margaret Schabas and Alan Richardson. Their helpful comments and advice shaped my drafts into a dissertation. Not only that, my work and life benefited from Margaret’s sharp remarks and new perspectives and Alan’s black humour and deep concerns about doing philosophy. I owe them much thanks. By good fortune, I happened to have a wonderful examination committee: Robert Brain, James G. Lennox and Chris Stephens. Their questions and feedback helped me improve this dissertation and develop further ideas. I deeply appreciate their time and good words.      I wish to thank all members of UBC Philosophy. My special thanks to Sylvia Berryman, Eric Margolis and Chris Stephens for their support and academic advice; to Rhonda Janzen, Kate Lewis and Tim Son for their administrative aid which made my life far easier.           Many thanks to my dear colleagues for making this journey merrier: Tyler DesRoches, Serban Dragulin, Emma Esmaili, Sina Fazelpour, Chris French, Kinley Gillette, Jihee Han, Cosima Herter, S. Andrew Inkpen, Katie Joel, Alirio Rosales, Mojtaba Soltani, Servaas van der Berg and Gerardo (Jerry) Viera. I especially thank Emma, Jerry and Serban for their traditional, cool, and dark support (respectively) and various, on-going, help which started in 2009.   I thank Geunjae, Gyumin, Hyunkyung, Jaejun, Kkotchssi and Minyoung for being special audiences for my presentation; Minsun for being such a perfect companion for my Huxley trip.  Finally, I am very grateful to Anne Barrett of the Archives and Corporate Records Unit at Imperial College London for arranging for me to visit and read Huxley manuscripts.  viii     To my parents    1  ▌CHAPTER 1  Introduction    1.1. A question for agnostics  In 1869, Thomas Henry Huxley (1825-1895) coined the term “agnostic,” because he could not find a proper label for his philosophical position.1 Since then, the term has quickly gained currency among both scholars and non-scholars. It is now further used to describe more than a position that a person may take. For example, in computer science, a program is described as “agnostic” when it is compatible with various operating systems.2     The most salient feature of agnostic position would be, ironically, refusal to take a position regarding the truth value of a given proposition. This shows that the mere absence of a belief or opinion does not sufficiently capture the state of being agnostic because it suggests that the state involves refusing.3 Consider, for example, the issue of the existence of a god, since                                                           1 The year is known based on Huxley’s own story. In 1869, Huxley did not use the terms “agnostic” and “agnosticism” in published essays and lectures. Later in 1889, Huxley explained how he had come to coin the term as follows:  This was my situation when I had the good fortune to find a place among the members of that remarkable confraternity of antagonists, long since deceased, but of green and pious memory, the Metaphysical Society. Every variety of philosophical and theological opinion was represented there, and expressed itself with entire openness; most of my colleagues were -ists of one sort or another; and, however kind and friendly they might be, I, the man without a rag of a label to cover himself with, could not fail to have some of the uneasy feelings which must have beset the historical fox when, after leaving the trap in which his tail remained, he presented himself to his normally elongated companions. So I took thought, and invented what I conceived to be the appropriate title of ‘agnostic.’ It came into my head as suggestively antithetic to the ‘gnostic’ of Church history, who professed to know so much about the very things of which I was ignorant; and I took the earliest opportunity of parading it at our Society, to show that I, too, had a tail, like the other foxes. ... That is the history of the origin of the terms ‘agnostic’ and ‘agnosticism’” (1889e, 239-40, his italics).  2 Oxford English Dictionary Online (3rd ed., December 2016). Also, some mathematical scientists have used the term “agnostic” to indicate the possibility of using a data analysis tool without an understanding of the structure of a phenomenon at issue (Napoletani et al. 2011).      3 However, in the literature of epistemology, the state of being agnostic is sometimes explained in terms of the state of having no belief, as opposed to the state of taking a position or a stance. See Friedman 2013 and Yoder 2013 for critical discussions of an account of agnostic position as non-belief.    2  agnosticism is most commonly thought of as a religious position. Holding an agnostic position about this issue means refusing to take both positions that a god exists and that a god does not exist. In this context, agnosticism is generally introduced as a middle position between theism and atheism.4 Agnostics neither affirm nor deny the existence of a god; they are not those who simply lack a belief, opinion or interest regarding the issue.  Why would one take an agnostic position toward the existence of a god? It has been said that agnostics are cowards. Richard Dawkins introduces a view of a preacher from his old school. An agnostic is a “namby-pamby, mushy pap, weak-tea, weedy, pallid fence-sitter.”5 The basic idea is that agnostics are too timid to take a side on the existence of a god. By taking an agnostic position, one can escape from implications or consequences of taking either position. For example, agnostics want to avoid the charge of being a disbeliever, or they do not want to give an impression that they are arrogant or cocksure.  Agnostics may be cowardly in another sense. According to one of Huxley’s contemporaries,  [t]here is an agnosticism which is simply the cowardly escaping from the pain and difficulty of contemplating and trying to solve the terrible problems of life by the help of the convenient phrase, “I don’t know,” which very often means “I don’t care.”6  Agnostics, then, refuse to take a position because they want to refuse a (negative) connotation associated with each position or because they want to refuse to go through some intellectual struggle. Huxley, who lived in Victorian society under the Anglican establishment, was not free                                                           4 Flint 1903; Woelfel 1998; Le Poidevin 2010; Dawkins 2006; Smart 2011. 5 Dawkins 2006, 69.  6 Magee 1889, 89.   3  from those charges.7 To religious critics, agnostics were just “infidels” who were afraid of the old name; or, agnostics were “freethinkers who had yet to learn to think.”8 An agnostic position is a noncommittal and self-serving position for cowards. In this way, Huxley’s coinage of “agnostic” has been sometimes explained.9   This account of “cowardly agnosticism” would not be satisfactory as a general account of why one would take an agnostic position. Of course, the account may be the most accurate description of some particular agnostics. Also, there would exist those who deliberately put forward an agnostic position in a diplomatic manner for some reasons. The account may turn out to be the most plausible (psychological) account, if an agnostic stance only concerns culturally significant issues like the existence of God.  However, there is also a reason that agnostics themselves have offered. Agnostics neither affirm nor deny, for example, the existence of a god, because of the nature of a given issue such as its insolubility or unknowability. The Oxford English Dictionary states that an agnostic is “a person who believes that nothing is known or can be known of immaterial things, especially of the existence or nature of God.”10 Agnostics refuse to take a position because they think that a given issue belongs to insoluble matters or because they think that the existence of God can be                                                           7 Egil Asprem briefly describes one aspect of the domination of the Anglican establishment: “[r]eligious tests were, for example, required for anyone wishing to obtain a position at the most prestigious British universities (i.e., Oxford, Cambridge, Durham) until 1871, when the University Test Act was passed. In practice these tests were designed to exclude Roman Catholics …, but also non-Christians and non-believers were affected by the requirement, and forced out of top institutions of education and research for purely theological reasons” (2014, 290). 8 For example, Wace 1888, 9, 10; Magee 1889. For more on the immediate reception of agnosticism, see Irvine 1968, 322-30; Jensen 1991, 118-25. Also see Huxley 1863b.  9 For example, Gavin Hyman writes as follows: “Thomas Huxley, for instance, was unhappy with ‘atheism’ because it was too dogmatic; ... [f]urthermore, it seems that ‘atheism’ was also being increasingly linked with far-left revolutionary politics, which further tainted the term in polite society” (2006, 30). See also Lightman 2015a. 10 OED Online (3rd ed., December 2016).  4  neither proved nor disproved.11 That is, an agnostic position is based on a general epistemological position to the effect that there are some subject matters about which human beings cannot have knowledge.12 Such an epistemological view of our ignorance, commonly labeled as “agnosticism,” leads one to an agnostic position about a specific issue.   An agnostic position, then, seems different from a skeptical position. Being agnostic implies a state of not knowing or being incapable of making a truth-value judgment, whereas being skeptical implies an active inquiry such as posing a question or casting doubt.13 When one takes a skeptical stance, she usually offers reasons to show that a given claim is doubtful; when she takes an agnostic stance, she would explain why it is not knowable. For example, we may say, a judge is skeptical about an indictment because submitted evidence is not conclusive, but thinks that there is no point of being skeptical about the existence of a god because the existence of a god is not knowable or the question regarding its existence is insoluble. In short, the judge was skeptical about the validity of the indictment but agnostic about the existence of a god.  Philosophical reasons for taking an agnostic position invite further questions regarding one’s epistemology. For example, we can ask why agnostics think that we cannot have knowledge about some subject matters such as the existence of a god. We expect that agnostics would have some views on what it means to know something and what specifically we cannot know. Similarly, we can ask Huxley about his epistemological position. What was the coiner of the term “agnostic” thinking? This is the question raised and answered in this dissertation.                                                            11 Smart 2011. 12 Flint 1903; Paterson 1932; Baumer 1960; Le Poidevin 2010; Smart 2011. There is also a different formulation. For example, see Sorensen 2009 for an understanding of agnosticism in terms of epistemic possibility and for a discussion of meta-agnosticism.  13 Baumer 1960; Woelfel 1998.  5  We might conjecture that Huxley’s notion of agnosticism and underlying epistemological position are settled issues. As we shall see from now on, many scholars have discussed Huxley and his agnosticism. I believe, however, there is an untold story that would help us to better understand what kind of agnostic he intended to be. Fifteen years after introducing the term “agnostic,” Huxley said, “If a General Council of the Church Agnostic were held, very likely I should be condemned as a heretic.”14 This remark, I think, remains valid. My aim is to understand Huxley’s agnosticism from his own voice.  1.2. Huxley and Victorian scientific community   The fact that Huxley was the person who first suggested the term “agnostic” would be one reason for paying attention to him to understand the position labeled as “agnosticism,” but there is another reason. Huxley engaged in philosophical discussions, and yet his philosophical struggle has not been appreciated as much as his social and political struggle.   Huxley, born in 1825 in London, can be seen as a typical example of a self-made person. His life started as a youngest kid of a financially insecure but large family; he attended a local evangelical school for two years but dropped out when he was only ten years old; he was unsupported and taught himself. The late-teenaged Huxley was a hard working medical school student in debt; he received scholarships for tuitions but he had no money for living expenses. In 1846, Huxley boarded on the HMS Rattlesnake as an assistant naval surgeon (unlike Charles Darwin who was on board as a gentleman paying his own expenses and later took the position of official naturalist of the Beagle). During the four-year voyage, Huxley examined marine invertebrates and sent his scientific papers and samples to London. Huxley after the journey was                                                           14 Huxley 1884a, 5.  6  a young man who had scientific reputation but no decent job, leaving his fiancée on the other side of the globe. Around the age of thirty, Huxley could eventually secure a regular job at the Royal School of Mines in London and get married after their eight-year engagement. The middle-aged Huxley was a famed “man of science,” busy with scientific works, lectures, and services for the government and various scientific organizations. Finally, in 1883, Huxley was elected as a president of the Royal Society, which was probably the most privileged and influential position in Victorian scientific community. He died in 1895.15 Huxley’s dramatic change of his social standing went hand in hand with the social standing of science in the Victorian era. The first half of the 19th century was the period in which scientific activity started to be recognized as socially meaningful activity; there were scientific organizations and journals that appreciated and disseminated scientific results and discoveries, but science was not a solid profession.16 In the mid-19th century, along with the movement of professionalizing science, there was a shift in generation started within Victorian scientific community.17 Although the nature of the transition and the degree of discord are arguable, it seems undeniable that there was a notable transition of generations within the community. Bernard Lightman describes the transition as follows:   The aristocratic gentleman of science, those Oxbridge-educated Anglicans who dominated the scientific scene in the first half of the century, provided Victorians with a vision of culture and social order based on natural theology. The middle-class Young Turks of science like Thomas Henry Huxley and John Tyndall, who came                                                           15 Huxley 1886d; 1890f; Clodd 1902; Leighton 1912; L. Huxley 1920; Peterson 1932; Bibby 1959; Ashforth 1969; Desmond 1997; Schwartz 1999; White 2002; Collie 2011. Also Darwin [1887] 1958; Desmond and Moore 1991.     16 Schweber 1981; Cahan (Ed.) 2003; White 2003; Snyder 2006. See also Turner 1997. 17 Turner 1974; 1978; Jacyna 1980; Desmond 1994; 2001; White 2003; Lightman 2004; Stanley 2011; Dawson and Lightman (Eds.) 2014; Abberley 2016. My statement simplifies the changes in Victorian scientific community around the mid-19th century, and historians have examined various aspects of the transition. For example, see Barton 1998a for a discussion of distinctive characteristics of scientific periodicals after 1860s.    7  from outside the Oxbridge environment, began at the middle of the century to vie with the gentlemen of science for the leadership of the British scientific world and the accompanying cultural authority.18   The transition from “gentlemen of science” to “young men of science” has been analyzed. Frank M. Turner emphasizes a class struggle initiated by the “middle class Young Turks” who had difficulty in securing a job and earning money despite the fact that their scientific achievements were well received and appreciated.19 Robert Young pays attention to the contrast between the natural theology of clerical scientists and the evolutionary theory of secular scientists.20 Ruth Barton draws attention to the growing political influence of the members of the X-Club upon the Royal Society, which was a dining club of nine young men of science who were against the “aristocratic patronage of science.”21 Adrian Desmond highlights the conflict between Anglicans and disgruntled nonconformists.22 Matthew Stanley argues that the transition was possible because “naturalists,” young men of science, put efforts into and had strong effects on science education.23 Recently, Lightman shows the contribution of their translation activities to weakening the power of the Anglican Church and “Oxbridge” education.24  Regardless of a focal point taken to examine the transition, Huxley has been the lead actor. Huxley was self-appointed “Darwin’s bulldog” and well known for his defense of the                                                           18 Lightman 1997b, 3. 19 Turner 1974; 1978.  20 Young 1985. See also Brock and MacLeod 1976. 21 Barton 1990, 53; 1998b.  22 Desmond 1997; 2001. 23 Stanley 2011, 540.  24 Lightman 2015b.   8  evolutionary theory;25 he was from the middle class family without the Oxbridge background; he earned “medals without money”;26 he was one of the most active members of the X-Club which lasted about thirty years; he felt strong affinity for nonconformists since he was young; he made large and various contributions to science education; he translated German scientific works and introduced them to Victorian society. It would probably be more correct to say that Huxley was in fact the source for inspiring different focal points. As the leading protagonist of the new movement in Victorian scientific community, Huxley’s role has been well recognized and examined.27 Relatedly, his rhetorical talents and strategies have been also studied.28  On the other hand, in philosophy, Huxley is invisible. Consider the analysis of the 19th century science in Britain. In the philosophy of science, the 19th century of Britain was the period during when the nature of scientific explanation was discussed and the term “scientist” was coined.29 Scholarly attention has focused on the three figures, John Herschel (1792-1871), William Whewell (1794-1866), and John Stuart Mill (1806-1873), and their views of Newtonian                                                           25 As far as I know, we cannot find Huxley calling himself “Darwin’s bulldog” from his published essays and letters. Huxley’s student Henry Fairfield Osborn reported the following in his Impressions of Great Naturalists: Huxley’s solicitude for Darwin’s strength was characteristic of him. He often alluded to himself as “Darwin’s bull dog” (1924, 58). He [Huxley] said afterward: “You know, I have to take care of him [Darwin]; in fact, I have always been Darwin’s bulldog,” and this exactly expressed one of the many relations which existed so long between the two men (Ibid., 78-9). See also Kaalund 2014 for a recent research on the reception of the famous Huxley-Wilberforce debate.  26 I borrow the expression “medals without money” from Houston Peterson (1932, the title of Chapter 3).  27 Interestingly, the recognition of Huxley’s role might have been too strong. Ruth Barton introduces “what can be described as ‘the Huxley problem’” in an article that analyzes how Victorians within the scientific community described themselves, and writes as follows: “I [Barton] have therefore been careful to avoid basing arguments on Huxley’s usage alone. For example, his usage of ‘workers’ for members of the scientific community must be checked against the usage of contemporaries” (2003, 75). 28 Block 1986; Jensen 1991; Paradis 1997; Wright 2016. 29 Cannon 1961; Ruse 1975; 2009b; Yeo 1979; 1993; Butts 1985; Hull 2003; Snyder 2006; Cobb 2011; Cowles 2016.  9  science, in particular, Isaac Newton’s (1642-1727) first rule of reasoning in philosophy, according to which science admits only both true and sufficient causes, or “true causes” (verae causae).30 This focal point deserves philosophical attention, because Newtonian science was then taken as an exemplary of science but yet the three philosophers had different understandings. Moreover, their views were reflected in their different evaluations of Darwin’s theory of natural selection, which has been considered to have important bearings on the philosophy of science. To Herschel and Whewell, Darwin’s notion of natural selection is not a scientific hypothesis that is worth considering whether it is confirmed or not; to Mill, it was a legitimate hypothesis.31 Mill, however, does not seem to have given Darwin or his followers enough time; a few years later, Mill considered a hypothesis of intelligent design to be better than Darwin’s hypothesis of natural selection.32                                                              30 Ruse 1975; 2009b; Yeo 1979; Hull 2003; Snyder 2006; 2009; Lewens 2009. 31 According to Herschel, there are two requirements for an alleged cause to be considered a vera causa (Herschel 1831; Lewens 2009; Snyder 2009). The first requirement is that its existence must be proved empirically; it must be either an already known cause or, if it is not observable, at least analogous to an observable cause. The second requirement concerns the causal sufficiency of an alleged cause. An alleged cause must be sufficient to bring about a phenomenon to be explained, and this sufficiency must be shown empirically by observation or analogy. The two requirements should be met in order for a hypothesis to be considered a scientific hypothesis. Herschel briefly mentioned Darwin’s theory in a footnote of his book: natural selection may well be an existing cause, but it is not sufficient (Herschel 1861; also see Darwin 1859b; Hull 2003).  Whewell also thought that a versa causa must have solid inductive grounds, but Whewell considered Herschel’s requirement for being analogous to a known cause to be too strict (Whewell 1840; Hull 2003; Snyder 2006; 2009). According to Whewell, it is better for a hypothesis not to invoke a cause familiar to us, because if we limit our search for true causes to known causes, science would hardly make progress (Snyder 2006). Whewell distinguished an adequate hypothesis from a “loose hap-hazard sort of guess” or “hasty and imperfect hypothesis” (Whewell 1849, 60). According to Whewell, an adequate hypothesis should be suggested by gathering facts and observations, superinducing a law, and generalizing the law (Snyder 2012, Ch. 2). Darwin’s natural selection was not an adequate hypothesis, rather a speculation, because Darwin did not provide inductive grounds for the transition from one to another species, nor did he show that the amount of time required by such a transition is available (Whewell 1864; Hull 2003; Snyder 2006; see Curtis 1987 for a different explanation of why Whewell rejected Darwin’s theory). Mill seems to have been the most lenient toward the adequacy of a scientific hypothesis. Mill wrote that “Mr. Darwin’s remarkable speculation on the Origin of Species [sic] is another unimpeachable example of a legitimate hypothesis” (Mill [1846] 1882, 614f; cited in Hull 2003, 186). Mill did not care whether a hypothesis already had solid empirical or “inductive” grounds. In fact, Mill said that “we ought not” to blame one for having “extremely slight grounds” or “bold” suggestion (Mill [1846] 1882, 614f, 615f). 32 Ruse 1975; Hull 2003.  10  There is another aspect of the 19th century science in Britain that has been noted: it was the period when the image of science as we now picture began to dominate. For example, regarding the 19th century natural sciences, Philippe Huneman notes that “[i]t became more and more possible to undertake a scientific discourse without taking sides on philosophical issues, such as metaphysical or epistemological problems.”33 In this regard as well, Huxley has not received scholarly attention from philosophers. To be sure, Huxley has been discussed as one of the most important spokespersons for Victorian scientific naturalism which was the “cult of science” that was so popular during the second half of the 19th century.34 Its proponents ruled out supernatural causes and promoted a scientific world view, but their philosophical views have been understood in terms of the positions of the early 19th century philosophers of science. Under the framework that Whewell’s philosophy of science was the representative philosophy of “gentlemen of science,” whereas Mill’s philosophy of science as the representative philosophy of “young men of science,” Huxley (and other Victorian scientific naturalists) has been assumed to simply adopt Mill’s position.35 Mill’s influence on Huxley is not deniable, and Huxley openly, and throughout his life, endorsed Mill’s System of Logic as the exposition of scientific method.36 Yet Huxley’s philosophical view should not be just taken as identical with that of Mill, as he said “though Mill’s ‘Logic’ was very good, empiricists were not bound by all his theories.”37 Huxley neither agreed with Mill’s final evaluation of Darwin’s theory, nor considered a theory of                                                           33 Huneman 2011, 202.    34 Lightman 1987, 28. The seminal work is Turner 1974. See Lightman 2009 for useful introductory discussions of Victorian scientific naturalism and the development of its scholarship. For recent researches, see Rectenwarld 2013; 2016; Lightman and Dawson (Eds.) 2014; Lightman and Reidy (Eds.) 2014.   35 Flint 1903; Ellegård 1957; Passmore 1957; Turner 1974; 1975; Stoddart 1975; Jacyna 1980; di Gregorio 1981; 1984; 1997; Knight 1997. 36 Huxley 1854a; 1878h; 1892c.  37 Huxley 1887a; L. Huxley 1908 (Vol. III), 354; partly cited in Irvine 1968, 356.  11  creation by an intelligent designer to be a scientific theory. Although Huneman does not mention Huxley, Huxley’s agnosticism can be appreciated as a philosophy of science that was intended to epistemologically liberate science from a particular metaphysics or theology.   1.3. Possible reasons for not paying philosophical attention to Huxley  Huxley as science publicist and social reformer has been emphasized in the scholarly literature, whereas Huxley as philosopher has been underrated. There seem to be at least two reasons for this asymmetry. The first seems to involve the conception that Victorian scientists were not serious philosophers. For example, John Passmore wrote that “[n]one of the nineteenth-century scientific publicists is of any great importance as a philosopher,” although, he added, their scientific works had an impact on philosophy.38 Huxley is no exception to that conception; as the most well-known Victorian public figure, he is often the main target. As Roger Smith writes:   He [Huxley] did not write systematically as a philosopher but opportunistically as a public intellectual with a large number of irons in the fire. Moreover, as an extraordinarily busy man, he wrote under intense pressure, often late at night, seeking immediate effect. He certainly succeeded; and we can admire this while not taking too formally the philosophical dimensions of what he wrote.39  Here Smith has in mind Huxley’s views on mind, but similar points have been made regardless of subject matters.40 James R. Moore characterizes Huxley and another leading Victorian scientific figure John Tyndall as “not first-rate philosophers but skilled controversialists.”41 Regarding attempts to examine Huxley’s agnosticism from a philosophical perspective, Barton,                                                           38 Passmore 1957, 46. See also Cockshut 1964; Dockrill 1964.  39 Smith 2015, 24. I am indebted to Piers J. Hale for this reference.  40 Paterson 1932; Copleston 1966; Levine 1990.  41 Moore 1988, 511.  12  who understands it mainly as a rhetorical device for Victorian scientific naturalism, writes as follows: “I disagree with the emphasis of [James G.] Paradis and D. W. Dockrill, both of whom, in seeking philosophical system in Huxley’s thoughts, pay insufficient attention to polemical intent.”42 In next chapter, I will discuss the reception of Huxley’s agnosticism in detail, including the views of the two scholars whom Barton mentions. When it comes to agnosticism, Huxley has been too famous as a polemicist and science popularizer to attract philosophical attention.  The second reason has to do with the reputation that Huxley is infamous for speaking paradoxically, to put it generously, or inconsistently. Huxley left numerous lectures and essays, and his interest was not confined to science. He wrote and talked about philosophical, theological, political and social issues. Many scholars have found his “thoughts and doings” puzzling. 43 For instance, Huxley was a strong defender of Darwin, as his famous nickname “Darwin’s bulldog” attests. According to Huxley, his initial reaction to Darwin’s Origin of Species (published in 1859) was, “How extremely stupid of me not to have thought of that!”44 However, Huxley did not accept the idea of gradual modification until around 1874, and moreover he was never fully committed to the idea of natural selection.45 Thus, scholars have been puzzled over what Huxley was defending and whether he was a “real Darwinian,” “genuine Darwinian,” “pre-Darwinian,” “anti-Darwinian,” “pseudo-Darwinian” or something else.46                                                            42 Barton 1983, 262. 43 “Thoughts and Doings” is the title of Huxley’s teenage diary (1840-45). 44 Huxley 1887g, 551. 45 Huxley 1859d; 1859e; Bartholomew 1975; di Gregorio 1981; 1982; 1984; 1997; Ruse 1997; Weiss 2004; Lyons 2009. 46 The labels are from Bartholomew 1975, 535; Mayr 1997, 250; di Gregorio 1997, 159; Bowler 1997, 120. See also di Gregorio 1981; Ruse 1997; Lyons 2009; Depew 2010.   13  Similar examples of puzzlement abound. Huxley’s materialistic writings appear irreconcilable with his insistence that he was not a materialist.47 Many of his writings are full of severe criticisms of the Bible, but he also endorsed the inclusion of the Bible in elementary school curricula.48 Regarding his lecture “Evolution and Ethics,” commentators have pointed out that it was inconsistent for him to claim that “cosmic processes” led to the evolution of ethical attitudes and yet are antagonistic to them once they have arisen; he replied by saying that “I’m sorry for logic.”49 As another example, Lightman states that “[i]t is puzzling to see Huxley praising [Georges] Cuvier,” since “Cuvier was an upholder of natural theology, and closely associated with Huxley’s enemy, Richard Owen.”50 Some scholars have even questioned Huxley’s integrity.51 Nevertheless, he said, “One thing people shall not call me with justice and that is – a liar.”52, 53 The common sentiment of puzzlement seems to reinforce or be reinforced by the conception that Huxley was just a clever opportunist, not a deep thinker committed to a system.54                                                           47 For example, Huxley 1868a; 1886c. 48 Huxley 1870d; 1879d; 1889f; 1893b; 1894e.   49 Huxley 1893a; 1893c; 1894c, 12. For discussions of the lecture, Mivart 1893; Carus 1894; White 1895; Dewey 1898; Simpson 1949; Helfand 1977; Paradis and Williams 1989; Allhoff 2003; Goslee 2004; Ruse 2009a.   50 Lightman 2014, 27. After expressing the puzzlement, Lightman introduces Gowan Dawson’s explanation (Dawson 2016). “Huxley pursued two inconsistent strategies when he discussed Cuvier. When he wrote research papers for fellow anatomists he rejected the validity of Cuvier’s law of correlation, but when writing for a popular audience he pretended to support it, since the notion of reconstructing the entire form of prehistoric creature from a single bone had caught the public imagination” (Lightman 2014, 27).  51 Houghton 1949; Richards 1987. See also Barr 1997a for a discussion of Huxley’s integrity. Huxley also knew that even his ally thought of him as a hypocrite (Huxley 1879d).   52 Huxley 1860c, 319. 53 In addition to the above examples, also note Barton’s report. “Historians have found it difficult to find a consistent viewpoint in Huxley’s many assertions about nature, its order, its laws, and its chains of causation. Some accuse him of deliberate equivocation. Others try to identify shifts of opinion, but no one agrees on what changed, when, or why” (1983, 268).  54 Peterson 1932, 294, 312-3; Ashforth 1969, 122; Barton 1983, 261, 265, 269; Lightman 1987, Ch. 1; de Waal 2013, 34-5. For reports of this sentiment, see Blinderman 1966, 50-2; Dockrill 1971, 462.   14  As we will see in next chapter, the idea that Huxley was naïve concerning the issues that he was addressing, unaware when he was contradicting himself is also found in the reception of his agnosticism.  Despite the dominant impression of Huxley, since it is not a hidden fact that he addressed philosophical issues, few sympathetic scholars have looked into his writings and tried to defend him.55 Charles S. Blinderman is an early notable example. In “T. H. Huxley: A Re-evaluation of his philosophy,” Blinderman paid attention to Huxley’s discussions of materialism and idealism to make sense of Huxley’s denial of being a materialist.56 Probably having a similar motivation, Sophie Forgan and Graeme Gooday write that their research on Huxley’s working places such as the Jermyn Street Museum “will help to articulate some of the contradictions in Huxley’s life and may help to some degree to account for them.”57 Also, some scholars have examined Huxley’s views on science, noting that, unlike other natural philosophers like Darwin, Huxley wrote essays that belong to what we now call “philosophy of science.”58 Although Huxley’s position has been introduced fragmentarily, the following points have been made: he thought that a direct empirical demonstration of hypothesis is highly important;59 he considered a skeptical attitude to be constitutive of scientific method;60 he praised Zadig’s method, that is the application of knowledge of cause to distant time and                                                           55 Few scholars think that Huxley appears to have tried constructing a philosophical system, but conclude that he failed in the end (Blinderman 1966; Dockrill 1964; Paradis 1978).     56 Blinderman 1966. According to Blinderman, Huxley held both idealism and materialism; the young Huxley was a more idealist whereas the old Huxley was a more materialist; Huxley was inconsistent in the sense that he did not commit himself to one position throughout his life by abandoning the other entirely. However, Blinderman did not view this only negatively: everyone tends to swing between the two perspectives, because there is some truth in both; going back and forth can be seen as a constructive process thanks to Huxley’s agnosticism.       57 Forgan and Gooday 1996, 438. 58 Ellegård 1957; Knight 1997. 59 di Gregorio 1984; 1997; Hull 2003; Paradis 1978; Lyons 1999; 2009. 60 Barton 1983; Knight 1997.  15  space;61 he rejected Auguste Comte’s idea of the hierarchy of sciences;62 his view was similar to Herschel’s and Mill’s, but not to Whewell’s;63 he was not a Baconian in the sense that he viewed the invention of hypothesis as crucial for science.64 Huxley’s meta-scientific thoughts have been discussed, but not in connection with his agnosticism.  I have pointed out the asymmetrical standing of Huxley in the scholarships as well as the conception of Huxley as a shallow and paradoxical thinker. Huxley’s agnosticism is no exception to this tendency. I take it to show that further research can be done to better understand Huxley’s philosophical notion of agnosticism. The aim of this dissertation is to offer a more comprehensive picture of his agnosticism, which I believe helps us to understand, first, his epistemological view underlying his proposal of agnosticism, second, the nature and structure of agnosticism, and last, the relation between his agnosticism and his meta-scientific thoughts. The main message of each can be summarized as follows:    1. Following modern philosophers, especially René Descartes and David Hume, Huxley engaged in an epistemological project which was to find a method for identifying knowledge that would work regardless of our limited condition. Huxley’s question concerns how to select mental contents that deserve the title of knowledge among all (possible) mental contents which certainly exist. His solution was to adopt the method of verification and material terminologies. To Huxley, it was not an epistemologically significant question how our mind comes to have mental contents that we consider knowledge.                                                             61 Knight 1997; Sommerville and Shortland 1997. 62 Eisen 1964; Knight 1997. 63 Ellegård 1957; di Gregorio 1984; 1997; Knight 1997. 64 Sommerville and Shortland 1997.  16  2. Huxley based agnosticism on conclusions of his epistemological project. I characterize his agnosticism as a quick guide to knowledge that consists of an account of legitimate evidence and an ethics of knowing (as opposed to an ethics of belief). Huxley’s agnosticism says that the knowledge status of proposition p should be determined by evidential reasons alone. It further requires epistemic agents to submit themselves to evidence and take a stance toward p accordingly. When presented with evidential reasons, agents involuntarily come to have a belief or opinion about the knowledge status of p; they should be simply honest about it. On the other hand, having a belief or opinion about the knowledge status of p based on non-evidential (e.g., prudential) reasons is done voluntarily; they should not pretend that p is knowledge.  3. Based on the notion of agnosticism above, Huxley defended science. Science is the only means of attaining knowledge, not because principles of science such as the uniformity of nature have been shown to be true or real, but because science has followed agnosticism. Science is metaphysics-agnostic in the sense that it makes use of metaphysical principles but does not base its epistemic value on any metaphysics. The proper stance toward principles of science is to put agnostic faith, which is trusting and taking risks. The notion that science needs the truth of a particular metaphysical or theological system to have any epistemic value comes from an application of an old philosophical view to science, which Huxley found not only improper but also potentially harmful to the progress of science.       By arguing that Huxley had a bigger picture behind his agnosticism, I hope to show that Huxley did not suggest agnosticism simply to underscore a certain psychological state (being ignorant or lacking a belief), a morally desirable attitude (humble or modest attitude), or an admission of the condition of human faculties (the confession of human ignorance or limits); also I hope to clarify why Huxley, in Hume where the term “agnosticism” was first mentioned in print, would have introduced agnosticism as “modern way of thinking.”65                                                             65 Huxley 1878h, 70.  17  1.4. Structure of the dissertation  Chapter 2 discusses interpretations of Huxley’s agnosticism that have been offered. Huxley associated more than one notion with his agnosticism, and most interpretations center on one of them. This leads me to categorize the interpretations into three groups. I also introduce recurrent worries about Huxley’s promotion of agnosticism, and suggest that the worries have arisen because of the adoption of a partial approach to his agnosticism. I propose instead taking an inclusive approach. To illustrate what it means to take an inclusive approach, I discuss Dockrill’s and Stanley’s understandings of Huxley’s notion of agnosticism. Yet I find their interpretations unsatisfactory in another aspect, and conclude that we need to pay more attention to Huxley’s views of what and how we can know, as opposed to what and how we cannot know.     In Chapter 3, entitled “Huxley’s Cartesian Project,” I examine Huxley’s interests in Descartes’s epistemological inquiry. Huxley understood Descartes’s inquiry into certainty as an attempt to introduce a method for identifying knowledge, and claimed that Descartes had offered two “paths” and one “maxim.” His explanations of Descartes’s inquiry and its development illuminate the structure and nature of his agnosticism. By showing that the two paths and the maxim also feature in Huxley’s later discussions of agnosticism, I support the two suggestions made in the previous chapter: we should take an inclusive approach and Huxley intended agnosticism to be a guide to knowledge, as opposed to a theory of nescience.   I move on to discuss how Huxley incorporated the paths of idealism and materialism into his agnosticism. Huxley revised the conclusion of Descartes’s method of doubt and took only an epistemological point from George Berkeley: thoughts (perceptions or mental contents appeared to consciousness) certainly exist and only thoughts are given to us. Huxley also disagreed with Descartes’s rationale behind the second path of materialism, because Descartes had a  18  metaphysical reason to employ physical way of thinking. I discuss Huxley’s epistemic reason to favor physical way of thinking over non-physical way of thinking, highlighting his distinction between metaphysical materialism and material terminology. I leave it unexplained how Huxley could think that physical way of thinking (material terminology) without an ontological commitment to metaphysical materialism gets us something that deserves the title of knowledge. This question is taken up in next chapter.  In Chapter 4, I mainly look into Huxley’s Hume to find an answer. We can see there Huxley explaining what kind of thoughts can be considered knowledge and clarifying what would be the condition for a thought to be knowledge. Given our epistemic situation, since we cannot but make use of the stream of thoughts in finding a method of identifying knowledge among all thoughts, we need to first sort out thoughts to see how to proceed. Thus, I explain Huxley’s categorization of thoughts to prepare the discussion of Huxley’s view of memory as a model of knowledge. Huxley’s conclusion was that we can only use experiencing of a set of sensory and relational impressions as a justificatory factor, which supplied him with another reason to adopt physical way of thinking (material terminology) over non-physical way of thinking (immaterial terminology) in expressing a thought that one wishes to claim to be knowledge. This chapter ends with Huxley’s defense of the method of verification as the only justificatory method, against two methods, one appealing to the notion of necessity and the other to religious faith.   Based on the previous discussions, in Chapter 5, I elaborate on the suggestion that agnosticism primarily concerns the issue of knowing and involves the three elements that Huxley drew from Descartes. I characterize agnosticism as a guide to knowledge, consisting of an account of legitimate evidence and two epistemic duties regarding evaluation of evidence. In  19  showing that Huxley’s concern was ethics of knowing, I argue against the interpretation of Huxley’s agnosticism as William K. Clifford’s ethics of belief; this discussion will resolve one of the two recurrent worries that will be introduced in Chapter 2. To further support my interpretation, I draw attention to Huxley situating agnosticism within the tradition of modern critical philosophy. The penultimate chapter concerns epistemological bearings of agnosticism on Huxley’s view of science. I re-visit the other recurrent worry that Huxley’s agnosticism was a hindrance to his view of science as the only way to knowledge because he could not but base science on faith and yet he criticized his religious opponents for their faith. I argue that this worry is not well-directed. I first identify assumptions underlying the worry and then discuss Huxley’s conceptions of science and of faith. This discussion aims to show the untenability of the assumptions and to illustrate Huxley’s stance toward science. Huxley defended the epistemic value of science on the basis of agnosticism and considered both agnosticism and faith to be essential to science and its advancement.      The final chapter, Chapter 7, summarizes discussions and arguments in the previous chapters.        20  ▌CHAPTER 2  The Reception of Huxley’s Agnosticism     2.1. A science defender’s nescience: a puzzling position  Huxley is often associated with agnosticism in the scholarly literature, in particular the literature concerning the relation between religion and science. Huxley has been introduced as the first person who used the term “agnostic” along with a brief note on his notion of agnosticism.66  If we look into discussions of agnosticism that do not pay much attention to Huxley, agnosticism seems to stand as a defensible, irreducible philosophical position.67 When we move from the discourse of agnosticism in general to that of Huxley’s agnosticism, on the other hand, criticisms abound. An interesting point is that almost all criticisms against his agnosticism have something to do with his conviction in scientific method.  Agnosticism seems to promote a stance of ignorance toward metaphysical and theological issues. Yet it is well known that Huxley committed himself to scientific method. If so, one question arises. How could Huxley be so sure that scientific method is the only means of acquiring knowledge? When we move on to his own explanations of agnosticism, it appears that he put himself in trouble by inducing the question, because his agnosticism seems to involve the acceptance of scientific method as the only way to knowledge. To hold agnosticism appears to mean to make confession of ignorance toward metaphysical matters and to have conviction in scientific method. Scholars have noted that there is something puzzling about Huxley’s                                                           66 van Fraassen 1998; Benn 1999; Hyman 2006; Dawkins 2006; Yoder 2013; de Waal 2013; Ruse 2014.  67 See for example Rosenkranz 2007.   21  agnosticism.68 If one wishes to be an agnostic, one needs to express both humility and dogmatic adherence. Thus, it seems that Huxley’s agnosticism does not help us to understand how he could be so sure about scientific method. The question above is left unexplained. Perhaps this should not be expected; agnosticism might not be the right place to look for his rationale, because agnosticism in general has been regarded as a view on the limits of knowledge, or our ignorance. Perhaps we might find the rationale for his conviction in science elsewhere.  Yet, there is a more serious set of challenges than the paradoxical aspect. Many scholars have pointed out that if Huxley held agnosticism, he could not have grounds for his conviction in science, or he should not have expressed the conviction. These worries concern his endorsement of agnosticism, as opposed to his agnosticism. Given that his larger project was to defend and promote science as the only knowledge for society and its progress, agnosticism seems not suitable for his objective. The worries can be formulated as two kinds of criticism, which were made by his contemporaries and still continue today. First, endorsing agnosticism means that Huxley was philosophically naïve because he was thereby undermining foundations of science.69 Second, endorsing agnosticism made him philosophically inconsistent because he held double standards, one for religious faith and the other for scientists’ faith.70 These points suggest that the fundamental claim underlying his project of promoting science is nothing but his own “gnosis.”   Before I elaborate on the two general criticisms, let’s look at Huxley’s own explanations of agnosticism. It is not quite straightforward to identify what he meant by agnosticism. “One of the problems facing the student of agnosticism, and in particular Victorian agnosticism,”                                                           68 Hutton 1895; Dockrill 1971; Gilley and Loades 1981; Barton 1983; Lightman 1987; Levine 2014.    69 Balfour 1895; Turner 1974; Lightman 1987; 1997a; 2001; Levine 2014; Stanley 201.  70 Ward 1899 [1915]; James 1879; Gilley and Loades 1981; Lightman 1987; 1997a; 2001; Greene 2003; Levine 2014. Numbers 2003 introduces this line of criticisms.    22  Dockrill says, “is to know what to make of T. H. Huxley’s definitions and explanations of the word.”71 As implied, a problem emerges because Huxley associated agnosticism with more than one idea. He seems to have entertained three ideas. First consider two ideas that concern what we can or cannot know.    The theological “gnosis” would have us believe that the world is a conjuror’s house; the anti-theological “gnosis” talks as if it were a “dirt-pie” made by the two blind children, Law and Force. Agnosticism simply says that we know nothing of what may be beyond phenomena.72   Here Huxley claims that we cannot know about theological and metaphysical matters because their subject matter concerns something “beyond phenomena.” Yet, Huxley also seems to have wanted to associate agnosticism with science:   It [Agnosticism] simply means that a man shall not say he knows or believes that which he has no scientific grounds for professing to know or believe.73  Those two quotes seem to express two different ideas. Based on Huxley’s discussion of the limits of human faculties, it has been suggested that he claimed that we are ignorant of metaphysical issues that go beyond phenomena, i.e., what appear to us in our consciousness. On the other hand, the second quote is deemed to express the idea that scientific method is the only means of attaining knowledge. Jointly, Huxley’s agnosticism appears to make the following claim:                                                           71 Dockrill 1971, 461. 72 Huxley 1884a, 6.  73 Ibid., 5.   23  regarding something beyond phenomena, we can know nothing because of our limited condition, and regarding phenomena, we can know only by using scientific method. How did Huxley arrive at this double-sided claim? This question has led to different interpretations of his notion of agnosticism, as we will soon see.  The third idea that Huxley seems to have associated with agnosticism involves what we should or should not do. As implied in the aforementioned quote, his agnosticism requires one to confess one’s ignorance. He proposed what he called the “agnostic principle.”   Positively the principle may be expressed: In matters of the intellect, follow your reason as far as it will take you, without regard to any other consideration. And negatively: In matters of the intellect do not pretend that conclusions are certain which are not demonstrated or demonstrable.74  The “essence” of agnosticism, according to Huxley, “lies in the rigorous application of” this principle.75 What does agnosticism prescribe? This has complicated the issue of how to understand Huxley’s agnosticism.     2.2. Partial approach to Huxley’s agnosticism: the common approach  Huxley, after all, appears to claim: regarding something beyond phenomena, we can know nothing due to our limited faculties; regarding phenomena, we can attain knowledge only by using scientific method; follow the agnostic principle. We may grant that Huxley’s agnosticism refers to all of them, without asking any further question regarding how they are related to each                                                           74 Huxley 1889e, 246.  75 Ibid., 245.   24  other.76 Yet scholars have attempted to reconstruct how he would have arrived at agnosticism. Distinctive reasons seem to have led to each idea of agnosticism, and different interpretations have been offered accordingly. We can find, I suggest, three types of interpretations.    2.2.1. Type one: an incidentally conjoined position   The main motivation that leads to this type seems to be to avoid making the notion of agnosticism incoherent. The claim that scientific method is the only way to knowledge would generate tension with the claim that we cannot know something beyond phenomena. If Huxley meant agnosticism to mean that we are ignorant of some metaphysical and theological matters because of our limited faculties that restrict the realm of knowledge to phenomena, how could he also want agnosticism to mean that scientific method, not other methods, is the only means for gaining knowledge? To make this claim, he would have to claim that he knew something that cannot be known due to our limited faculties. Therefore, Huxley’s discussions of agnosticism turn out to involve incompatible ideas, not just different ideas. Consequently, proponents of this type tend to separate the two ideas. Agnosticism is identified with an account of the limits of human knowledge based on the notion of the limited human faculties; his commitment to scientific method is now connected with Victorian scientific naturalism. Huxley’s agnosticism is, thus, labeled as “scientific agnosticism.”77 Huxley loosely connected the two ideas to the same                                                           76 Van A. Harvey, for example, characterizes Huxley’s agnosticism as skepticism and underscores its two components: “skepticism regarding the claims of metaphysicians or theologians to have produced knowledge and skepticism regarding the historical tradition about Jesus in the New Testament” (Harvey 2012, 536). Harvey also discusses Huxley’s agnostic principle as the third component, and his view will be addressed in detail in Chapter 5.  77 For example, Desmond 1997, 500, 525; Lightman 2002, 271, 289. See also Turner 1974. There is another rationale for the label “scientific agnosticism.” As Huxley himself admitted, the idea that we are ignorant of metaphysical matters is also found in other thinkers like Mansel. The label “scientific agnosticism” is used in contrast with “religious” agnosticism. In this regard, it should be noted that Turner’s and Lightman’s books concern more than Huxley. A related distinction is the distinction between “right wing” agnosticism and “left wing” agnosticism (Passmore 1959; Harvey 2012).     25  term “agnosticism,” and he did simply because he endorsed both; he may or may not have acknowledged a philosophical complication coming from holding the two ideas.   Huxley endorsed the idea of science as the only way to knowledge, and it seems quite straightforward why he came to endorse the idea: he was a devoted scientist. A more interesting question regarding his coinage of the term “agnostic” would be how he arrived at the idea of metaphysical and theological ignorance grounded on the limits of human faculties (call it “metaphysical ignorance doctrine”) and why he found it worth endorsing and promoting enough to bother himself to give a thought on a new name. We have been given two accounts.   Let me start with a more hostile reading, which was once popular. Huxley promoted agnosticism, the metaphysical ignorance doctrine, not because he wanted to express his genuine view. Agnosticism conveys the philosophical notion which had been already well known without the help of the new label, but the notion with the new label was to him a device invented for polemical use. Accordingly, a “purely opportunistic element in agnosticism” has been examined.78 For example, agnosticism is claimed to be a red-herring, because his intention was to hide or soften his genuine position: atheism, materialism, or scientism.79 Or, his intention behind promoting agnosticism was to win a debate with his religious opponents, and thus he used it as a tool of showing that they cannot know about what they claimed to know.80   This uncharitable account has been shown to be untenable by Bernard Lightman, because of its main point on Huxley’s purpose of the coinage. According to Lightman, “Huxley’s invention of the term ‘agnostic’ is often seen as a brilliant rhetorical strategy,” under the                                                           78 Moore 1988, 511.  79 Engels 1892; Lenin 1908; Bibby 1959. See also Dockrill 1971; Lightman 2015a. 80 Paterson 1932; Irvine 1968; Moore 1988; Reed 1997. Ruth Barton accepts Lightman’s view which I will introduce shortly and yet claims that polemical intention is the most important in understanding Huxley’s agnosticism (1983).   26  assumption that “Huxley’s role as a neologist was well known to his contemporaries during the 1870s and 1880s, before the publication of his trilogy of essays on agnosticism.”81 However, Lightman has shown that this assumption is incorrect. Although Huxley coined the term in 1869, he did not mention the terms “agnostic” and “agnosticism” in his essays and lectures until 1878. Most of his contemporaries, albeit familiar with agnosticism, did not even know that Huxley was the coiner of the terms until the mid-1880s when his letter revealing the fact was published without his permission in the inaugural issue of the Agnostic Annual, partly because he had not publicly engaged in controversies over agnosticism.82 In the meantime, however, Huxley continued to express the position of agnosticism in his letters and published works.83 Lightman also points out that Huxley in the end revealed himself as the coiner in public debates and tried to control the meaning of agnosticism. The story fully uncovered by Lightman renders the “rhetorical strategy” account implausible because, if it were correct, we would arrive at an odd conclusion that Huxley invented a rhetorical device that he did not use for around twenty years in public and polemical contexts.84                                                            81 Lightman 2002, 272. Lightman is referring to Huxley’s “Agnosticism,” “Agnosticism: A Rejoinder,” and “Agnosticism and Christianity,” all of which were published in the Nineteenth Century 25 in 1889. 82 Huxley 1884a; Desmond 1997, 566-7; Lightman 2002; Lyons 2012. However, this does not mean that Huxley was not associated with agnosticism around that time. The first place where the terms “agnostic” and “agnosticism” appeared in print is Richard H. Hutton’s essay, “Pope Huxley,” which was published on January 27, 1870, less than one year after Huxley’s coinage. Hutton was also a member of the Metaphysical Society and in that essay, Hutton identified Huxley as an agnostic and his position as agnosticism (Hutton 1870).     83 Lightman 2002; 2004. Even before inventing the word “agnostic,” Huxley held agnosticism. This point has been noted by his son Leonard Huxley (1920) and other scholars including Lightman (e.g., Clausen 1976; Lyons 1999; 2012; White 2003; Beatty and Hale 2008).     84 Lightman asks, “Why did he [Huxley] wait so long, only six years before his death, to state what he meant by the term as originally conceived?” and suggests the following: “At first, Huxley had remained relatively silent after 1869 in order to allow his coinage ‘agnostic’ to become part of intellectual discourse, but he later had an additional reason for not speaking at length about the meaning of the term: he would have had to reveal his significant gap between himself and his friend [Herbert Spencer]. Huxley wanted to present a united front, until 1889 when he was in the middle of an acrimonious quarrel with Spencer” (Lightman 2002, 272, 287). See also Huxley 1889i.   27   This “rhetorical strategy” account does not tenably account for Huxley’s adoption of agnosticism (identified as the metaphysical ignorance doctrine), but we can find a more balanced version from Lightman’s book Origin of Agnosticism. Although Lightman notes, in a footnote of his essay that I discussed just above paragraph, that his book, published in 1987, “follows” the “line of approach” that assumes that “agnosticism constituted a well-thought-out strategy which Huxley resorted to often during the 1870s and 1880s,” his account explains how Huxley would come to hold agnosticism and use it as a tool to attack religious targets.85 Lightman defines agnosticism as “a species of skepticism built upon Kantian principles”:   Huxley therefore conceived of agnosticism as a theory that restricted knowledge to the phenomenal realm and that was based on Kant’s notion that human mind is subject to inherent limitations. … Any object that could be termed part of the transcendental or noumenal world was considered to be beyond the limits of human knowledge.86   Huxley claimed that we are ignorant of some metaphysical or theological matters such as the existence of God or the constituents of reality, because these issues are not about the phenomenal world revealed by (the Kantian sense of) experience. This is Huxley’s agnosticism, and he was led to appreciate the destructive power of this “Kant’s notion” via a Christian thinker, Henry Longueville Mansel (1820-1871). Mansel’s Limits of Religious Thought, published in 1859, was mentioned more than one time in Huxley’s discussions of agnosticism and in an approving manner. Lightman’s account has been widely recognized and shared by other scholars.                                                             85 Lightman 2002, 273, n. 11.  86 Lightman 1987, 15. Also Lightman 1983; 2012; 2014.   28   To summarize, the first type of interpretation of Huxley’s agnosticism views his discussions of agnosticism as a juxtaposition of different ideas, and only counts the metaphysical ignorance doctrine as agnosticism.87 The most elaborate account of how Huxley came to endorse the metaphysical ignorance doctrine is Lightman’s account that highlights Mansel’s influence.   2.2.2. Type two: an application of scientific method  The second type of interpretation of Huxley’s agnosticism understands it as an expansion of his adherence to scientific method.88 The main motivation seems to be that an interpretation of Huxley’s agnosticism should make sense of his stance toward issues that do not require us to go beyond phenomena. If we equate his agnosticism with the metaphysical ignorance doctrine based on our conditioned faculties, agnosticism would have nothing to say about a claim concerning the phenomenal world, except that it does not fall within the boundary of ignorance. However, his three 1889 essays on agnosticism primarily concern alleged miracles like Jesus walking on water. Huxley intended to apply agnosticism not only to claims about the nature of reality but also claims about immaterial entities such as soul or vital force and even seemingly legitimate scientific claims. This motivation seems to have led some scholars to focus on Huxley’s illustration of agnosticism as the “essence of science.”89  According to the second type, Huxley’s agnosticism follows from his conviction in scientific method, confirmation by sense experience. Huxley would arrive at the notion of                                                           87 Lightman’s reason for the exclusion seems to be the following. “Although there were times when Huxley himself, carried away by the heat of controversy and his own polemical skill, used the word agnosticism rather loosely, it is fairly clear what he intended. In those key sections of Huxley’s work where he deals with his conception of agnosticism, two elements will always be found: a discussion of Kant or a thinker profoundly influenced by Kant, and an elaboration of Kant’s notion of the limits of knowledge” (1987, 14).  88 Hibben 1903; Turner 1974; MacLeod 1982; Jensen 1991; Lyons 2014.  89 Huxley 1884a, 5.   29  metaphysical ignorance because we cannot solve metaphysical matters, for example, whether materialism or immaterialism is true, by applying scientific method. Although Huxley spoke of Kant, Hamilton, and Mansel in connection with the idea of limited human mind, this would be just his careless way of expressing the point that scientific method is inapplicable. It would be better to look at what his practice of agnosticism was like.      Sherrie L. Lyons, whose research focus is Huxley’s scientific works, holds this interpretation. To examine “what he meant by it [agnosticism],” she says, we need to understand “the context of his life as a scientist.”90 The following quote from Lyons shows her interpretation. “On those questions [like the existence of God] that were not amenable to the scientific method, i.e., those that went beyond the cognizance of the five senses, he [Huxley] declared himself an agnostic.”91 This, she highlights, should not be understood to imply that Huxley was not critical about scientific claims. The core idea underlying his agnosticism is that any – metaphysical, theological, and scientific – claim should be evaluated by the same standard, and the standard is that of science. In this sense, according to Lyons, “Huxley’s agnosticism provided the framework for his scientific view,” but also “his experience as a scientist provided the framework for his agnosticism.”92   A similar but interestingly different view has been suggested by James G. Paradis.93 Lyons portrays Huxley qua agnostic as a confident scientist, but in Paradis’s interpretation, the agnostic is illustrated as a frustrated scientist. According to Paradis, Huxley’s declaration as an                                                           90 Lyons 2012, 86.  91 Ibid., 87. In her earlier essays, Lyons explains Huxley’s agnosticism rather unclearly: “Building on the Kantian principle that the human mind had inherent limitations and further elaborated by Hume, Huxley maintained that our knowledge of reality was restricted to the world of phenomena as revealed by experience. It also became the cornerstone in defining what constitutes the practice of science” (2009, 152; also 2010, 435). 92 Lyons 2012, 102.     93 James G. Paradis, T. H. Huxley: Man’s Place in Nature (1978).  30  agnostic dramatically shows his identity as a scientist who underwent a kind of epistemological crisis. “Huxley,” Paradis states, “had recognized that as physical science moved closer to the material objects of its attention, achieving greater clarity and certainty, the ontological considerations were becoming impossible to deal with definitely.”94 Huxley expressed a kind of ambivalent feeling which Paradis thinks captures the historical aspect of agnosticism. Huxley’s strict application of empirical method led him to declare that we cannot know the truth of doctrines that had been supplying us with intellectual and moral principles. Thus, Huxley wanted to have an epistemology that can validate some of his beliefs in such principles, in particular the order of nature, but he could not find one. Agnosticism excused Huxley’s failure. Agnosticism was a “scanty gown” but yet “suitable attire” for Huxley as a scientist.95         2.2.3. Type three: an intellectual code  This type of interpretative approach pays more attention to what Huxley’s agnosticism asks one to do or not to do, than to how exactly Huxley grounded his account of the limits of human knowledge. The scholars who take this approach tend to focus on Huxley’s “agnostic principle.”  One cluster of interpretations of this type highlights the notion of free inquiry implied in agnosticism. Huxley could arrive at the state of being ignorant, because, in the first place, he questioned beliefs that people had been forced to hold uncritically. Moreover, he always emphasized the freedom of inquiry and wanted it to be realized in society: the state of being free from traditionally inculcated beliefs and being free to question. In this context, David Knight                                                           94 Paradis 1978, 101. 95 Ibid., 103. As I will show in Chapter 6, Huxley expressed his instrumentalistic conception of science too strongly and systemically to characterize his position as an excuse for no epistemology.     31  connects Huxley’s agnosticism with the “Royal Society’s motto, ‘Nullius in verba [Taking nothing on authority]” and points out that he wanted science to be “an agnostic enterprise.”96   Another related cluster draws our attention to intellectual honesty. Agnosticism does emphasize free inquiry, but further, Huxley associated free inquiry with intellectual and moral virtue. His contemporary Robert Flint considered Huxley’s agnosticism to mean “simply honesty in investigation” or “merely the conscientious exercise of intelligence in the pursuit of truth.”97 Flint appears not to have been impressed by this “unquestionable” or “self-evident” principle.98 On the other hand, according to Christopher Clausen, the significance of Huxley’s agnosticism is that it has turned having doubts and the absence of beliefs, by redefining them as honesty or even humility, into a more superior or at least alternative position that one may take.99 Clausen thinks that it was Huxley’s contribution; no one had been successful in this regard.100  Yet another version of this type is the one popular in the literature on the ethics of belief. Huxley’s agnosticism is understood to express evidentialism, according to which we should have a belief based on only evidential reasons as opposed to non-evidential reasons such as prudential reasons. The fact that Huxley was a close friend and supporter of William K. Clifford (1845-1879) further has invited scholars to interpret his agnosticism in this way. In the famous essay, “Ethics of Belief,” Clifford asserted: “[i]t is wrong always, everywhere, and for anyone to believe anything on insufficient evidence.”101 This is a quite clear example of evidentialism,                                                           96 Knight 1997, 57, 58.  97 Flint 1903, 43, 44.  98 Ibid., 43.  99 Clausen 1976.  100 Alan P. Barr, in his essay on Huxley’s passion for truth, also understands Huxley’s agnosticism as intellectual honesty in the sense that agnostics undertake “not to assert the truth of anything about which he or she is uncertain” and points out that “this honesty of word or deed” was what Huxley took as the most valuable quality (1997a, 17-8).  101 Clifford 1876, 295.  32  although contemporary evidentialists tend to claim only for being epistemically wrong, unlike Clifford who presented his imperative as both epistemic and moral. And, when William James (1842-1910) argued against Clifford in another famous essay “Will to Believe,” he also cited Huxley and considered them to have the same view.102 Since then, the understanding of Huxley’s agnosticism as Clifford’s ethics of belief has been received without a serious challenge.103   This “intellectual code” type also has accounts of how Huxley would have arrived at agnosticism. Paul White’s works offer a notable account.104 White understands that agnosticism was “a more dynamic orientation toward belief as a practice: an active questioning, an openness toward beliefs of others or toward evidence contrary to one’s own beliefs, a process of conscientious doubt and inquiry.”105 According to White, the Metaphysical Society, an enclosed debate club of social and intellectual elites with fundamentally opposed views, can illuminate how Huxley would have been led to agnosticism. Indeed, Huxley said that the meetings of the members of the Society had triggered the coinage of “agnostic.”106 White suggests, based on topics and manners of discussions done in the Society, that the Society could function due to the shared agnostic code of openness and inquiry.107 Agnosticism, then, emerged as a “bridge between intellectual groups, a means of drawing them together despite difference of belief.”108                                                            102 James 1896.  103 For example, Kauber 1974; Doore 1983; Weolfel 1989; Aikin 2008; Yoder 2013. This is probably because philosophers’ focus has been put on a position called “evidentialism,” as opposed to Clifford’s or Huxley’s position. I will discuss the interpretation of Huxley’s agnosticism as Clifford’s ethics of belief more thoroughly in Chapter 5. 104 Thomas Huxley: Making the ‘Man of Science’ (2003) and “The Conduct of Belief: Agnosticism, the Metaphysical Society, and the Formation of Intellectual Communities” (2014). 105 White 2014, 222.  106 See footnote 1.    107 The Metaphysical Society was initially suggested in 1868 as the “Theological Society” to discuss questions on “speculative subjects, especially theology,” “with the freedom of an ordinary scientific society” (Hutton 1885, 177; also cited in White 2014). 108 White 2014, 237.    33  2.3. Huxley’s promotion of agnosticism  “Many studies,” Barton says, “have found inconsistencies and inadequacies in Huxley’s agnosticism.”109 Many of the findings have to do with Huxley’s project of defending science: this project seems not to hang together with espousing agnosticism.110 We can situate recurrent worries that are thought to bear on philosophical value of agnosticism as follows: agnosticism poses two problems for Huxley who wanted to claim that scientific method is the only means of acquiring knowledge. Depending on scholars’ interpretation of agnosticism, the two problems are presented as criticisms or illustrated as its limitations.  The first problem has to do with agnosticism as an account of the limits of human knowledge. It is problematic, because this account left him with no way to show the reality of objective validity of axioms of science. Science has fundamental assumptions: for example, the uniformity of nature, causation, and the existence of an external world. These assumptions would seem to fall under the realm of our ignorance. Consider Type One interpretation: if we understand Huxley to base his account of the limits of knowledge on the notion of human condition restricted to phenomena, scholars have pointed out, he would not be able to show the reality or objective validity of the fundamental principles of science.111 Huxley cannot claim that he can, because he would end up claiming to know something that cannot be known according to his own doctrine of metaphysical ignorance. Regarding this self-destructive aspect of Huxley’s endorsement of agnosticism, Lightman writes, “[a] philosophical justification of the axioms upon                                                           109 Barton 1983, 261. 110 Hutton 1895; Flint 1903; Paterson 1932; Ashforth 1969; Barton 1983; Lightman 1987; Levine 2014. 111 Lightman 1987; Levine 2014.   34  which science must be based could not be undertaken by the agnostics if they restricted knowledge to the same degree as did Mansel.”112  Type Two (an application of scientific method) leads us to the same concern. If Huxley based his account of the limits of knowledge on the applicability of scientific method, again, he would not be able to show the reality or objective validity of axioms of science, because, in this case, these claims cannot be shown to be true by using scientific method, inference based on observations. One of his contemporaries rhetorically asked, soon after Huxley’s death, “How could Professor Huxley be an ‘Agnostic’ if he knew as much as that?”113 In short, agnosticism is relied on scientific method and it is supposed to be the only way of knowing, but, unfortunately, it cannot help Huxley; as have been said, scientific method cannot justify its fundamental principles.  Thus, regardless of whether Type One or Type Two is accepted, Huxley was undermining the grounds of science by proposing agnosticism. In this sense, his endorsing agnosticism is philosophically short-sighted given his larger project.114 The second problem troubling Huxley is that his agnosticism appears to prohibit him from doing what he did during his whole life – believing and claiming that science is the only                                                           112 Lightman 1987, 9.  113 Hutton 1895, 105. Hutton concluded that “he [Huxley] is the great Agnostic who has tried, and, as I hold, tried in vain, to regard physical science as the one sure guide of life, …” (ibid., 106).   114 Recently George Levine summarizes the issue and reports Turner’s reaction to his earlier thought as follows: …, thanks to the work, among others, of Bernard Lightman and Frank Turner, historians and philosophers of science have recognized fundamental intellectual inconsistencies in the [Victorian scientific] naturalists’ program. Lightman has demonstrated the inherent contradictions in naturalist thought and its vulnerability to philosophically strenuous questioning of its metaphysical bases. Epistemologically, scientific naturalism seems not to have had a leg to stand on. Turner has talked of “the existential, intellectual, and moral bankruptcy of scientific naturalism,” echoing judgments made by many of the naturalists’ contemporaries. Nevertheless, some years ago when I cited the phrase at a conference Turner was attending, he blanched at his own rhetoric and asked, “Did I write that?” It is now, alas, too late to find out how he might have wanted to diminish the severity of the dismissal (2014, 80).   35  means of gaining knowledge. Agnosticism involves the idea that we should confess our ignorance about what we cannot know. Huxley criticized Victorian clerics because they claimed to know what they cannot know. However, to his critics, at the end of the day, the belief in axioms of science is on par with the belief in assumptions of theology.115 For example, Flint wrote, Huxley’s agnosticism “was confined to beliefs not drawn from and confirmed by sense-perceptions, but was not hostile to such as were.”116 Huxley’s promotion and defense of science seem to suggest that he did not follow what his own agnosticism demands: question passed-down beliefs or confess ignorance. Huxley only vehemently pushed his enemies to “critically examine” their faith and further to admit their ignorance, while letting his allies hold and express faith (in science). If we understand Huxley’s agnostic principle as Clifford’s imperative, the situation gets worse, because the imperative regulates one’s belief formation. Huxley should not even have said that he had faith in assumptions of science, because there seems not to be sufficient evidence to demonstrate them. Thus, Huxley has been seen as having double standards for his faith in science and against theological faith.117  The interpretations discussed in the previous section either generate the two problems or provide no solution to the problems.118 For this reason, Huxley’s promoting agnosticism has been considered to have damaged his conviction in scientific method. Yet, as I will discuss shortly and many scholars have pointed out, it seems obvious that Huxley used agnosticism to promote his conception of science and to raise his social status, and his agnosticism did make                                                           115 Balfour 1895; Ward [1899]1915; Flint 1903; Gilley and Loades 1981; Lightman 1987. 116 Flint 1903, 46.  117 For example, Flint 1903; Lightman 1987; Levine 2014. 118 This depends on the issue of how to understand agnosticism – whether Victorian scientific naturalism is ultimately constitutive of or independent of agnosticism. Thus sometimes the problems have been presented as criticisms of scientific naturalism or Victorian scientific naturalists’ empiricism.     36  some contributions to his project. This has invited research on benefits that Huxley would have gained by endorsing agnosticism, or roles that his promoting agnosticism played. Turner’s analysis on Victorian scientific naturalism has set up a concrete historical stage for such research: Victorian scientific naturalists’ struggle with established Anglican authority over scientific community and society in general.119   According to Turner, Victorian scientific naturalists promoted their conception of science, with the belief that the method and results of science can solve social problems and lead to social progress. This promotion happened to involve a kind of class struggle. The scientific community was then under the control of clerical scientists or theologically accommodating scientists who “could and did directly influence evaluation of work, patronage of research, and appointments in scientific institutions, the universities, and the public schools.”120 Many of Victorian scientific naturalists, as mentioned, were from the middle class without Oxbridge background or “proper” religion. To secure their position as a scientist and to propagate their conception of science, Victorian scientific naturalists had to undermine social and political influence of the Anglican establishment.  If we approach Huxley’s agnosticism as part of his larger project of promoting scientific naturalism, we can see some roles that agnosticism served for Huxley. First, take a look at Turner’s own discussion.121 Turner’s understanding of Huxley’s agnosticism ultimately falls under Type Two: Huxley’s agnosticism, the doctrine of metaphysical ignorance, was an application of the epistemological position that knowledge should be “verifiable by observable                                                           119 Turner 1974; 1975. 120 Turner 1978, 364.  121 Turner 1974; 1975. Turner’s analysis is not just about Huxley. Turner is more interested in investigating Victorian scientific naturalism of which Huxley was one of leading protagonists.  37  empirical facts.”122 Victorian scientific naturalists “had chosen” the work of John Stuart Mill as the epistemological basis of science.123Applying the epistemological view to some theological matters and further promoting an ignorance stance toward such matters in the name of “agnosticism,” according to Turner, were a strategically useful move for Victorian scientific naturalists. Turner highlights three advantages. First, “any religious belief that could not be sustained by verifiable empirical facts became open game for the agnostic challenge”; second, “agnosticism was an instrument for clearing away certain metaphysical remnants in practical scientific research”; “[b]y asserting that men lacked sufficient knowledge to decide whether the universe was material or spiritual or whether it was ruled by a deity, the agnostics rejected a culture and cultural values that depended upon answers to such questions.”124 Because of these advantages, their arguments for scientific enlightenment and against the dominance of Christian church could be convincing.  Turner’s understanding of Huxley’s agnosticism as “self-serving agnosticism” based on the agenda of Victorian scientific naturalists has been shared by many scholars.125 Strictly speaking, the terms “scientific naturalism” and “agnosticism” do not convey the same idea, but many scholars, including Turner, use “scientific naturalists” and “agnostics” interchangeably because in general they are co-extensive. Ruth Barton says, “Behind the agnostic was a defender of naturalism”; Roy MacLeod views scientific naturalism as an “agnostic ideology”; Martin                                                           122 Turner 1974, 19.  123 Ibid., 20.    124 Ibid., 21-2. When Turner makes the first point, he refers to Annan 1951 and Irvine 1968.   125 Ibid., 21. For example, MacLeod 1982; Barton 1983; Fichman 1997; Desmond 2001; Stanley 2011.  38  Fichman says, “Huxley was one of the first to see the polemical advantage of adapting agnosticism.”126  Turner has identified contributions of agnosticism mainly by using the idea that Huxley used agnosticism to undermine the authority of his social and political enemies, clerics and theologically oriented scientists. While keeping the basic historical setting of Victorian scientific naturalists’ struggle for their cultural authority, other scholars have added or revised contributions of agnosticism to Huxley’s larger project of promoting science, by uncovering other roles.127 For example, according to Adrian Desmond, Huxley’s agnosticism contributed to creating an image of scientific knowledge as neutral.128 As seen, White’s account tells us that Huxley’s agnosticism functioned as a tool for “sociability.”129 Lightman’s later study on Huxley’s agnosticism as a “Nonconformist Sect” shows how agnosticism furnished Huxley with a rationale for the nonconformists’ task of dismantling the established dominance of aristocratic Anglican church, which Huxley also shared.130 To conclude, we can summarize the reception of Huxley’s endorsement of agnosticism as follows: it made contributions to his cultural project of promoting science, but it was not, philosophically speaking, a good move because of the two problems that he imposed upon himself. Agnosticism helped him to increase his voice for science, but it cannot help him to show                                                           126 MacLeod 1982, 3; Barton 1983, 279; Fichman 1997, 103.  127 There has been growing concern over the simplicity of Turner’s framework – young middle class men of science competing with established clerics and theologians with aims to take over social and cultural authority; for example, it leads us to pay attention to conflicts, competitions, and divergence, and excludes or marginalizes other groups (see Lightman 2009; 2014; White 2014). This kind of concerns may be led by, or generate, different accounts of roles of Huxley’s endorsement of agnosticism, other than the role of attacking. Yet, the concerns involve characteristics of Huxley’s project of promoting science – say, belligerent, cooperative, or something else – rather than the existence of his project. Still, roles of Huxley’s endorsement of agnosticism have been explored within his larger project.         128 Desmond 2001. Also Moore 1988; Fichman 1997. 129 White 2003, 119. 130 Lightman 2004, 198.  39  science as the only way to knowledge. This may not be a surprise, because Huxley basically repackaged a well-known notion in philosophy with a catchy new label.    2.4. Inclusive approach to Huxley’s agnosticism  I take a different approach to Huxley’s agnosticism. The main aim of this dissertation is to take all seemingly constitutive elements of agnosticism into consideration and to unearth his epistemological – not social – project behind. I believe that my interpretation of Huxley’s agnosticism can resolve the two worries. This is because the worries involve elements of agnosticism: the first worry that concerns how to secure foundations of science is set up by the metaphysical ignorance doctrine; the second worry that concerns Huxley’s faith in science is set up by the agnostic principle.  Taking an inclusive approach strikes me as a desideratum if we aim to understand Huxley’s agnosticism as he intended, although many scholars have offered an account that focuses on only part of his discussions of agnosticism. If Huxley associated different ideas with agnosticism, associating itself would more illuminate Huxley’s notion of agnosticism than does each idea associated. To understand “associating” – why Huxley put different ideas together under the same label, all the ideas should be taken into consideration.   There have been few attempts to make sense of Huxley’s agnosticism in this way. An early attempt was made by David W. Dockrill.131 Dockrill admits that Huxley’s discussions are confusing and misleading, yet argues that Huxley’s agnosticism indeed consists of two elements: “the doctrine of the necessity of metaphysical ignorance” and “the acceptance of what Huxley                                                           131 Dockrill 1964; 1971.   40  regarded as the method of scientific inquiry.”132 In order to understand Huxley’s connection of the two ideas of agnosticism, Dockrill suggests, we should look at Huxley’s essay on Descartes.133 According to Dockrill, the essay shows that Descartes’s method of doubt, which is another representation of the agnostic principle, is the common root of the two ideas of agnosticism:  In epistemology, the method has led to the theory that man’s knowledge is limited to his own mental states and cannot reach to matters which lie outside them. Applied to physical science the method has given rise to the doctrine that all natural phenomena can be represented and understood in materialistic terms.134   Dockrill’s account of Huxley’s rationale for combining the two ideas appears to be interesting, but he offers no further clear explanation. I will return to Dockrill’s account in next chapter. For the current purpose, it should suffice to note that Dockrill understands the doctrine of metaphysical ignorance and the acceptance of scientific method as the consequences of applying Descartes’s method of doubt.   As for another attempt, we have Matthew Stanley’s account found in his recent book, Huxley’s Church and Maxwell’s Demon, although its main concern is not Huxley’s agnosticism. Stanley suggests taking agnosticism “seriously as a philosophical stance that can provide insight into Huxley’s thinking about the limits of science.”135 Stanley considers Lightman’s account of Huxley’s metaphysical ignorance doctrine to be “definitive,” and yet he further mobilizes the                                                           132 Dockrill 1971, 463, 461.  133 Thomas H. Huxley, On Descartes’ “Discourse Touching the Method of Using One’s Reason Rightly and of Seeking Scientific Truth” (1870). 134 Docrkrill 1971, 475.  135 Stanley 2014b, 81, my emphasis.   41  notion of the limited human faculties to make sense of Huxley’s association of agnosticism with the acceptance of scientific method.136  According to Stanley, “[a]cknowledging [human] limitations,” rather than scientific method based on observation and experiment, was what Huxley meant when he spoke of agnosticism as the “essence of science, whether ancient or modern.”137 The basic idea seems to be that Huxley drew various “limits” of human cognitive faculties from the thoughts of philosophers (Stanley introduces Locke, Hume, Kant, Hamilton and Mansel), and suggested that accepting all the limits is the “essence of science”: for example, “Locke’s warning against universal knowledge” is one of them.138 The scope of limits even includes historical failure of human activities. Huxley’s discussion of the unproductivity of an inquiry with “spiritualistic” or non-materialistic terminology is understood by Stanley to show one of human limitations.  Acknowledging the limits involves taking certain attitudes, and, Stanley appears to suggest, these are expressed in Huxley’s umbrella principle, “agnostic principle.” For instance, Huxley’s warning against certainty taken from Locke implies “an obligation to always have one’s mind open for new evidence.”139 Once we admit all the limits and put them together, we arrive at Huxley’s conception of science: “science should restrict itself to statements and ideas that were accessible to experimentation, quantification, provided for further investigation, and could be represented in materialist terms without complete allegiance to materialism.”140 Huxley’s agnosticism is an account of various human limitations and the “agnostic sense of”                                                           136 Ibid.  137 Ibid., 83; Huxley 1884a, 5.    138 Stanley 2014b, 82.  139 Ibid., 84.  140 Ibid., 87.   42  such limitations.141 The role of agnosticism for Huxley, then, was to limit scientific practice and thus Huxley could accuse some practice such as invoking God’s direct intervention to explain phenomena of being unscientific.142           As I said, I also take an inclusive approach to Huxley’s agnosticism. However, I further suggest, we need to see that his agnosticism more concerns knowledge than belief, and what we can know than what we cannot know. I believe that the two philosophical difficulties ultimately stem from the understanding of Huxley’s notion of agnosticism as an account of ignorance or limits. An account of “being ignorant” is not an account of “ignoring,” and thus agnosticism as an account of ignorance hardly makes sense of his strong claim about science.   It is my view that his agnosticism concerns how we can move from “being ignorant” to “ignoring,” and this is what Huxley intended to convey. His agnosticism, as I understand it, was a guide to knowledge consisting of an account of legitimate evidence and an ethics of knowing, which is supposed to work regardless of our ignorance. In following chapters, I examine whether there is a good reason to think that Huxley had a philosophical project behind his agnosticism, if he did, what it was, and how each idea associated with agnosticism works together. I hope to show that Huxley’s agnosticism was a constructive position about how to attain knowledge, rather than an idle position about our ignorance.                                                                141 Ibid., 84.  142 This interpretation leaves the possibility of having knowledge by other means, such as revelation. Huxley argued not only that science should be naturalistic, but also that because only scientific method can supply society with knowledge, science, not theology, should have social and cultural authority.    43  ▌CHAPTER 3  Huxley’s Cartesian Project    This chapter discusses what I call “Huxley’s Cartesian project” which can be described as an epistemology of inquiry. Huxley found Descartes’s inquiry into certainty illuminating, because Descartes had left us ways to knowledge: two “paths” of idealism and materialism, and his maxim (only accept something clear and distinct as certain). By mobilizing the three elements all of which he associated with agnosticism, Huxley attempted to come up with a method for identifying knowledge that would work despite our ignorance or limits. However, Huxley understood and combined the three elements in his own way. This chapter focuses on the first two elements, “paths.” Huxley’s defense for his method for identifying knowledge and the discussion of the third element will be addressed in the following chapters.   The important points of this chapter are as follows. First, Huxley’s discussions of Descartes shed considerable lights on Huxley’s notion of agnosticism. Second, Huxley’s own way to a Cartesian guide to knowledge is an epistemological journey without devising and relying on a metaphysical or theological argument. Third, Descartes’s first path puts us in the situation where we cannot but select some mental contents as knowledge by which we construct our best understanding of nature; Descartes’s second path shows us that we can have mental contents that have explanatory value by understanding given phenomena in a materialistic-mechanistic way.       44  3.1. Dockrill’s account of Huxley’s agnosticism  In his 1971 essay, David W. Dockrill points out a widespread assumption in understanding of Huxley’s agnosticism, and attempts to challenge it. According to Dockrill, Huxley indeed presented his agnosticism in two very different ways: first, agnosticism was presented as an account of metaphysical ignorance on the basis of the necessary limits of human faculties; second, agnosticism was presented as an account of metaphysical ignorance on the basis of the notion of scientific method as the only means of attaining knowledge. Since the two accounts are different account of human ignorance, commentators tend to assume that Huxley’s agnosticism should be limited to either of the two accounts. Dockrill admits that Huxley’s mode of presentation looks confusing, but, he argues, this does not have to show that he was inconsistent. Dockrill further admits that Huxley sometimes discussed only one of the two ideas, but, he explains, Huxley did so because of the context or topic of essays. According to Dockrill, Huxley’s agnosticism is the “combination” of the two ideas, because they have the same origin.143    Dockrill pays special attention to Huxley’s lecture essay on Descartes’s Discourse, because, as briefly noted in the previous chapter, he considered the essay to reveal the common root, Descartes’s method of doubt. Dockrill’s basic idea is that applying Descartes’s method of doubt results in two ideas and Huxley later presented them as the two components of his agnosticism. On one hand, the method results in the thesis of the necessity of metaphysical and theological ignorance; on the other hand, the method results in the thesis of scientific method as the only means of attaining knowledge. Thus, Huxley’s rationale for combining the two ideas                                                           143 Dockrill 1971, 474.  45  and labeling them with a single name is that they are consequences of employing the same method – Descartes’s method of doubt.  However, Dockrill offers no other clear illustrations on how Descartes’s method of doubt results in the two components. Instead, Dockrill introduces the following passage from Huxley:   In truth, Descartes’ physiology, like the modern physiology of which it anticipates the spirit, leads straight to Materialism, so far as that title is rightly applicable to the doctrine that we have no knowledge of any thinking substance, apart from extended substance; and that thought is as much a function of matter as motion is. Thus we arrive at the singular result that, of the two paths opened up to us in the “Discourse upon Method,” the one leads, by way of Berkeley and Hume, to Kant and Idealism; while the other leads, by way of De La Mettrie and Priestley, to modern physiology and Materialism. Our stem divides into two main branches, which grow in opposite ways, and bear flowers which look as different as they can well be. But each branch is sound and healthy and has as much life and vigour as the other.144  Without further explanations, Dockrill concludes that “Huxley’s stress on the common methodological basis of his theories of science and knowledge” supports the claim that Huxley intended his agnosticism to mean both ideas.145 This, according to Dockrill, undermines the widely shared assumption: “[t]he conflict between the accounts is not, despite appearances, between different claims about what agnostic doctrine is; rather it is a conflict between different ways of viewing this doctrine.”146     Huxley’s passage does little to help us to understand Dockrill’s account of how Huxley arrived at his double sided agnosticism from Descartes’s method of doubt. Dockrill’s earlier work partially suggests what he has in mind: the application of the method of doubt led Huxley to the so-called cogito argument, just as it had led Descartes himself; the acceptance of the                                                           144 Huxley 1870c, 190; also cited in Dockrill 1971, 475. 145 Dockrill 1971, 476. 146 Ibid., 477.  46  conclusion, cogito ergo sum, led Huxley to the doctrine of idealism that we can know nothing beyond phenomena, which would be in turn formulated as the first component of agnosticism that we cannot know metaphysical and theological matters.147 This appears to be one aspect of Huxley’s journey that Dockrill has in mind. Indeed, we can see Huxley discussing Descartes’s cogito argument directly in this regard (I will return to this later).  On the other hand, Dockrill’s second connection between Descartes’s method of doubt and the idea of scientific method as the only way of attaining knowledge is less straightforward. Dockrill seems to rely on the fact that Huxley considered the method of doubt to be “the great first commandment of science.”148 Dockrill’s idea seems to be that science which adopts a materialistic perspective can be said to be the practice of the method of doubt, and for this reason science is claimed to be the only way of attaining knowledge. However, in addition to an exposition of what it means for science to practice Descartes’s method of doubt, a further explanation should be provided to see how practicing the method of doubt is related to adopting a materialistic perspective, and further to the claim that scientific method is the only way of attaining knowledge.   Apart from the lack of explanations, it is not clear whether Dockrill’s account succeeds in resolving the scholars’ worry concerning the two different accounts of human ignorance, which would have motivated the adoption of a partial approach to Huxley’s agnosticism. Most commentators have not denied the fact that Huxley associated more than one idea with his agnosticism. Huxley’s agnosticism should mean only one of the two ideas, because, as we have seen in the previous chapter, they are distinctive accounts and the tension can be generated by                                                           147 Dockrill 1964. 148 Huxley 1870c; also cited in Dockrill 1971, 475.  47  holding the two accounts. We may grant Dockrill’s claim that the two components of Huxley’s agnosticism are the two results of applying Descartes’s method of doubt, and yet we can still maintain that the two results put Huxley in trouble. Since 1971 when Dockrill’s essay was published, the worry appears to have gained a bigger voice. Although Dockrill wrote that “Huxley’s inconsistency is less serious than most of his critics have thought,” it looks more serious than Dockrill thought.149    Nevertheless, I also think, with Dockrill, that Huxley intended his agnosticism to include both ideas. Furthermore, I believe that Dockrill is correct in claiming that Huxley’s essay on Descartes helps us to understand Huxley’s notion of agnosticism. I share Dockrill’s approach to Huxley’s agnosticism, however, I do not accept his account. My worry is that Dockrill’s account does not adequately support the interpretation that Huxley’s agnosticism is a combined thesis. To be sure, Dockrill states that Huxley’s agnosticism is a “combination” of the two ideas and once uses the expression “integrally related parts of the one whole” to characterize them.150 However, Dockrill’s notion of combination amounts to sharing a common origin. Furthermore, his claim that Descartes’s method of doubt was the common root cannot make sense of the following remark from Huxley: “Descartes’ two paths meet at the summit of the mountain, though they set out on opposite sides of it.”151 Pointing to Descartes’s method of doubt as the common root of the ideas of Huxley’s agnosticism can account for “diverging” but not for “meeting.” We need to examine if “combining” played a greater role.                                                              149 Dockrill 1971, 477. 150 Ibid., 475. 151 Huxley 1870c, 194.  48  3.2. The importance of Huxley’s discussions of Descartes  As Dockrill’s literature review shows, Huxley’s essays concerning Descartes have been neglected by the commentators most interested in Huxley’s agnosticism.152 This is unfortunate because these essays shed considerable lights on the nature and structure of his agnosticism.  To be sure, Huxley did not mention the terms “agnostic” and “agnosticism” in his essay on Descartes’s Discourse, which was published in 1870. Nonetheless, given that Huxley later stated that he had invented the term around 1869 and that, as mentioned in the previous chapter, Huxley had held the position from his earlier days without the label, it would be highly probably that Huxley had agnosticism in mind when he was writing the Descartes essay.  Even though Huxley held agnosticism in his earlier days, one might point out, this fact does not necessarily mean that Huxley’s Descartes essay reflects his conception of agnosticism. However, as we will see soon, his Descartes essay reveals the structure underlying agnosticism, not just fragmentary ideas. I see this essay as his proposal for how he would like to reconstruct his agnosticism as a more or less unified position. We can regard the essay as a place where Huxley took an occasion to organize loosely related but separated thoughts.   Huxley started his essay on Descartes’s Discourse by introducing a metaphor for “the intellectual filiation of mankind”: instead of the metaphor of “one great chain,” the “thoughts of                                                           152 Dockrill 1971. Huxley discussed Descartes in more than one essay (e.g., Huxley 1868a; 1878h; 1878d; 1881). Although Dockrill’s paper was published in 1971, his conclusion still holds. Salvatore Vasta’s 2012 paper would be one addition, because he discusses Huxley’s philosophical project in the Descartes essays; he connects Huxley’s Descartes essays not with agnosticism, but with his efforts to undermine Comtean positivism. Also, to be sure, Huxley has been sometimes mentioned by philosophers of mind and neuroscientists as a proponent of epiphenomenalism (about the mind-body problem), as a follower of Descartes who offered a physiological and chemical account of vital phenomena, or as the owner of the famous Aladdin lamp illustration about so-called “Hard Problem” (e.g., Kim 2008; Block 2002; Gulick 2014). In this context, Huxley’s essay on Descartes’s Discourse and other related essays are often cited. However, most of them simply introduce Huxley’s expression and his essays in passing without discussion. Notable exceptions are Dennett 1991, N. Campbell 2001, and Greenwood 2010.    49  men seem rather to be comparable to the leaves, flowers, and fruit upon the innumerable branches of a few great stems, fed by commingled and hidden roots.”153 Then, he wrote:  It seems to me that the thinker who, more than any other, stands in the relation of such a stem towards the philosophy and the science of the modern world is Réné Descartes. I mean, that if you lay hold of any characteristic product of modern ways of thinking, either in the region of philosophy, or in that of science, you find the spirit of that thought, if not its form, to have been present in the mind of the great Frenchman.154   What Huxley meant by “modern ways of thinking” are idealism and materialism. The modern way of thinking in philosophy is the adoption of the doctrine that we cannot go beyond phenomena, and the modern way of thinking in science is the adoption of a materialistic perspective. Huxley later claimed that the “two paths” to truth introduced in Descartes’s Discourse had led to the two modern ways of thinking.   One and the same person, Descartes, had exhibited the spirits of both ways of thinking. Since then, Huxley observed, “Our stem divides into two main branches, which grow in opposite ways, and bear flowers which look as different as they can well be.”155 After Descartes, thinkers had taken only one of the two ways of thinking. As Huxley called them, they were either “metaphysical thinkers” or “physical thinkers.” Although branching had led to development of each way of thinking, the state of exclusiveness was not ideal to Huxley.    If a botanist found this state of things in a new plant, I imagine that he might be inclined to think that his tree was monœcious – that the flowers were of different sexes, and that, so far from setting up a barrier between the two branches of the tree,                                                           153 Huxley 1870c, 166. 154 Ibid., 167. This is the only occasion when Huxley spells “Réné.” Huxley sometimes misspelled a name. He was once accused of pretending to have read Mansel’s book, because he wrote “Mansell” (Huxley 1895a). I think that his misspelling should not be considered to indicate his unfamiliarity with authors whom he addressed.    155 Ibid., 190.  50  the only hope of fertility lay in bringing them together. I may be taking too much of a naturalist’s view of the case, but I must confess that this is exactly my notion of what is to be done with metaphysics and physics. Their differences are complementary, not antagonistic; and thought will never be completely fruitful until the one unites with the other.156  The way to fruition is to unite the two ways of thinking. That is why Huxley found Descartes worthy of attention, rather than others who had employed just one of the two ways of thinking.   Later in the essay, Huxley hinted at how to unite metaphysical and physical thinkers:  The reconciliation of physics and metaphysics lies in the acknowledgment of faults upon both sides; in the confession by physics that all the phænomena of Nature are, in their ultimate analysis, known to us only as facts of consciousness; in the admission by metaphysics, that the facts of consciousness are, practically, interpretable only by the methods and the formulæ of physics: and, finally, in the observance by both metaphysical and physical thinkers of Descartes’ maxim – assent to no proposition the matter of which is not so clear and distinct that it cannot be doubted.157   What physical thinkers need to confess seems to be related to the thesis of metaphysical and theological ignorance based on conditioned human faculties; what metaphysical thinkers need to admit seems to be related to the thesis of scientific method as the only way of attaining knowledge; finally, Descartes’s maxim is the rule complied with by those whom Huxley regarded as past agnostic thinkers like Socrates. Given that the three items that Huxley picked up from Descartes look closely related to the three ideas underlying his agnosticism, it seems clear that                                                           156 Ibid, 190-1. 157 Ibid., 194. The “faults” of metaphysical and physical thinkers were the reasons why Huxley refused to endorse the metaphysical versions of Idealism and Materialism. According to Huxley, “the Idealist, not content with declaring the truth that our knowledge is limited to facts of consciousness, affirms the wholly unprovable proposition that nothing exists beyond these and the substance of mind,” whereas “the Materialist, holding by the truth that … material phenomena are the causes of mental phenomena, asserts his unprovable dogma, that material phenomena and the substance of matter are the sole primary existences” (Huxley 1879a, 318; 1863c; 1868a).   51  our understanding of his agnosticism would benefit from an examination of why and how he mobilized the three Cartesian items.   Huxley appreciated Descartes not only because Descartes had introduced and followed two paths leading to the modern ways of thinking, but also because, as Huxley quoted Descartes, he “always had an intense desire to learn how to distinguish truth from falsehood in order to be clear about [his] actions, and to walk surefootedly in this life.”158 To Huxley, Descartes sought for a method for identifying knowledge and did so to live; he was neither a dogmatist of traditional beliefs nor merely a destructive sceptic. Satisfying the “intense desire,” Huxley appears to have thought, has something to do with Descartes’s having two paths (and “maxim” which I will discuss later). Huxley, who also shared the desire, wished to have a guide to knowledge, by uniting the two modern ways of thought.   3.3. Two paths that Descartes opened  Before examining Huxley’s own way of using the two modern ways of thought, we need to look into his claim that Descartes’s Discourse shows the “spirits” of each of them. The plausibility of his interpretation of Descartes is not my concern. Huxley’s discussions can clarify components of his agnosticism and notions that he used to make claims about agnosticism, thereby preparing us to see how his guide to knowledge is supposed to work.   3.3.1. The first path: legitimate Idealism  First consider the modern way of thinking in philosophy. Huxley called it “legitimate Idealism,” according to which “whatever the universe may be, all we can know of it is the picture presented                                                           158 Cited in Huxley 1870c, 168.  52  to us by consciousness.”159 Huxley claimed that the “ultimate issue” of Descartes’s cogito argument is the doctrine of legitimate Idealism.160  Descartes had tried to find something certain in the sense of being indubitable, and thus he had exercised the so-called “method of doubt.” The basic idea is that if we cannot doubt something, we would be entitled to be certain about it. We can doubt a given belief, according to Descartes, if we can entertain the possibility that it is not true. Huxley illustrated this process as follows.   As the record of his [Descartes’s] progress tells us, he was obliged to confess that life is full of delusions; that authority may err; that testimony may be false or mistaken; that reason lands us in endless fallacies; that memory is often as little trustworthy as hope; that the evidence of the very senses may be misunderstood; that dreams are real as long as they last, and that what we call reality may be a long and restless dream. Nay, it is conceivable that some powerful and malicious being may find his pleasure in deluding us, and in making us believe the thing which is not, every moment of our lives. What, then, is certain?161  As is well known, Descartes eventually found one thing that he could not doubt: he was doubting. Doubting is one type of thinking, and this finding is applicable to any other types of thinking activity. Whatever we are thinking, we cannot deny that we are thinking. Thus, while thinking, we cannot doubt the existence of us thinking: cogito ergo sum.  According to Huxley, however, Descartes had not carried out the method of doubt thoroughly. When Descartes was arriving at “I am thinking, therefore I exist,” he should have pushed further. Huxley first pointed out that the term “therefore” plays no role in Descartes’s reasoning because what Descartes had found certain is the existence of a “thinking I.” Huxley                                                           159 Ibid., 178. 160 Ibid., 176. 161 Ibid., 172.  53  then claimed that Descartes’s conclusion consists of three different theses: “something called I exists,” “something called thought exists” and “the thought is the result of the action of I.”162 According to Huxley, only the second thesis can pass the method of doubt. It would be an important issue in the literature on Descartes how “I” should be understood, but Huxley understood that Descartes’s term “I” was supposed to refer to some entity governing thoughts (“masterful entity, the Ego”), something more than “I” being momentarily aware of itself thinking.163 Huxley thought that the existence of such an entity can be doubted. Also, he claimed that we can entertain the possibility that thoughts are not the result of the action of such an entity. On the other hand, doubting that a thought exists proves the existence of a thought. Thus, according to Huxley, the method of doubt leads us to the conclusion that we cannot doubt the existence of thoughts.164 What does it mean that we cannot doubt the existence of thoughts? According to Huxley, “Descartes uses ‘thought’ as the equivalent of our modern term ‘consciousness.’”165 It refers to mental phenomena or states of consciousness (regarding the latter expression, he added: “‘Consciousnesses’ would be a better name, but it is awkward”).166 Whatever we are aware of – thoughts, feelings, or sensations − is something that certainly exists. It cannot be doubted that the states of consciousness exist, because doubting itself is such a state.                                                             162 Ibid., 177. 163 Ibid.; Huxley 1878h, 87. 164 Huxley had made the same point earlier in his letter to Kingsley: “Cogito, ergo sum is to my mind a ridiculous piece of bad logic, all I can say at any time being ‘Cogito.’ The Latin form I hold to be preferable to the English ‘I think,’ because the latter asserts the existence of an Ego – about which the bundle of phenomena at present addressing you knows nothing” (1863c, 349). 165 Huxley 1881, 359. 166 Huxley 1878h, 87, n18; 1870a; 1870c; 1874a; 1880a.   54  Thus, the lesson of the revised argument of Descartes is, according to Huxley, that “a thought is existence.”167 The conclusion that we cannot doubt that thoughts exist means that thoughts are “real and existent.”168 To be sure, this does not mean that we can take the content of thoughts as true, as Huxley said, “thoughts may be delusive, but they cannot be fictitious.”169   The lesson, Huxley conjectured, had been further developed into the doctrine that “existence is thought.”170 Whatever we conceive as an existence or being takes a form of thought. When we say that a thing or property exists, what we can say with certainty is only that it exists as a state of consciousness, and our understanding of a thing or property comes from contents of thoughts. To turn “a thought is existence” into “existence is a thought,” one needs to show that nothing can be known to exist to us without being a thought. This is, according to Huxley, what George Berkeley tried to show when he argued that it is contradictory to talk about an unperceived being.171 Agreeing with Berkeley, Huxley also found it impossible for us to think of unperceived objects or properties. However, Huxley’s concern is entirely epistemological. Unlike Berkeley, he did not further claim that the state of consciousness is the only ontological mode that objects and qualities take because this is an extra point that overstates the epistemological point that all we can come to know is what appears to our consciousness; it cannot pass the method of doubt, and thus cannot be taken as certain. Huxley’s “legitimate Idealism” refers to an epistemologically understood idealism, not metaphysical idealism.                                                            167 Huxley 1870c, 172. 168 Ibid. 169 Ibid. 170 Ibid. 171 Huxley 1871c.   55  As Huxley saw it, Descartes’s cogito argument had been advanced to Berkeley’s famous thesis that to exist is to be perceived. The advancement had gone further: if we highlight our capability or incapability implied in Berkeley’s thesis, we would arrive at a Kantian idea that we are capable of knowing only what is experienced or the phenomenal.172  Huxley’s understanding of the modern way of thought in philosophy will help us later to understand the first component of his agnosticism more precisely. Huxley mentioned Kant, Hamilton, Mansel, and Spencer as those whom he thought had endorsed legitimate Idealism, but Descartes and Berkeley were those whom Huxley discussed to explain a way to the notion.     3.3.2. The second path: legitimate Materialism  The other path opened in Descartes’s Discourse is what Huxley called “physical ways of thought.” According to Huxley, physical thinking involves adopting a materialistic-mechanistic view that seeks to offer a causal or structural account in material terms. As Huxley noted, Descartes’s predecessors and contemporaries had employed this way of thinking to physical phenomena. However, Descartes turned it into a “grand conception”; he “sought to resolve all the phenomena of the universe into matter and motion, or forces operating according to law.”173 Huxley also referred to Descartes’s other essays where we can see a more detailed development of the materialistic-mechanistic view, but he thought that Descartes’s discussion of blood circulation in Discourse sufficiently shows the view.  Huxley appears to have thought that the materialistic-mechanistic view comes in degrees, reflecting three component ideas. First, it involves accounting for a given phenomenon as a                                                           172 Huxley 1878h.  173 Huxley 1871c, 181.  56  physical operation or its outcome, with the idea that a materialistic-mechanistic account can do explanatory work. Second, it involves the notion of a materialistic-mechanistic account as a sufficient account. These first two points, as we will see later, partially constitute Huxley’s reason for favoring physical thinking over non-physical thinking. Last, it involves commitment to its universal applicability. At this point, a materialistic-mechanistic view becomes a materialistic-mechanistic worldview. According to Huxley, Descartes should be considered more “modern” than William Harvey, because Harvey had only endorsed the first conception whereas Descartes also endorsed the second and a qualified version of the third.174 In the following, I will illustrate each idea in turn.  Consider the first idea. Physical thinkers attempt to explain a given phenomenon in material terms and in a mechanical way. To offer a materialistic-mechanistic account is to consider a given phenomenon to be an operation that results from a particular arrangement of matter and motions. For example, according to Descartes (understood by Huxley), Harvey, known to discover the circulation of the blood, had offered a physical account of how the blood circulates in terms of valves, muscles, and contraction. The heart, the main organ responsible for the phenomenon, is seen as a physical-chemical machine. Seemingly holistic processes or parts are thought to be decomposable into a set of more basic motions and elementary physical units, and their combination or coordination instantiate a particular mechanism. To identify the correct mechanism among possible, imaginary mechanisms, numerous observations and dissections need to be done. These are what Harvey had done, and Descartes also did the same work.  According to Huxley, a materialistic-mechanistic account has explanatory value in the sense that it renders a given phenomenon intelligible. For Huxley, to explain or to make a thing                                                           174 Huxley 1871c; 1874a; 1881.   57  intelligible is to show a causal chain leading to a phenomenon expressed in an explanandum.175 Here, showing a causal chain does not merely mean enumerating events in a temporal order; it also involves unpacking events in terms of relevant objects’ properties and behaviors so that a particular series of events can be determined. For example, to explain the circulation of the blood, one may speak of a certain event, say, muscles squeezing the blood; but one may further show how such an event can happen by utilizing, say, properties and behaviors of the muscles, valves, and blood. In this way, events are concatenated in a certain way, and not in other ways. In this manner, Huxley argued, a mysterious and holistic phenomenon, the circulation of the blood, becomes “at once intelligible” as it is shown to be an invariant or determinate effect of a particular set of interrelated local events; the function and importance of the valves become “at once apparent.”176  Let’s turn to the second idea. Huxley’s physical thinkers also believe in the sufficiency of a mechanical account using material objects and properties. In other words, physical thinkers refuse to appeal to a special agent, essence, quality or organ that is usually assumed to be non-material and govern a whole process, which some people claim completes a materialistic-materialistic understanding of phenomena. This point is well illustrated in Huxley’s discussion of the difference between Harvey and Descartes.  According to Huxley, the “founder of modern physiology” is Descartes, not Harvey.177 Although Harvey had influenced Descartes and Descartes praised him in the Discourse (and moreover, as Huxley pointed out, Harvey was right and Descartes was wrong about the cause of                                                           175 Huxley 1880a. According to Huxley, “[a]nything is said to be explained as soon as we have discovered its cause, or the reason why it exists; the explanation is fuller, if we can find out the cause of that cause; and the further we can trace the chain of causes and effects, the more satisfactory is the explanation” (Ibid., 7).  176 Huxley 1878b, 335. 177 Huxley 1881, 356; 1874a.  58  the blood circulation), Harvey belonged to the “ancient physiology” tradition. This is because Harvey invoked immaterial or animistic stuff to account for vital phenomena, as the ancient had done.178 Huxley’s point concerns Harvey’s discussion of “innate heat.” It is a kind of vital power that is thought to reside in the blood, which Harvey invoked to explain why the heart circulates the blood at all. According to Huxley, Harvey conceived the blood as “the seat of a soul,” and, with its power, “maintains and fashions all parts of the body.”179 What Huxley found “ancient” about Harvey is the idea that organs or local processes related to a vital activity need a soul or some special entity to operate. The invocation of non-materialistic entities or properties, according to Huxley, usually adds no explanatory value to a given causal account but simply exemplifies the “ingrained tendency of the human mind to suppose that a process is explained when it is ascribed to a power of which nothing is known except that it is the hypothetical agent of the process.”180 In short, Huxley appears to conclude that Harvey did not regard the heart as an automatic machine although he thought that it is working like a machine.  On the other hand, Descartes clearly conceived of the heart as a machine. That is, Descartes considered a materialistic-mechanistic explanation to be sufficient. Huxley quoted Descartes to highlight his manner:     “I [Descartes] shall try to explain our whole bodily machinery in such a way that it will be no more necessary for us to suppose that the soul produces such movements as are not voluntary than it is to think that there is in a clock a soul which causes it to                                                           178 Huxley 1870c; 1874a; 1878b; 1881. 179 Huxley 1881, 357.    180 Ibid. However, this does not mean that Huxley underestimated Harvey. In his “William Harvey,” Huxley was at pains to show the significance of Harvey’s achievement and to clear away some misconceptions about him and his theory. Huxley was firm in his view that Harvey had made contribution to modern physiology because Harvey was the first person who took quantitative considerations into account (for example, “the comparison of the quantity of blood driven out of the heart, at each beat, with the total quantity of blood in the body”), but cannot be regarded as a modern physiologist (1878b, 335; 1878c).          59  show the hours.” These words of Descartes might be appropriately taken as a motto by the author of any modern treatise on physiology.181   According to Huxley, Descartes was, because of his strict dualism, “logically compelled to seek for the explanation of the phenomena of the material world within itself” where he saw “nothing but extension and motion.”182 At least in the case of behaviors of living things without a soul, materialistic-mechanistic accounts must suffice. This idea, Huxley suggested, had left modern physical science with the two options: “[i]t offers physical explanations of vital phenomena, or frankly confesses that it has none to offer.”183 Lastly, Huxley’s conception of physical thinking involves an attitude of expanding a materialistic-mechanistic approach. According to Huxley, Descartes approached vital phenomena that had not been traditionally accounted for solely in material and mechanical terms, with the clear “conception that the physical universe, whether living or not living, is a mechanism, and that, as such, it is explicable on physical principles.”184 However, Descartes did not expand his materialistic-mechanistic approach to mental phenomena. For Descartes, thoughts are the operation of a soul. In this sense, Huxley concluded, Descartes was not a fully physical thinker, although he did expand the scope of a materialistic-mechanistic approach to vital phenomena. Those whom Huxley regarded as the followers of Descartes’s second path had gone farther than Descartes. They had attempted to explain the operations of “thinking organ” in terms of matters and motions of a brain and other physical parts. Descartes had thought that animals are                                                           181 Huxley 1881, 362. Huxley, however, noted that the analogy between a living body and a machine like a clock is not perfectly apt, because it can suggest that “there is a central source of power and the parts of the machine are merely passive distribution of that power” (Ibid., 362-3; also, see 368-9).   182 Ibid., 359. 183 Ibid., 358. 184 Huxley 1878d, 205.  60  unconscious automata whereas human beings who have a soul are not thus automata; to Huxley, animals are conscious automata and human beings are one of them. The scope of physical thinking had been larger enough to lead to “a robust faith in the universal applicability of the principles laid down by Descartes.”185  We have seen what Huxley meant to suggest by introducing the two paths, legitimate Idealism and legitimate Materialism, which he found in Descartes’s Discourse. However, I have not clearly identified how Huxley wanted to mobilized the two paths and why he thought that they should be structured in a guide to knowledge. Descartes would have his own reasons, but Huxley did not agree with all of his reasons. For example, Descartes’s metaphysical commitment to dualism did not persuade Huxley to apply physical thinking to vital phenomena. The following section discusses Huxley’s view of how the two paths can contribute to our search for knowledge.   3.4. Huxley’s reshaping of Descartes’s project  I have said that Huxley wanted to find a method for identifying knowledge. How did Huxley understand the notion of knowledge? Since he accepted Descartes’s two paths to knowledge, we may also expect that he would have adopted Descartes’s notion of knowledge, and indeed he did.186 Descartes had understood knowledge in terms of certainty, which had been in turn understood in terms of the impossibility of doubting. Huxley adopted that notion of knowledge. This is also reflected in the fact that he spoke of certainty in most of the formulations of agnosticism or its principle.                                                           185 Huxley 1881, 371. 186 Huxley 1870c; 1878h.  61  Yet, a further clarification is needed because the notion of being certain is ambiguous. It can mean a mental state (having no doubt) or a property of a proposition (being true). One issue that can arise immediately would concern a relation between them. To Huxley, our being certain of p is prior to p being certain in the sense that we reach the conclusion that p is certain from our being certain of p.187 If we are in the state of having no doubt regarding the truth value of p, p is considered to be true, and we have knowledge.  However, we cannot freely switch from our feeling of certainty to our state of being certain. Sometimes we feel certain even though there are reasons that show we should not be certain. According to Huxley, to know p is to be certain of p, thus we need to make sure whether we are in the state of being certain or just feel certainty.188 Phenomenologically speaking, it might be hard to distinguish between being certain and feeling certain. The notion of being certain should be understood normatively. It is one thing that we feel certainty and it is another thing that we deserve to feel certainty. If we want to find something certain, something that deserves the title of knowledge, we should not simply point to our feeling of certainty but justify it. In other words, being justified is the most important connotation of the concept of knowledge.   As we have seen, Huxley shared Descartes’s concern: we need a guide that leads us to knowledge. Huxley also wanted a guide that basically has Descartes’s two paths of Idealism and Materialism (and maxim). If so, then Huxley could have just embraced all of Descartes’s claims. However, he did not. As we have seen, he did not accept Descartes’s conclusion of the cogito argument (similarly, he dismissed Berkley’s metaphysical idealism as well). As we will see, he did not think that a guide to knowledge needs to be supported by a theological argument that is                                                           187 Huxley 1878h; 1889e. 188 Huxley 1878h; 1889e; 1889g.  62  supposed to assure us of the reliability of our faculties such as Descartes’s argument from undeceiving God.  Huxley’s selective endorsement of Descartes (and other modern philosophers) means two things. First, Huxley would have needed to offer a different story for how the two paths of idealism and materialism are to be incorporated into a guide to knowledge, which I will illustrate soon. Second, Huxley seems to have thought that he could make a case for his guide to knowledge without constructing metaphysical or theological underpinnings. Descartes likened philosophy to a tree and metaphysics to its root that constitutes principles of knowledge.189 For Huxley, as I introduced earlier, Descartes himself is a stem “fed by commingled and hidden roots,” from which the branches of Idealism and Materialism have diverged.190 Huxley wanted them to be commingled again. As he put it, the task is to “breed” the flowers of “monoecious” trees to produce “fruits.” Huxley seems to have thought that we, not a metaphysical or theological system, do the breeding, and he could get our agreement on his guide to knowledge because his discussion is epistemological and based on something that we all can be certain about.    According to Huxley, if we employ the method of doubt, we would be able to arrive at irrefragable facts which Huxley labeled as “legitimate Idealism,” which were introduced in the previous section.191 One of them is the revised conclusion of the cogito argument: the certain existence of the states of consciousness. In other words, what appear in consciousness – either propositional or non-propositional contents – are real and existent. For example, when I think of ice being cold, the mental content that ice is cold exists in my consciousness and I am not deluded about the content appearing in my consciousness; similarly, when I feel pain, the                                                           189 Descartes 1982 [1644]. 190 Huxley 1870c, 166. 191 Huxley 1886c.  63  sensation of pain exists in my consciousness and I am not deluded about the sensation appearing in my consciousness.  The other “irrefragable” fact is Berkeley’s thesis that to exist is to be perceived. To Huxley, Berkeley’s message is that all we come to know about existence ultimately comes from what consciousness shows us. We speak of objects such as marbles and their qualities like solidity, but our understanding of them is about how they appear to us. Huxley wrote that “whatever may, or may not, exist in the thing [marble], all that we can know of these qualities is a state of consciousness.”192 According to Huxley, Berkeley’s view can be best received as an epistemological effort to search for “the limits of our faculties,” which should not be carelessly dismissed by “stamping on the ground.”193 These two “irrefragable” facts, Huxley urged, bear on the search for knowledge. First, because it is certain that the states of consciousness exist, if we make use of the states of consciousness, we can be assured that we are dealing with something real and existent (which may or may not be “delusive,” as Huxley said). On the other hand, the second fact tells us that the stream of consciousness is all we can have, because it denies the possibility that we can come to know something that does not take a form of the state of consciousness. Jointly, the two facts inform us that we cannot but make use of what appears to consciousness, but what we are about to make use of is something that certainly exists. To Huxley, our knowledge is “knowledge of states of consciousness.”194 Incorporating the two irrefragable facts into the search for knowledge hints at how Huxley would like to proceed. One may take the certain existence of thoughts as the first thesis and want                                                           192 Huxley 1870c, 175. 193 Huxley 1871c, 251. 194 Huxley 1870c, 176.  64  to examine what can be inferred from it. Valid reasoning will make us sure that its certainty is transmitted along with the reasoning. However, this is not the way that Huxley proceeded. What we learn is that each thought certainly exists. The “first thesis” tells us where to look to find more knowledge although such knowledge would not be certain in the highest degree.    As just implied, for Huxley, there are two kinds of knowledge: immediate and mediate knowledge.195 What Huxley labeled as “immediate knowledge” or “intuitive knowledge” is about what happens in our consciousness. Immediate knowledge has two characteristics. First, because we cannot doubt our having a state of consciousness, this type of knowledge enjoys the highest degree of certainty, indubitability.196 Second, this type of knowledge is known by virtue of being in a state of consciousness. In other words, we need not come up with some special method to have immediate knowledge.  A method for identifying knowledge, which Huxley wanted to find, is a method that we need to have what he called “mediate knowledge.” Immediate knowledge would take the form of, say, “It seems to me that the marble is round,” whereas mediate knowledge would take the form of “The marble is round.” We cannot, of course, simply claim that any mental content is mediate knowledge, and thus we need a criterion or method for distinguishing mediate knowledge from not mediate knowledge. According to Huxley, mediate knowledge, unlike immediate knowledge, is not certain in the sense of being impossible to doubt.197 Instead, mediate knowledge is certain in lesser degree, in the sense of being implausible to doubt or having no reason to doubt.198                                                           195 Huxley 1879a, 319, n 6. 196 Huxley 1870c, 178. Huxley interchangeably used “indubitable” and “irrefragable” (1860d; 1862c; 1868a; 1871c, 279; 1874a; 1874b; 1878h; 1886c).   197 Huxley 1870a; 1879a.   198 Huxley’s notion that certainty comes in degree can be summarized as follows. If either we cannot doubt p or we have no reason or grounds for doubting p, we are justified to be certain of p and thus p deserves the title of either  65  Huxley loosened the notion of being beyond doubt to make room for less certainty that mediate knowledge would enjoy.  The motivation of Huxley’s Cartesian project, then, can be re-illustrated as follows: we want to make use of contents of immediate knowledge to have mediate knowledge. Thus, he asked the following question: “now the question arises, whether any, and if so what, portion of these contents of the mind are to be termed ‘knowledge?’” 199 Numerous and various thoughts appear in consciousness. It is certain that I am in the state of having, for example, the following mental contents: a benevolent god exits; a red elephant starts running when it sees the moon; my finger hurts; there must be a tree at the corner. Among all possible mental contents present to us, Huxley wanted to select mental contents that deserve the title of (mediate) knowledge (hereafter, “knowledge” will mean mediate knowledge unless otherwise noted). How can we find such mental contents? One might try to single out mental contents that correctly capture reality. However, we have not been given any metaphysical or theological truth that can help us. Huxley did not welcome the suggestion because he did not think that there are metaphysical or theological facts that can pass the method of doubt with the highest standard. Furthermore, given the irrefragable fact about our limited condition, he did not think that we can come up with a way of identifying mental contents that convey a genuine description of something outside our consciousness or a god’s plan. Instead, Huxley suggested looking for a way of identifying mental contents that convey a trustworthy connection of mental states.   To see what Huxley suggested, let’s focus on the state of consciousness as a mental state. When we are in a state of consciousness, a particular mental content appears to us. We can speak                                                           immediate knowledge or mediate knowledge. Accordingly, Huxley’s method of doubt involves either checking the possibility of doubting p or checking the plausibility of doubting p by examining reasons to doubt p. 199 Huxley 1878h, 85.   66  of a state of consciousness in terms of a mental content, but we can also talk about a state of consciousness as a mental state. Huxley drew attention to modes in which mental contents are present to consciousness by likening consciousness to a kaleidoscope. Mental contents appear and disappear in succession, and a stream of contents looks sometimes orderly and sometime disorderly.200 In other words, all possible mental contents are not present to us at once. Huxley thought, because the stream of mental contents makes up “our whole life” and we cannot go beyond the stream, it is good enough for us if we can make sure which mental state follows or accompanies with which mental state(s).201 In a sense, Huxley aimed to find arrangements about mental streaming and looked for mental contents that capture such arrangements in a reliable and useful way to us. With this type of knowledge, Huxley stated, borrowing Descartes’s expression, “we are enabled ‘to walk surefootedly in this life.’”202 Huxley’s suggestion that we should aim for having mental contents that convey trustworthy arrangements of mental states, as opposed to those that convey correct representations of something outside consciousness, raises questions. One question would be whether mental contents that “merely” convey an arrangement of mental states indeed deserve to be considered “knowledge.” The following chapter addresses this issue. We need to see first how such mental contents can be identified according to Huxley.   For the task, it is necessary to organize the stream of mental contents. Huxley did not think that this is a strange idea. We have been, without much awareness and via unknown processes, sorting out mental contents by using concepts, and this is one type of organizing. For                                                           200 Huxley 1878h. 201 Huxley 1870c, 176. 202 Ibid., 178.  67  example, some contents are about what we call “self” and others are about “non-self.”203 To Huxley, self and non-self are hypothetical entities that play a certain explanatory or categorizing role. For instance, take the concept of self: self has been deemed to consist of corporeal and mental parts and to interact with non-self, as opposed to “ego” which is supposed to only mean the mental part of self.204 Another example relevant to the current section is the distinction between the material and the immaterial. Based on our notions of material object (matter) and immaterial object (spirit), we group mental contents into contents about the material and contents about the immaterial.205 Thus we have been saying, “All the phenomena of nature are either material or immaterial, physical or mental.”206  As the expression “the phenomena of nature” just quoted shows, a concept of nature is used. Huxley observed that people say that they live in nature and various things happen around them; they also say that mental contents are about nature. This, for Huxley, is another act of organizing mental contents. Organizing mental contents involves interpreting mental contents, that is, regarding them as information about something. We accommodate the stream of mental contents by situating a conscious agent or “thinking thing” within a cosmological framework and this kind of stage set is, according to Huxley, what people have called “nature” (or “universe” or “world”). Huxley’s metaphorical expression, “the shifting patterns of a kaleidoscope” becomes “the shifting scenes of the world’s stage” or “the shifting scenes of the phantasmagoria of Nature.”207 To organize mental contents, we use the minimal conception of nature as a framework                                                           203 Huxley 1870c, 176; 1879a.   204 Huxley 1879a.  205 To Huxley, a material object is “[e]verything which occupies space, offers resistance, has weight, and transfers motion” and an immaterial object is everything that is not a material object (1880a, 92). 206 Ibid., 94.  207 Huxley 1878h, 73; Huxley 1892c, 3.   68  accommodating consciousness, and fill nature with hypothetical, often loosely defined, concrete and abstract entities such as trees, water, material objects, self, non-self, etc. The specific layout and makeup of nature, according to Huxley, have been serious issues throughout human intellectual history among religious and philosophical thinkers.     Huxley conjectured that people had shared this way of understanding of nature, but there is one point to highlight regarding his minimal conception of nature. Conscious agents are cast to play both roles of “interpreters” of nature and of “actors” in nature.208 Huxley’s metaphor expands to illustrate how we should conceive ourselves: we are looking into a kaleidoscope; but also we have “got between the bits of glass of the kaleidoscope.”209  This point deserves attention because it can help us to see how Huxley saw the role of the concept of nature in his epistemological project. To proceed with an inquiry on what to count as knowledge, we organize mental contents while assuming that we are part of events in nature and objects in nature are responsible for our having mental contents. This assumption need not be shown to be true, because some mental contents will not be judged as knowledge because of their being correct descriptions of nature. If they were, one might justly claim that Huxley ended up following the suggestion he did not welcome by taking it for granted that, for example, we have mental contents and they represent nature. To Huxley, we conclude that some mental contents are correct descriptions of nature after judging them as knowledge. In next chapter, we will see his proposal for an internal way of determining knowledge. For now, it should suffice to note that the concept of nature is used to translate mental contents into objects or events in nature so that we can have a systematic understanding of nature.210 That is to say, Huxley’s epistemological inquiry                                                           208 Huxley 1887a, 81. This point has been neglected; an exception is Eng 1978. 209 Huxley 1887a, 74.  210 Huxley 1880a.  69  concerns not a relation between a “real” nature and a phenomenal nature made up of our everyday mental contents, but a relation between such a familiar phenomenal nature and a constructed nature or what Huxley called a “new Nature” made up of trustworthy mental contents or knowledge.211  Descartes’s first path led Huxley to the task of identifying trustworthy mental contents among all possible mental contents. This task involves examining a given mental content to see if it organizes mental states in a reliable and useful way. At this point, we meet Descartes’s second path, physical thinking. We may organize mental contents in various ways, but physical thinking is the way to knowledge. Let’s return to the issue of how to organize mental contents to clarify what it means to adopt physical thinking.  As I introduced earlier, one manner of organizing mental contents is to re-illustrate them by using hypothetical entities comprising nature. We somehow connect mental contents with objects and their properties.212 For example, smelly yellowish pointy globular stuff presented in our consciousness may be described in terms of an object with particular properties, onion; spooky damp feeling may be stated as, say, the presence of a ghost; the benevolence of a god may be associated with the usefulness of eyes to animals. Of course, re-illustrations are also mental contents. The mental content that a ghost is hovering around me may be further linked with other mental contents: I have skipped meals; or I am paying my karma. We use hypothetical entities to translate given mental contents and we also have mental contents about them.  Another manner of organizing mental contents that concerned Huxley has to do with how to sequence mental contents, that is, how to arrange phenomena of nature when linking one to                                                           211 Huxley 1887f, 51.  212 Huxley 1880a.  70  another. All phenomena of nature have been, for some reason, grouped by people repeatedly into material and immaterial phenomena with reference to two hypothetical entities, the material and the immaterial. This means to Huxley that we have been making sense of our phenomenal nature in two ways. We interpret some phenomena in terms of the operation of material objects or events, while regarding them as material phenomena; we interpret other phenomena in terms of the operation of immaterial objects or events, while regarding them as immaterial phenomena. This does not suggest, according to Huxley, that we should not mix-match two types of phenomena when arranging them. There is no barrier against it, because we deal with the same thing – contents appeared to our consciousness. The point of arranging phenomena is to get some systematic understanding of what-follows-what in the stream of our consciousness.   When we interpret and arrange phenomena, we can employ either physical thinking or non-physical thinking. We may want to link a given phenomenon to immaterial phenomena by understanding it as being produced by or correlated with the operation of an immaterial object. When, borrowing Huxley’s example, we attempt to understand a phenomenon that water tends to flow in terms of an immaterial object, “aquosity,” we are employing non-physical thinking.213 Or, we may try the other way. We may attempt to link the phenomenon with material phenomena by understanding it as being produced by or being correlated with material objects. When we understand the behavior of water in terms of properties and structure of molecules, we are employing physical thinking.   Which way of thinking, then, should we adopt? Or, should we adopt both? According to Huxley, there is no correct way of thinking. Which to adopt between the two ways of thinking is not an issue that has been pre-determined by an alleged ontological truth which is called                                                           213 Huxley 1868a, 152.  71  “materialism” or “immaterialism.” Moreover, Huxley did not think that debating over which to adopt can be settled by a metaphysical inquiry. Rather, it is the matter of choosing “terminologies” or “one set of symbols,” which can be settled by taking into consideration the fittingness and effectiveness of adopting each in the light of users’ epistemic condition and aim.214 A language and way of thinking are employed by human beings to organize mental contents and further to achieve a certain goal. There is no truer choice, but there could be a more rational choice.  Of course, Huxley preferred and argued for the thoroughgoing adoption of physical thinking, because he was interested in a way of thinking for a guide to knowledge, as opposed to a guide for, say, casual chats. When our concern is knowledge, it is better to use one way of thinking, because we cannot adopt both ways of thinking at the same time. If we deal with water tending to flow from a non-physical perspective, it means that we have given up taking the physical perspective. Thus, Huxley said, if “using one terminology, or one set of symbols, rather than another” turns out to be more conducive to our epistemic goal, “it is our clear duty to use the former.”215  He recommended physical thinking.216 As discussed, Huxley thought that physical thinking likely offers a definite causal and structural chain which makes an account explanatory. He found non-physical thinking incapable of offering a definite causal and structural chain; what spiritual thinkers usually do is attach an immaterial entity to a given (hard-won) physical chain and claim that it is a governor or beginner of a whole process. This kind of work can add no explanatory value, and thus Huxley concluded that physical thinking offers a sufficient                                                           214 Ibid., 164. 215 Ibid. 216 Huxley 1854c; 1868a; 1878h.   72  account.217 A guide to knowledge should promote a way of thinking that can render phenomena intelligible, and this is one reason why Huxley recommended physical thinking or material terminology; “the alternative, or spiritualistic, terminology,” Huxley said, “is utterly barren, and leads to nothing but obscurity and confusion of ideas.”218 We are trying to identify knowledge to be used to construct a new Nature, and, although not all materialistic-mechanistic accounts turn out to be a trustworthy mental content (I have not discussed how Huxley determined a trustworthy mental content; it will be explained in next chapter), we should try to come up with a materialistic-mechanistic mental content so that our candidate for knowledge can have explanatory value.    Thus, Huxley insisted that he did not commit himself to materialism.219 Interpreting phenomena in material terminology appears to adopt the doctrine of materialism, according to which all phenomena are resolved into matter and motion. However, when the doctrine is adopted as part of the package of physical thinking, it is not a true ontological statement about reality. It is rather a statement about what one will do: a physical thinker will view all phenomena in terms of phenomena of matter and motion. Huxley preferred the expression “material terminology” over “materialism” not just because he intended to keep distance from those who had taken the doctrine as an ontological truth; but also because expression “material terminology” reflects his instrumentalistic understanding of language as a tool to capture and fix an idea. To Huxley, for example, scientists’ “matter” is analogous to mathematicians’ “x,” and words used in an ordinary                                                           217 Huxley understood the “law of parcimony” [sic] in terms of explanatory value (1879a, 307). If two accounts do the same explanatory work, and one account has as its component an entity or event that makes no explanatory contribution, the “law of parsimony” guides us to favor the other account.       218 Huxley 1868a, 164. 219 Ibid.; 1894d.   73  life are no less symbolic.220 Huxley did not speak of “terminology” to disguise his ontological commitment to materialism; metaphysical materialism and material terminology do different works, and as we will see in Chapter 6, he thought that it is better not to endorse metaphysical materialism for the sake of science.   Huxley’s distinction between metaphysical materialism and material terminology can be also explained in terms of the difference between Huxley and Descartes in their push for the physical ways of thought. Huxley highlighted Descartes’s practice of physical thinking, but noted that Descartes had expanded his physical thinking to and only to vital phenomena because of his dualism. For Descartes, there is a metaphysical reason to study animals physically; animals are material objects without a soul and thus it would not make sense to adopt immaterial thinking to study them. Huxley could not find any metaphysical necessity that would compel us to study animals physically. Instead, Huxley had an epistemic reason to study animals physically: explanatory value understood in terms of offering a definite causal and structural account. What Huxley had for his guide to knowledge are only the two paths (in addition to Descartes’s maxim) and no other ontological commitment. Thus, he could say that he had methodological commitment to materialism, while denying metaphysical materialism.   Although Huxley thought that his guide to knowledge lacks any metaphysical commitment (including metaphysical materialism), as I briefly introduced, some might think that he would need some eventually. Can we claim to have epistemic value without grounding it on reality? Huxley would have to explain why some mental contents are knowledge and others are not, and how using physical thinking can get us something that deserves the title of knowledge. These issues will be addressed in the following chapter.                                                            220 Huxley 1868a, 165; 1852; 1877b.  74  ▌CHAPTER 4  Huxley’s Epistemology: Memory, Verification, and Physical Thinking    As seen in the previous chapter, Huxley had an epistemological project of offering a guide to knowledge. I have described the project as “Cartesian,” because he shared Descartes’s aim and wanted to mobilize the three elements derived from Descartes: legitimate Idealism, legitimate Materialism, and Descartes’s maxim which I have not yet examined. Huxley thought that any guide to knowledge should embrace legitimate Idealism (we can have access only to mental contents), because it is an irrefragable fact obtained by the method of doubt and it delineates our epistemic condition. Huxley also wanted to endorse legitimate Materialism (a materialistic-mechanistic account is sufficiently explanatory and universally applicable), because physical thinking, unlike non-physical thinking, tends to offer an account that has explanatory value. The discussions of this chapter supply us with another reason: physical thinkers’ account can become a justified account by the way of verification, which he defended as the only proper method of justification. For those two epistemic reasons, physical thinking alone should be a constituent of a guide to knowledge.   Thus, Huxley’s second reason for favoring physical thinking over non-physical thinking hinges on his claim that the method of verification is the only proper method of justification, and this chapter reconstructs how he would have made a case for the claim. Huxley first attempted to establish the method of verification is a method of justification. Because of our epistemic situation, we should single out a particular group of mental contents that we may use as a model of knowledge, and an “anatomy of mind” would help us in this regard. According to Huxley, memory is such a group, and thus we may come up with a process by which we can make sure  75  that a given mental content takes the form of memory. This process is the method of verification, and we can use this process despite our uncertainty about what is outside of our consciousness. Huxley further showed that the method of verification is the only method of justification, by arguing that other methods cannot function properly as a method of justification. Finally, Huxley claimed that the method of verification works well with physical thinking, but not with non-physical thinking.       4.1. How to identify knowledge   Recall the irrefragable fact that the states of consciousness are only things that are given to us. This fact was fundamental to Huxley’s epistemological project in the sense that every epistemologist should start with the fact. Also, recall that the goal of his project was to offer a guide to knowledge. Because he wanted a guide to knowledge, he needed to have a method for identifying knowledge among all possible thoughts or mental contents. However, because we can only rely on our “observation and reflections” about what is happening in our consciousness, Huxley tried not resort to other ways that are incompatible with the irrefragable situation of human beings. A kind of pattern or criterion can be internally seen if we carefully examine thoughts, as he could see a shared form from a messy pile of similar and dissimilar marine invertebrates.221  Thus, Huxley found it necessary to categorize and examine thoughts, mental contents, based on similarities and dissimilarities among them and on nothing else. A philosophical purpose of conducting an analysis of mind is to understand characteristics of a model class of                                                           221 Mario A. di Gregorio, after examining Huxley’s scientific works, notes as follows. “It should be already apparent that Huxley was consistent in method and approach throughout his career. The basis, despite changes of discipline and subject matter over the years, is always the detailed structural examination of specimens, leading to a classification on the basis of the affinities revealed” (1984, 115).    76  mental contents that has been considered to be trustworthy and further to come up with a method of justification, which will base his guide to knowledge. According to Huxley, as we will see shortly, memory is a class of mental contents that we may use as a model of knowledge; a structural characteristic of memory as a class can serve as an indicator of being knowledge. Thus, if we can come up with a process by which we can make sure that a given mental content has the characteristic of memory, we would have a method of justification (this process is the method of verification). Then, Huxley seems to have concluded, we will have obtained an epistemological tool that works within our limits and regardless of our ignorance of something external to our consciousness.  4.1.1. Huxley’s Hume The best place to see Huxley’s analysis of mental contents and his reasoning behind it is Hume where he discussed Hume’s and other modern philosophers’ analysis of mind. There we can see Huxley explaining the purpose and necessity of examining mind: “it is obviously impossible to answer the question, What can we know? unless, in the first place, there is a clear understanding as to what is meant by knowledge,” and this issue “cannot be approached without the examination of the contents of the mind; and the determination of how much of these contents may be called knowledge.”222 Also, Hume was the first place where the term “agnosticism” appeared in print. Huxley situated agnosticism within the tradition of critical modern philosophy which he thought Descartes had started. It seems beyond doubt that the book contains philosophical views and lines of reasoning that lay behind Huxley’s agnosticism.                                                             222 Huxley 1878h, 58-9.   77  Huxley’s Hume has not received sufficient attention from Huxley scholars. Apart from the reviews of his contemporaries, early scholarly biographical works have covered the essay to various extents: P. Chalmers Mitchell, Huston Peterson, William Irvine and Albert Ashforth discussed the essay; James A. Davis reproduced parts without much discussion; Cyril Bibby mentioned it in passing.223 Turning to rather recent scholars’ works, the book has been barely mentioned in connection with his epistemology and agnosticism. For example, in the case of book-size works, Hume is not mentioned in Frank M. Turner 1974, Mario A. di Gregorio 1984, J. Vernon Jensen 1991, Adrian Desmond 1994 and Sherrie L. Lyons 1999. Desmond’s later version of Huxley provides information regarding the publication of Hume with general comments such as that Huxley made Hume “the voice of Victorian Scientific agnosticism.”224 Paradis discusses Huxley’s view on innate ideas presented in Hume.225 Lightman comments with reference to Hume that Huxley tried to solve the difficulties of Humean epistemology by using a Kantian idea of the structure of mind and that Huxley denied the necessity of causal relations.226 Matthew Stanley’s recent book, Huxley’s Church and Maxwell’s Demon, would be a notable exception in the sense that Huxley’s Hume is one of Stanley’s main texts as he cites it many times, yet his focus is not on Huxley’s work on mind and its bearing on epistemology, but on introducing Huxley’s views on law, miracle, will, and the nature of philosophy, which I consider to be grounded on Huxley’s discussions of mind and language.227 In the case of article-size works on Huxley, Huxley’s Hume is sometimes mentioned in passing or briefly introduced in a footnote:                                                           223 Mitchell 1900; Davis 1907; Peterson 1932; Bibby 1959; Irvine 1968; Ashforth 1969. 224 Desmond 1997, 500.  225 Paradis 1978.  226 Lightman 1987.  227 Stanley 2014b.   78  Ellegård 1957, Dockrill 1971, Barton 1983, Block 1986, Smith 1998, Levine 1999, and Lyons 2010. The following works use some of Huxley’s thoughts in Hume in connection to their subject matter: Stanley 1957, Blinderman 1966, Clausen 1976, Gilley and Loades 1981, Knight 1997, Campbell 2001 and Lightman 2002. On the other hand, within Hume scholarship, Huxley has been from time to time mentioned and discussed as an interesting interpreter of Hume’s thoughts.228 Within Darwin scholarship, some of Huxley’s thoughts in Hume are mentioned in regard to Hume’s influence on Darwin (Huntley 1972) and Darwin’s view on consciousness (Smith 1978).            In addition to its title, the fact that many parts of Hume address the topic of mind may prevent us from seeing its relevance to Huxley’s agnosticism. Indeed, the book appears to be all about Hume and it starts with chapters on Hume’s life and political and historical writings. Huxley’s Hume does help us to understand Hume’s thoughts on several issues, but it is not all about Hume. Huxley also introduced other philosophers’ views, but more importantly, Hume is filled with Huxley’s own views and reasoning. Huxley’s contemporary philosopher James McCosh said “Professor Huxley has now in this work on Hume given his own philosophy.”229 For instance, consider Huxley’s discussion of mind, which is one of the main topics of this chapter. His discussion addresses in various degrees the classical modern philosophers’ account of mind. We might expect Huxley to have examined their views separately and thoroughly, but he did not. For example, he brought in Spinoza, but only Spinoza’s analysis of emotion, in connection with an elementary group of mental content. This kind of cherry-picked discussion might be seen as philosophically unprofessional, and scholars of each modern philosopher might                                                           228 For example: Maidment 1939; Kemp Smith 1941; Flew 1959; Noxon 1964; Wright 1983; Pitson 1993; Fosl 1994; Russell 2008; McGrew 2014.   229 McCosh 1884, 43. Also Desmond 1997.   79  find it misleading. However, Huxley’s aim was not to offer a complete exposition of these philosophers’ positions. He presented their distinct views in the way we can arrive at the most plausible and up-to-date account of mind, which is, not surprisingly, the account endorsed by himself. It would be more correct to say that this book illustrates Huxley’s thoughts by using philosophical issues that Hume was tackling. That is why Ashforth describes the book as “a kind of collaboration between Hume and Huxley.”230 I cannot agree more. It appears that Huxley saw a development or even progress in modern philosophers’ understanding of given issues, and his hidden aim was to incorporate and present his view by taking the opportunity of writing a volume on Hume for the series, English Men of Letters.231  In the following, I introduce Huxley’s analysis in detail, because this would put us in a better position to understand his well-known claim for verification as a testing process for putative knowledge. His conception of verification is grounded on his lesser known idea that knowledge should take the form of memory, which in turn owes its plausibility to his categorization and analysis of mental contents and to his selection of memory as a model class of trustworthy mental contents.    4.1.2. Anatomy of mind  Huxley reported that things that appear in consciousness had been termed differently, “thoughts” by Descartes, “ideas” by Locke, and “perceptions” by Hume. These different names, according to Huxley, refer to “certain events, facts, or phenomena (whichever name be preferred) which pass                                                           230Ashforth 1969, 98.  231 John Morley, editor of English Men of Letters, asked Huxley to write a volume on Hume (Huxley 1874c; 1878a). Huxley wrote to Morley that “Hume is frightfully tempting” and that Hume’s Inquiry “touches all the problems which interest us most just now” (1874c, 149; 1878c). See also Desmond 1997, 497-500.  80  over the inward field of view in rapid and, as it may appear on careless inspection, in disorderly succession, like the shifting patterns of a kaleidoscope.”232 This conception of the states of consciousness, Huxley said, is based on our observation, and “without any hypothetical admixture.”233 In this respect, Huxley found Hume’s terminology the most proper.  Huxley not only adopted Hume’s term “perception,” but he also embraced Hume’s general categorization of perceptions. I will explain Huxley’s view on the contents of the mind while highlighting characteristics of Hume’s categorization that Huxley seems to have found attractive. The following two questions can help us to see Huxley’s rationale underlying proposing of his own analysis of mental contents. First, why did Huxley accept some aspects of Hume’s categorization? Second, why and how did he revise others?   Past philosophers with whom Huxley acquainted himself also had offered a categorization of mind, but Huxley adopted Hume’s. Hume used two distinctions. First, perceptions are either impressions or ideas. Second, each class is grouped into simple and complex ones. Accordingly, there are four categories: simple impressions, complex impressions, simple ideas and complex ideas. Huxley found Hume’s categorization appropriate. For one thing, the two sets of distinctions seem to have been made without appealing to something other than states of consciousness, such as mental contents given by God or a distinction like Locke’s primary and secondary properties. For another, Hume had not simply taken our ordinary state of consciousness as a basic unit of mental contents, but he had attempted to identify what Huxley called “primary irresolvable” mental contents, as Huxley introduced Hume’s letter to Francis                                                           232 Huxley 1878h, 73. 233 Ibid.  81  Hutcheson where Hume had contrasted an anatomist’s and a painter’s approaches to the mind.234 These points will be illustrated more later on.   Consider the first distinction: perceptions are divided into impressions and ideas. Following Hume, Huxley took it for granted, based on our experience and reflection, that quite similar images or contents are present in consciousness but somehow differently. Hume had formulated the felt difference as the difference in vividness. The more vivid contents are impressions which we have when we sense or feel, and the less vivid or “faint” ones are ideas which we have when we recollect or think.   Huxley noted that Hume’s criterion, the degree of vividness, had been criticized. Huxley added that there are occasions where the degree of vividness misleads one about whether a given perception is an idea or an impression: for example, a person who is undergoing visual afterimage may take, say, the idea of redness as a new impression of redness because of the vividness of the afterimage.235 Nevertheless, Huxley admitted that it is hard to find a better criterion than Hume’s. For Huxley, the criterion of vividness has the advantage that it does not appeal to anything outside the states of consciousness, which are only things given to us as certain.       When Huxley stated that “the psychologist dissects mental phenomena into elementary states of consciousness, as the anatomist resolves limbs into tissues, and tissues into cells,” he appears to have in mind Hume’s distinction between impressions and ideas.236 Limbs are formed from tissues, and tissues from cells. This is supposed to correspond to an elementary-derivative relationship that Hume had proposed for impressions and ideas. According to Hume, ideas are “copied” from impressions, and this in turn accounts for the greater vividness of more elementary                                                           234 Ibid., 88; 71, n1.   235 Ibid. 236 Ibid., 59.  82  perceptions relative to more derivative ones. Although Huxley was wary of Hume’s illustration of the relationship between impression and idea as “copying” or “copied” and preferred instead “metamorphosis” from impressions to ideas, Hume had been, to Huxley, right in capturing an elementary-derivative relationship from the felt difference in vividness among similar perceptions.237 The important message regarding Hume’s impression-idea distinction is that we can group mental contents into elementary mental contents (“impressions”) and derivative mental contents (“ideas”) by using vividness, without reference to their assumed origin. When grouping mental contents, we need not understand impressions in terms of, say, some interaction with objects that really exist outside.   Hume had used an additional distinction to sort out perceptions: simple and complex perceptions. When a perception can be broken into simpler perceptions, it is considered to be complex. For example, using Huxley’s example, the perception of red is simple whereas the perception of a red circle is complex. This distinction is supposed to capture a single-composite relationship, and Huxley agreed with Hume that perceptions exhibit such a relationship. Huxley added that “[t]he one [anatomist] traces the development of complex organs from simple rudiments; the other [psychologist] follows the building up of complex conceptions out of simpler constituents of thought.”238 Huxley was referring to a compositional relationship expressed in Hume’s distinction between simple and complex perceptions.                                                            237 Ibid., 78, 88, 102. Some Hume scholars understand the copying relation to hold only between a simple impression and a simple idea, but not between a complex impression and a complex idea. Huxley seems either to disagree or to think that this issue does not stand as important. To Huxley, after all, Hume’s so-called “copy principle” is a crude attempt to describe unknown processes generating one group of perceptions (idea) in connection with the other group of perceptions (impression). What is important about Hume’s discussion of “copying” relation or principle, to Huxley, is the notion that an idea is derived or transformed from an impression. Thus Huxley appears to have reasoned that it is no risk to loosely say that a complex idea is a copy of a complex impression. Huxley stated that the idea of red rose is the copy of the impression of red rose, whereas the idea of blue rose is “not an actual copy of any complex impression though all its elements are such copies” (Ibid., 76).  238 Ibid., 59.  83  Now turn to the second question concerning Huxley’s revision of Hume’s categorization. Huxley was willing to adopt Hume’s two distinctions to sort out perceptions, but it is another issue to determine what to count as an impression or idea and what to count as simple or complex. Huxley’s revision involves the extension of each category. Huxley summarized Hume’s categorization of impressions:   According to Hume, these [impressions] are of two kinds: either they are impressions of sensation, or they are impressions of reflection. The former are those afforded by the five senses, together with pleasure and pain. The latter are the passions or the emotions (which Hume employs as equivalent terms). Thus the elementary states of consciousness, the raw materials of knowledge, so to speak, are either sensations or emotions; and whatever we discover in the mind, beyond these elementary states of consciousness, results from the combinations and the metamorphoses which they undergo.239   Huxley moved on to point out that Hume had erred in failing to include a certain type of impression, namely perceptions of relations. They should have been regarded as elementary perceptions (i.e., impressions), not derivative perceptions (i.e., ideas).240 This revision is worth highlighting, given the concern regarding Hume’s account, which Alan Richardson succinctly summarizes as follows: “a succession of impressions is not an impression of succession, so there seems to be nothing for the concept of succession to be the copy of.”241 As we will see in following discussions of this chapter, this inclusion of the impression of relation seems to have allowed Huxley to take steps that would not have been available for Hume.                                                               239 Ibid., 77-8. 240 Huxley also pointed out that Hume should not have counted “impressions of reflection” as impressions. Following Spinoza, Huxley claimed that emotions are complex perceptions containing the idea of pleasure or pain and other ideas (Ibid., 78). The seeming complexity of a given perception does not show that the perception is an idea and thus should be excluded from the list of impression; yet, once we notice that a given perception looks complex not simple, we can proceed to examine whether its parts are derived ideas or elementary impressions.   241 Richardson 2003, 59.   84  Huxley claimed that Hume had characterized the notion of relation in an inconsistent manner. According to Huxley, Hume’s discussion makes it unclear whether a perception of relation results in, or results from, an association between two ideas. In some places and initially, Hume had characterized relations such as resemblance or cause-and-effect in terms of the “associating qualities” of an idea, which “attract” another idea; in other places, relations are complex ideas generated by an associating process.242 Yet Hume had offered another characterization when discussing the relation of equality (a relation of quantity), which Huxley found the most adequate as an account of relation: it is a perception that arises when the mind compares two impressions. This characterization should have suggested to Hume himself that perceptions of relations belong to impression, elementary group of perceptions. However, even when entertaining the notion that relations are not attracting qualities but perceptions that emerge when two impressions are present, Huxley said, Hume had failed to count them as impressions.  Hume had regarded the perceptions of relations as ideas, but Huxley argued, they should be counted as impressions. This is because perceptions of relations also have the characteristics that had led Hume to think that sensations belong to the elementary group of perceptions, impression. A perception of a relation also, Huxley claimed, is describable as a perception of distinctive single quality and is not a derivative of some other perceptions. Thus, by the same reason, Hume should have concluded that relations are impressions, rather than ideas, qualities of an idea, or some kind of principle.  According to Huxley, there are four impressions of relation: perceptions of co-existence, succession, similarity, and dissimilarity. The impression of relation is different from the impression of sensation only in that at least two of the other impressions must be present, and                                                           242 Huxley 1878h, 83.  85  additional mental faculties such as remembering would be required. This does not mean that a perception of relation is derivative, like an idea which is a copied impression. If we stick to Hume’s understanding of impression, nothing prevents us from calling, for example, the feeling of co-existence “impression of co-existence.” Moreover, given Hume’s thesis that all ideas are copied from impressions, if perceptions of relation are ideas, there must exist impressions of which the ideas are copies.243 In a sense, Huxley suggested, we may describe impressions of relation as “impressions of impressions” or “the sensations of an inner sense.”244  Accordingly, Huxley revised Hume’s category of impression.245 Huxley excluded impression of reflections and instead added impressions of relations.246 For Huxley, the contents of mind consist of an elementary group of perceptions, termed “impressions,” and a derivative group of perceptions, termed “ideas.” All the perceptions are either impressions or ultimately resolved to impressions.                                                            243 Ibid.  244 Ibid., 82. 245 Huxley’s revision is as follows (the list as appeared in Huxley 1878h, 85): A. IMPRESSIONS. A. Sensations of a. Smell. b. Taste. c. Hearing. d. Sight e. Touch. f. Resistance (the muscular sense). B. Pleasure and Pain. C. Relations. a. Co-existence. b. Succession. c. Similarity and dissimilarity. B. IDEAS. Copies, or reproductions in memory, of the foregoing.   According to Huxley, Hume had not included resistance as a kind of sensory impression. Huxley did not think that Hume’s omission is a serious mistake because it was not known in Hume’s days.   246 See footnote 240 for why Huxley excluded reflections from an elementary group of perceptions.   86  The categorization of perceptions by itself does not tell us what to single out as trustworthy perceptions that we may use as a model of knowledge. Let me highlight once more the purpose of this sorting out work. Huxley was probably interested in learning about how mind exactly works and how mind comes to have a perception, but this kind of proto-scientific or metaphysical inquiry can be set aside. For an epistemological inquiry, an “anatomy” of mind is necessary. However, Huxley categorized perceptions not because he wanted to conclude that impression as an elementary unit should be correct information of a world external to consciousness and thus impressions have epistemic value. The categorization of perceptions and the two criteria used for the categorization would be instructive when we examine distinctive characteristics of a certain type of perception that has been more or less trusted. If we determine its distinctive characteristics by comparing and contrasting it with other types, we can examine epistemic value of such characteristics and further we may be able to use them to formulate a methodical process for identifying trustworthy perceptions, which would work independently from our ordinary trust or distrust in particular perceptions.  In this regard, according to Huxley, memory is a type of perception that deserves our attention.    4.1.3. Memory  Memory is one group of complex ideas. Hume had grouped all complex ideas into two, memory and imagination. Huxley introduced Hume’s two criteria for the division, and dismissed one of them, the degree of vividness. This criterion works well enough for distinguishing impressions and ideas, but does not work well in this case, because, Huxley reasoned, many memories are not vivid and some imaginations are not faint. Huxley however agreed to the other criterion. Some  87  complex ideas retain the arrangement of a complex impression and others do not; the former group is what we call “memories” and the latter group is what we call “imaginations.” This difference, Huxley explained, amounts to the presence or absence of the idea that a corresponding complex impression existed, which captures our feeling of happening or occurrence when we were perceiving a complex impression. Hume had viewed the group that lacks the idea as a single type, and labeled all its members “imaginations.”  However, Huxley pointed out, Hume should have divided the latter group into two kinds, imagination and expectation. Expectations, Huxley wrote, “differ from simple imaginations in being associated with the idea of the existence of corresponding impressions, in the future, just as memories contain the idea of the existence of the corresponding impressions in the past.”247 If, as Hume appears to have suggested, we have a reason to distinguish some complex ideas by utilizing the presence of the idea that its corresponding complex impression existed in past, then we are also given a reason to utilize the presence of the idea that its corresponding complex impression will exist in future. Thus, to Huxley, a complex idea is one of three kinds: memory, expectation, and imagination. One might think that memories are not perceptions about generalized objects and events, and thus memory is not a class of perceptions worthy of attention for our task of determining a model of knowledge. However, Huxley disagreed. If we look into the stock of perceptions, Huxley said, we could see that we indeed have both specific and generic memories. A memory about a particular object or event is “specific” because its complex idea is derived from a single                                                           247 Huxley 1878h, 110.  88  complex impression about a particular object or event.248 There is also another type of memory: a complex idea that is not derived from a single complex impression. When several, more or less different, complex impressions are present in consciousness, a complex idea is consequently formed (by unknown processes). This type of complex idea is a generic idea, not a specific idea derived from one of complex impressions. Huxley illustrated how a generic idea might be generated to explain the conception of generic memory.     This mental operation may be rendered comprehensible by considering what takes place in the formation of compound photographs – when the images of the faces of six sitters, for example, are each received on the same photographic plate, for a sixth of the time requisite to take one portrait. The final result is that all those points in which the six faces agree are brought out strongly, while all those in which they differ are left vague; and thus what may be termed a generic portrait of the six, in contradistinction to a specific portrait of any one, is produced.249          What Huxley meant by “compound photograph” is corresponding to Francis Galton’s “composite portrait,” a kind of blended photograph produced by superimposing several photographs of individuals, which Galton made to see an “average” face of a certain group of people, such as criminals.250 As images are superimposed, Huxley conjectured, we form an idea representing a “general image” (which does not take a linguistic form like a general proposition), different from                                                           248 Ibid., 112. 249 Ibid., 111.  250 Galton 1878, 97. Huxley did not mention Francis Galton in Hume which was written and printed during the latter half of 1878, but Huxley was probably aware of Galton’s work on composite portraits, because Galton had introduced his work in his presidential address in the previous year and also in Nature in early 1878 (Galton 1878; Huxley 1878e; 1878f; 1878g). Yet, in a later essay, Galton mentioned Huxley and wrote the following: “Professor Huxley, from whom I have borrowed the apt phrase [“generic image”], has expressed himself to a similar effect in his recent Life of Hume, …” and “he [Huxley] has, quite independently of myself, adopted a view which I also entertained, and had hinted at in my first description of composite portraiture, though there was not occasion at that time to write more explicitly about it” (1879, 164).  89  all of specific “superimposed” perceptions. Huxley considered many of what people have commonly labeled as “abstract or general ideas” to be generic ideas.251 Huxley provided several examples to persuade readers. One example is worth noting.   An anatomist who occupies himself intently with the examination of several specimens of some new kind of animal, in course of time acquires so vivid a conception of its form and structure, that the idea may take visible shape and become a sort of waking dream. But the figure which thus presents itself is generic, not specific. It is no copy of any one specimen, but, more or less, a mean of the series.252  Huxley, versed anatomist, regarded a biological type as a generic idea. di Gregorio, whose research focus is Huxley’s zoological works, notes that Huxley aimed for a conception of type that has “a purely empirical meaning,” but as for the product, di Gregorio concedes, “[o]ne might venture to say that here Huxley is either a bad Platonist or a bad empiricist – or both.”253   Huxley’s conception of generic idea may look less strange if his claim is taken into consideration that the formation of a generic idea does not require using a language or conscious efforts, especially in the case of a generic idea of sensible objects as opposed to abstract objects. He argued that babies and some animals undergo the same process. This is hard to deny, he said, given behavioral and anatomical similarities. They are equipped with organs that can perform basically the same function, and they exhibit behaviors that appear to be acted upon by “abstract” or “general” ideas. His point is that such memories are generic ideas that are not expressed in a statement of generalization. A baby cannot formulate the sentence that a sugarplum is sweet, but                                                           251 Huxley 1878h, 112.  252 Ibid., 113.  253 di Gregorio 1984, 33-4.  90  the sentence is “merely the verbal expression.”254 When she is trying to grab a sugarplum on the table, Huxley said, she is acting upon a generic memory.255  Once a generic memory is formed, its corresponding expectation follows. The object of a memory, the part of a complex idea, is turned to the object of an expectation. A partial complex idea, say, that a candy being coexistent with sweetness, remains but it is now combined with the idea of the future existence of the corresponding impression instead of the past existence of that impression. If the impression of sweetness follows after eating a candy and this happens repeatedly, according to Huxley, we cannot think of candy without thinking of sweetness. In this way, we form a corresponding expectation. A point to note here is that for Huxley, the memory that a candy was sweet and the expectation that a candy will be sweet share the same component that a candy is co-existent with sweetness. To see whether a given memory is a correct memory, we sometimes recall what happened in the past, but if we are dealing with a generic memory, we can instead check on its corresponding expectation.     4.1.4. Memory as a model of knowledge According to Huxley, we generally trust, both consciously and unconsciously, our memories in daily lives. As Huxley understood it, believing is basically remembering. When we distinguish “true” beliefs from “false” beliefs, we are separating memories from imaginations. We value and treat expectation and imagination differently, and this also reflects our trust in memories.                                                            254 Huxley 1878h, 114. 255 Elsewhere Huxley also wrote: “No child has recourse to imaginary personifications in order to account for the ordinary properties of objects which are not alive, or do not represent living things. It does not imagine that the taste of sugar is brought about by a god of sweetness, or that a spirit of jumping causes a ball to bound” (1869c, 319).  91   Given that memory is our best candidate as a group of perceptions that we trust among all groups of perceptions, we may move on to argue that all memories that people claim to have are therefore knowledge. This is not Huxley’s move, however. Our ordinary memories or beliefs about a phenomenal nature do enjoy our trust in general, but they tend to be imprecise and ambiguous (I will return to similarities and differences between ordinary reasoning and scientific reasoning in Chapter 6). As explained in the previous chapter, Huxley wanted to construct a new Nature out of such a familiar and everyday phenomenal nature; this was the reason why he found it necessary to come up with a method for identifying knowledge among all possible perceptions.    In other words, in Huxley’s epistemology, memory as such is a model of trustworthy perception. For any perception to be considered knowledge, it is shown that a given perception takes the form of memory. Accordingly, Huxley highlighted the structural difference of memory from expectation and imagination: only memory has the idea that a complex impression, corresponding to what is remembered, existed in the past. This essential component would illuminate why we tend to trust memories and further form and trust expectations based on memories. This information, Huxley thought, should guide us when we attempt to formulate a justificatory method for identifying knowledge. What is special about the essential component of memory? We have an idea of the past existence of corresponding impressions when we perceived a set of impressions of sense, relation, pain and/or pleasure in the past (for now, let’s ignore the case of a false memory). To be sure, all the parts of an imagination are resolved into impressions. For any imagination, say, consisting of three ideas, A, B and C, there are supposed to be “originals,” corresponding impressions, a, b, and c. Thus, the fact that all contents of a given perception can be ultimately traced back to impressions does not differentiate memory from imagination. The imagination has  92  been formed without the experience of perceiving a, b, and c altogether. That is to say, we experience some arrangements of impressions but we do not experience others. For example, using Huxley’s example, we come to get the information that we are having the impressions of “sugarplum,” “sweetness,” and “co-existence” together, which would be later featured in a memory as the part that the complex impression appeared to us in the past.  Huxley appears to have found the essential component of memory epistemically valuable in two respects. On one hand, the component seems to have rendered memories credible to us, by supplying us with the feeling of happening. If a perception is accompanied with such feeling, we may have a reason to be certain about the perception. The essential component may be well used in our justificatory method for identifying trustworthy perceptions.     On the other hand, this essential component can well function for our justificatory method for identifying trustworthy perceptions. Huxley’s discussions of memory so far and his examination of other methods which will be introduced in the following sections suggest that he appears to have found two kinds of merit. First, the essential component of memory supplies us with information on a particular arrangement of impressions, not simply information on each impression. That is, the idea of the past existence of a certain arrangement, in a sense, picks out to us which arrangement of ideas to trust. For example, we have the memory that candies were sweet, not that candies were bitter. Just as the result of a so-called “critical experiment” tells us which hypothesis to prefer, the essential component of memory tells us which complex idea to prefer. One merit of the essential component is that it can function as a differentiator.  The essential component of memory has another merit. By virtue of the component, we can check a dubious memory, by try having another memory of the same kind. If one claims that an unsupported object soars high based on a memory, we can examine its corresponding  93  expectation by checking whether the complex impression of an unsupported object soaring high will exist. If we cannot have the impression, the person’s perception would turn out to be an imagination, not a memory. This process can be done because we can experience again a complex impression of the same generic memory. When we are not certain and want to determine if an unsupported object falls to the ground, we can put ourselves in the mental state of perceiving an unsupported object and wait and see what appears in our consciousness; in short, we can let go of an object again. The other merit of the essential component of memory is that it allows us to test a given complex idea. Because of the epistemic value of the essential component of memory, memory deserves special attention as a model of knowledge. Huxley’s suggestion is that we can use experiencing an arrangement of impressions when we want to check whether a given arrangement of ideas is trustworthy or not. Thus, to determine whether a given complex idea is knowledge or not, we need to figure out how we can have a corresponding or relevant complex impression.  Some might doubt: we want to have a method for identifying knowledge, because we want to determine whether or not, say, a sugarplum is sweet, not because we want to determine whether or not we have ever perceived “sugarplum,” “sweetness,” and “co-existence” together. However, according to Huxley, having a complex impression is a fitting criterion given our epistemic situation and aim. We have acquaintance with nothing but what appears in our consciousness, and what we need to do is to distinguish trustworthy complex ideas from non-trustworthy ones. Here, a complex idea is, in the end, a perception. A trustworthy complex idea is just a particular perception; its arrangement has appeared to us as an arrangement of impressions. We cannot and need not know, for example, if an idea that an unsupported object falls to the ground really reflects reality. Instead, it should suffice for us if we can be certain about a  94  particular sequence of mental states: for example, the mental content of an unsupported object is followed in our consciousness by the mental contents of succession and the object falling to the ground. This is the context in which epistemic value should be found and appreciated. According to Huxley, in addition to “our trust in representation of consciousness,” nothing more is needed to be pursued as our epistemic aim and nothing more can be achieved due to our condition.256  The notion that experiencing a complex impression is epistemically valuable can illuminate why Huxley considered observations and experiments to be essential to science. The point of that kind of activity is to put ourselves in a certain situation or condition so that we can learn which particular arrangement of impressions appears and whether a series of impressions under discussion indeed appears. To Huxley, a hypothesis is a supposition about what we would observe.257 It is, thus, a necessary task to identify what it means or involves to observe a given (rather theoretical) hypothesis and to formulate a plan on how to make ourselves observe it.258 In other words, these are efforts to turn an expectation into a memory.         Huxley used his discussions of mind to explain the notion of verification, and we can see that he attempted to understand the method of justification in terms of ordinary mental processes.  The process of strengthening generic memories of succession, and, at the same time, intensifying expectations of succession, is what is commonly called verification. The impression B has frequently been observed to follow the impression A. The association thus produced is represented as the memory, A → B. When the impression A appears again, the idea of B follows, associated with that of the immediate appearance of the impression B. If the impression B does appear, the expectation is said to be verified; while the memory A → B is strengthened, and                                                           256 Huxley 1870c, 178. 257 Huxley 1880a. Huxley’s conception of scientific method will be more explained in Chapter 6.  258 It is well known that Huxley had disputes with Darwin over experimental proof of Darwin’s theory of natural selection. I will return to this issue in Chapter 5.     95  gives rise in turn to a stronger expectation. And repeated verification may render that expectation so strong that its non-verification is inconceivable.259  As explained here, if the impression B, say, sweetness, follows, it is said that the expectation is verified, not just met. This highlights that verification works on a link between mental states, as opposed to a relation between expressed contents. To check whether a candy is sweet or not is to check whether we eat a candy with or without tasting sweetness. Repeated verification means that a given arrangement of mental contents has appeared in consciousness repeatedly, but it also has a consequence of making the arrangement inseparable by making the arrangement unbreakable in our consciousness. When we say that the concept of candy is not separable from the concept of sweetness, it means to Huxley that we cannot think of a candy without thinking of sweetness. According to Huxley, (some of) what people call “laws of nature” are generic memories that have been always verified and thus are expected to be always verified, although they tend to consider a law of nature to denote “a thing” that is “endowed with certain powers,” not a careful statement of an arrangement of ideas.260    To summarize, a memory is usually accompanied with an idea of occurrence, picking out an arrangement of ideas, which can be tested and strengthened. These characteristics, I believe, constitute the reason why Huxley thought that memory is a type of perception that deserves epistemological attention and that memories, as such, have default credibility.261 Experiencing an arrangement of impressions is epistemically valuable to us, not because it supplies us with sundry “building blocks”; rather it supplies us with building blocks as a set or series, and strengthens or                                                           259 Huxley 1878h, 120, his italics. 260 Huxley 1887a, 74; 1870a; 1878h; 1880a. As for another example, Huxley wrote: “Calling our often verified experience a ‘law of nature’ adds nothing to its value, nor in the slightest degree increases any probability that it will be verified again, which may arise out of the fact of its frequent verification” (1878h, 155, my emphasis). Huxley’s conception of law of nature will be discussed in Chapter 6.  261 Huxley 1870a.  96  weakens a set or series of mental contents. What we need to do with the lesson is not to conclude that any claimed memory deserves the title of knowledge, but to come up with a procedure by which we can make sure that a perception at issue is memory – the method of verification.   4.2. Huxley’s defense of the method of verification   Huxley’s suggestion for a method for identifying knowledge among all possible thoughts is to test a given derivative thought (idea) against a series of elementary thoughts (impressions) by trying to have an experience of a series of elementary thoughts, that is, by conducting the process of verification. According to Huxley, the method of verification is the only method of justification that is available to us. How did Huxley show that?  Huxley’s strategy can be described as reduction and elimination. On one hand, Huxley discussed what he regarded as philosophers’ traditional method. Huxley argued that their search for necessary truths amounts to finding out repeatedly verified propositions. On the other hand, Huxley discussed religious thinkers’ notion that faith yields knowledge. According to Huxley, their appeal to faith cannot be considered to be a legitimate justificatory process because it is arbitrary and often produces contradictory conclusions. These points will be discussed in detail.     4.2.1. Metaphysicians’ appeal to necessity   Huxley’s discussion of necessary propositions, which is not just an exposition of Hume’s view, deserves attention in two respects. First, the fact that there were propositions that had been called “necessary propositions” showed to Huxley that some contents of mind indeed had been considered to have the status of knowledge. This was assumed when he was examining the  97  contents of mind. Traditional metaphysicians, non-physical thinkers and Huxley shared the assumption that some mental contents had been accepted as knowledge.  Second, the traditional search for necessary propositions utilizes the epistemic resources of Huxley’s way of identifying knowledge, but in a cruder way. According to Huxley, to be certain of p means to be certain of the link between ideas expressed in p, and the more strongly the link has been verified, the more certain we are entitled to be. He acknowledged that some people are suspicious about the claim that repeated experiences can be used to find what is certain, but argued that the method of appealing to the notion of necessity is in fact an attempt to find repeatedly verified ideas. In this way, he showed why some “necessary propositions” had been considered knowledge, and further argued that the method of verification is better.  Huxley was well aware that propositions judged to be true by experience had been considered to be inferior to so-called “necessary truths,” especially in the realm of philosophy. 262 Huxley regarded this as a poorly established preconception, and thus it was an essential task for him to undermine it. He seized an opportunity when he came to discuss Hume’s distinction between “relations of ideas” and “matters of fact.” Huxley introduced the following passages from Hume’s Inquiry Concerning Human Understanding:  All the objects of human reason and inquiry may naturally be divided into two kinds, to wit, relations of ideas and matters of fact. Of the first kind are the sciences of geometry, algebra, and arithmetic, and, in short, every affirmation which is either intuitively or demonstratively certain. That the square of the hypotenuse is equal to the square of the two sides, is a proposition which expresses a relation between these two figures. That three times five is equal to the half of thirty, expresses a relation between these numbers. Propositions of this kind are discoverable by the mere operation of thought without dependence on whatever is anywhere existent in the                                                           262 Huxley 1878h.   98  universe. Though there never were a circle or a triangle in nature, the truths demonstrated by Euclid would for ever retain their certainty and evidence.  Matters of fact, which are the second objects of human reason, are not ascertained in the same manner, nor is an evidence of their truth, however great, of a like nature with the foregoing. The contrary of every matter of fact is still possible, because it can never imply a contradiction, and is conceived by the mind with the same facility and distinctness, as if ever so conformable to reality. That the sun will not rise tomorrow, is no less intelligible a proposition, and implies no more contradiction, than the affirmation, that it will rise. We should in vain, therefore, attempt to demonstrate its falsehood. Were it demonstratively false, it would imply a contradiction, and could never be distinctly conceived by the mind – (IV. pp. 32, 33).263    Hume had divided “all the objects of human reason and inquiry” into two groups, “relations of ideas” and “matters of fact,” because they are not “ascertained in the same manner.” What concerned Huxley was not the distinction per se, but an intention behind the distinction. As Huxley understood it, the distinction had been made to claim that the certainty given by checking a relation of ideas is different from, and further superior to, the certainty given by checking “whatever is anywhere existent in the universe.” What Huxley wanted to reject is an epistemological implication that the distinction bears on experience: “the assertion that the evidence of matter of fact is not so strong as that of relations of ideas.”264  Let’s first clarify Hume’s line of reasoning. The difference in the degree of certainty that we are entitled to have stems from the difference in manner in which we ascertain the truth value of a proposition. The truth value of propositions of “relations of ideas” can be shown “demonstratively,” whereas that of propositions of “matters of fact” cannot. What is the condition                                                           263 Hume 1748, as quoted in Huxley 1878h, 137-8 (Hume’s italics). Huxley’s citations of Hume’s works are based on the four volume edition published by Black and Tait in 1826. The citation comes from the beginning of Section IV. Sceptical Doubts Concerning the Operation of the Understanding.  264 Huxley 1878h, 140.  99  for the truth of a proposition to be shown demonstratively? It is when the negation of a given proposition implies a contradiction. On the other hand, if the negation of the proposition does not imply a contradiction, according to Hume, we cannot affirm the proposition demonstratively, and this means that we need the aid of observation to affirm it. At the end of the day, we would affirm or deny any given proposition, but sometimes we need to consult experience and sometimes we do not. The difference in the manner, it has been said, has something to do with, not us, but a proposition – whether its negation involves a contradiction.    A contradiction is a proposition that is not “intelligible” or that cannot be “distinctively conceived.” Huxley examined what it means to say that a proposition is distinctively conceivable. It may mean that the link expressed in a proposition is indissoluble in our consciousness. If our assurance ultimately comes from the indissolubility of a link between ideas in our consciousness, there is no reason to think that we should be entitled to have “greater” assurance regarding the propositions of “relations of ideas” than we should regarding the propositions of “matters of fact.” The factor that determines the degree of certainty would be how strong a link expressed in a proposition is, not to which group a proposition belongs. Huxley illustrated in what sense the propositions of “relations of ideas” are on par with the propositions of “matters of fact.”   To the assertion that the evidence of matter of fact is not so strong as that of relations of ideas, it may be justly replied, that a great number of matters of fact are nothing but relations of ideas. If I say that red is unlike blue, I make an assertion concerning a relation of ideas; but it is also matter of fact, and the contrary proposition is inconceivable. If I remember something that happened five minutes ago, that is matter of fact; and, at the same time, it expresses a relation between the event remembered and the present time. It is wholly inconceivable to me that the event did not happen, so that my assurance respecting it is as strong as that which I have respecting any other necessary truth.”265                                                            265 Ibid., 140-1.  100  Huxley’s reply may look strange without taking into consideration his account of perceptions. As discussed, a proposition is a verbal expression for a perception, and we cannot go beyond our perceptions. A proposition of “matter of fact,” in his terms, reports a complex idea, and a complex idea includes an idea of relation which is derived from an impression of relation. Also, it appears in our consciousness as a mental link of ideas. When we affirm a proposition of “matter of fact,” our certainty is about a mental link; when verified repeatedly, the link becomes indissoluble in our mind. Huxley made the same point regarding the propositions of “relations of ideas.” He reminded readers of Hume’s position: perceptions are divided into ideas and impressions, and ideas are ultimately derived from impressions. Thus, if we follow the position strictly, Huxley asserted, we should conclude that perceptions referred to in the propositions of “relations of ideas” can be traced back to our having some impressions. For example, “if there were no impressions of straight lines and triangles there could be no ideas of straight lines and triangles.”266 Here Huxley’s point of bringing up the generation of ideas is not to state that experience of impressions is needed to learn the propositions of “relations of ideas.” Rather, his point is that propositions of “relations of ideas” involve mental links in our consciousness just like propositions of “matters of fact,” and experience of impressions has equally played a role in forming a mental link.  Given that both propositions of “relations of ideas” and propositions of “matters of fact” involve mental links, Huxley would argue, it is a red herring to point to the different manners in which their truth value is ascertained, based on which the distinction between “relations of ideas” and “matters of fact” is drawn. Note that Huxley speaks of a proposition concerning the past                                                           266 Ibid., 139.  101  event in the quote above. A proposition concerning the past event (e.g., the sun rose yesterday) is as certain as a proposition used to demonstrate that “three times five is equal to the half of thirty.” A proposition that has not been experienced (e.g., the sun will rise tomorrow) is as uncertain as, say, a mathematical proposition that has not been demonstrated. Our experience turns an expectation into a memory, and our “mere operation of thought” turns an unproved mathematical proposition into a proved one. Once experienced or demonstrated, we can be equally certain about them. The point that there is no contradiction between propositions concerning our past experience of the sun and the proposition that the sun will not rise tomorrow is irrelevant to our entitlement of being certain.       To summarize, propositions of “relations of ideas” and propositions of “matters of fact” are on equal footing when it comes to certainty that we are entitled to have; for, regardless of their membership, our certainty targets on ideas linked in consciousness and comes from an indissoluble link formed by verification. Hume’s expression, “whatever is anywhere existent in the universe,” Huxley noted, should be understood, given his own doctrine, to refer to “the sum of our actual and possible impressions”267 To Huxley, as far as certainty is concerned, we have no grounds to think that propositions concerning relations of ideas enjoy higher certainty by virtue of their belonging to a particular type, than do propositions concerning matters of fact.  Some might still argue that Hume’s distinction concerns the mode of truth. That is, (in)conceivability, which is supposed to differentiate the nature of certainty, should be understood in terms of the mode of truth, not (in)dissolubility in consciousness. As Huxley also noted, Hume’s distinction concerns the concept of necessity. Necessary propositions and non-necessary propositions are mutually exclusive, and this suggests that the necessity of proposition can be                                                           267 Ibid.  102  used to differentiate the nature of certainty. As Huxley further admitted, our belief about the mode of truth is one thing and our belief about truth is another. Yet, while acknowledging that necessity is meant to be understood in terms of inconceivability, Huxley argued that there is no other way to understand (in)conceivability than in terms of (in)dissolubility in consciousness.   According to Huxley, Hume had illustrated necessity in terms of (in)conceivability but had failed to show how (in)conceivability can be understood differently from (in)dissolubility in consciousness. Huxley critically examined Hume’s claim that it is not a necessary proposition that “whatever event has a beginning must have a cause.”268 This proposition had been regarded as a necessary truth, but Hume had argued that it is not, because its negation is conceivable. Huxley highlighted Hume’s rationale for thinking that its negation is conceivable: “all distinct ideas are separable from each other.”269 According to Huxley, Hume was supposed to show, not presuppose, that “all distinct ideas are separable in thought.”270 Hume’s “circular” argument does not show how necessity can be understood other than in terms of an indissoluble mental link; introducing the notion of (in)conceivability that is presupposed to differ from the notion of (in)dissolubility did not impress Huxley.    Moreover, the notion of (in)conceivability, as Huxley understood it, primarily concerns our ability, not a property of a proposition. Thus, we should not focus on the “distinctiveness” of given ideas, but instead directly on the dissolubility of a given link of the ideas. According to Huxley, despite the “circular” argument, Hume in the end intended to endorse conceivability in                                                           268 Ibid., 142. 269 Hume as quoted in Huxley 1878h, 143. 270 Ibid., 144, my emphasis.   103  terms of dissolubility, because he had made the distinction between “conjoined” ideas and “connected” ideas, which captures the degree of the separableness of a mental link.271  Huxley’s main complaint about the distinction between “relations of ideas” and “matters of fact” was its implication for certainty rendered by experience. The claim that the membership of a given proposition determines our entitlement of certainty, he traced, ultimately hinges on the notion of (in)conceivability. His point is that we cannot but understand (in)conceivability in terms of (in)dissolubility in consciousness, which depends on how strongly experience has made a mental link of ideas inseparable or separable.  Huxley pressed his point further to claim that the traditional search for necessity has nothing special or superior as an epistemological approach. Throughout his life, Huxley was hostile toward those whom he called “pure metaphysicians.”272 Pure metaphysicians “attempt to base the theory of knowing upon supposed necessary and universal truths.”273 Their approach has been to search for necessary propositions on the basis of the notion that only necessity shows whether a given proposition is certain; once a necessary proposition is somehow identified, the next step is to make logical inferences by which certainty can be safely passed on. According to Huxley, they thought that they have proved a derived proposition by “fine-drawn deductions from axiomatic assumptions” which they took as necessary propositions.274 Speculation and deduction are their main work to find knowledge, and what Huxley found especially annoying was that they “assert that scientific observation is impossible unless such truths are already known or                                                           271 Ibid., 144, 146. 272 For example, Huxley 1871b; 1878h, 62, 87; 1891c; 1892a. 273 Huxley 1878h, 62. 274 Huxley 1854a; 1891c, viii.  104  implied.”275 Huxley called their approach “the high a priori road of mere philosophical speculation,” or “a priori method.”276  To debunk the search for necessity, Huxley first clarified what so-called “necessary truths” turn out to be.     Either they depend on the convention which underlies the possibility of intelligible speech, that terms shall always have the same meaning; or they are propositions the negation of which implies the dissolution of some association in memory or expectation, which is in fact indissoluble; or the denial of some fact of immediate consciousness.277  Propositions that we think we cannot deny have been considered to be “necessary propositions,” and they can be grouped into different types in accordance with a rationale why we cannot deny them. According to Huxley, propositions like “A is A” belong to the first type; we cannot deny them because the denial destroys intelligible speech.278 As an example of the second type, Huxley offered “two straight lines cannot inclose a space,” and claimed that to say that the proposition is necessary means to say that “we have no memory, and can form no expectation of their so doing.”279 The revised conclusion of Descartes’s cogito argument is of the last type; its denial “involves the denial of consciousness.”280   The second type is Huxley’s main concern, because “pure metaphysicians” have primarily relied on this type. Even Hume can be seen to have failed to resist it as seen from him approving the distinction between “relations of ideas” and “matters of fact.” According to Huxley, pure                                                           275 Huxley 1854a; 1878h, 62. 276 Huxley 1871b, 133; 1890b; 1891c, viii; 1894e; 1894f. 277 Huxley 1878h, 140. 278 Ibid. 279 Ibid. 280 Ibid.  105  metaphysicians’ approach was to find necessary truths, because they used the necessity of a given proposition as a criterion for checking whether our feeling of certainty toward the proposition is legitimate. Necessity and thus certainty are understood in terms of something we cannot conceive. Although some might claim that they are dealing with metaphysical possibility and impossibility, to Huxley, their reasoning has only concerned whether the negation of a proposition is conceivable or not. That is, they have invoked nothing but (in)conceivability in order to account for, or to single out, metaphysical (im)possibility which is supposed to inform us about whether we are entitled to be certain. Huxley summarized that “to say that an idea is necessary is simply to affirm that we cannot conceive the contrary.”281   As seen, Huxley understood inconceivability as indissolubility of a mental link. Accordingly, he claimed, those who search for necessity to find something that we are entitled to be certain of, unwittingly, show that certainty that they claim to have found is grounded on repeated experiences. Huxley could not see any legitimate additional signification by calling some propositions “necessary propositions,” instead of “repeatedly verified propositions.” If those who search for necessity claim that we are entitled to be certain that “the square of the hypotenuse is equal to the square of the two sides,” Huxley thought, we should be equally entitled to be certain that the sun rises. Thus, he concluded that the question of whether a given proposition is necessary is “really of very little importance”; what matters is how well and often it has been verified and how much we are willing to verify it.282    Huxley’s criticism went further. It turns out that the pure metaphysicians’ search for necessity is not superior as insisted. What is worse, their proposal of taking necessity as truth can                                                           281 Ibid., 144.  282 Ibid.  106  be misleading. This is because the search for necessity exclusively focuses on the bridge notion – conceivability – on the assumption that it serves as a proof of truth. Huxley found this myopic focus problematic, because in some situations, running an inconceivability test can lead us to a false proposition. Huxley introduced a case in which a false proposition is necessary in the sense of its opposite being inconceivable and thus can be claimed to be true instead.     In the well-known experiment of touching a single round object, such as a marble, with crossed fingers, it is utterly impossible to conceive that we have not two round objects under them; and, though light is undoubtedly a mere sensation arising in the brain, it is utterly impossible to conceive that it is not outside the retina. In the same way, he who touches anything with a rod, not only is irresistibly led to believe that the sensation of contact is at the end of the rod, but is utterly incapable of conceiving that this sensation is really in his head. Yet that which is inconceivable is manifestly true in all these cases. The beliefs and the unbeliefs are alike necessary, and alike erroneous.283   In other words, showing the necessity of a given proposition in the sense that its opposite is inconceivable does not guarantee its truth. p being inconceivable (~p being necessary) is thought to show that ~p is certain and thus true. If we follow this reasoning, we end up saying that the belief that “we have two round objects under fingers” is true because its contrary is not conceivable. But it is not true. Focusing only on inconceivability sometimes misleads us regarding the truth value of a proposition; it has the same problem of the claim that we should take any claimed memory as true. Testing necessity in terms of inconceivability is not always                                                           283 Ibid., 144-5.  107  reliable to Huxley.284 He concluded, “the fact that we cannot conceive the contrary of any belief may be a presumption, but is certainly no proof, of its truth.”285   To have a better method of justification, we should instead put our attention directly on the existence of a set of impressions at issue and conduct the process of verification. This is why we tend to consider propositions of which negation is inconceivable to be certain in the first place. We should aim to find invariable links between mental states, but this does not mean that we can point to indissoluble links and readily accept them as knowledge. An idea conveying an indissoluble link has been regarded as certain, but as seen in the previous section, Huxley wanted to use that to come up with an epistemological tool of finding something that deserves certainty. Thus, we need to clarify further what renders a mental link indissoluble. This, according to Huxley, has been neglected by those who search for necessity; what they have been doing is finding actual indissolubility as if it is a sufficient condition for a link to be invariable. Rather, Huxley suggested the method of verification: we should aim for invariable links and for this, we need to go through testing by utilizing a refined and independent procedure modeled on a process leading to an indissoluble link of mental states in consciousness.                                                              284 Some might find Huxley’s account of inconceivability too idiosyncratic or liberal, even after we accept his understanding of inconceivability in terms of indissolubility of a mental link. However, according to Huxley, our ability to conceive is constrained by our physical constituents, as our locomotion or respiration is (1878h). In other words, the indissolubility of a mental link depends not only on experiences of repeated verification but also on our physical conditions. To be clear, issues regarding the tenability of his account of inconceivability are irrelevant to my concern here. Huxley’s point is that, no matter how one likes to define or reduce (in)conceivability (by invoking, for example, “metaphysical impossibility”), one cannot but rely on feelings about inseparableness or irresistibleness to single out what is conceivable and what is not. As long as Huxley’s targets, traditional philosophers, are in the same situation, his point can be made. However, to defend Huxley, there seems no fair reason to exclude his account of inconceivability at the outset. As an unfair reason, I have in mind some meta-constraint on an account of inconceivability like the following: an account of inconceivability should be neither too weak nor too liberal because it should be able to accommodate the philosophical practice that philosophers have used the notion of (in)conceivability as a sort of grounds when they decide to let a given thesis occupy or not occupy a philosophical discourse space.  285 Ibid., 144.  108  4.2.2. Religious thinkers’ appeal to faith  Huxley criticized his contemporary religious thinkers’ claims in many places, and his criticisms have been well discussed.286 My focus is on Huxley’s discussion of the inadequacy of using faith as a method for identifying knowledge. He recognized that many religious thinkers assumed that having faith can serve as evidence for the truth of the content of faith. However, according to Huxley, having faith can make no contribution to our task of distinguishing trustworthy perceptions from non-trustworthy perceptions. Huxley made two points: first, the appeal to faith is based on arbitrary grounds; second, the appeal to faith leads us to incompatible conclusions. Because of these two defects, faith is not qualified as a justificatory factor.   As Huxley understood religious thinkers, they had a fundamentally different view on what kind of knowledge we need to have to lead our lives.287 They seem to have disagreed with proponents of the method of verification who generally hold the position that our life in the natural world only matters and thus knowledge about that world suffices. Huxley labeled this position as “naturalism”: we need knowledge only about a world that we can experience – “tangible, commonplace, orderly world of Nature.”288 If naturalism is granted, the method of verification might be accepted as the only way of identifying trustworthy perceptions, but many religious thinkers reject the position. Instead, Huxley stated, they claim that nature is “surrounded and interpenetrated by another intangible and mysterious world” which is “not                                                           286 For example, Turner 1874; Barton 1983; Macleod 1982, Lightman 1987; Harvey 2013.   287 Huxley 1892c; 1895b.  288 Huxley 1892c, 3.   109  merely beyond, but above, Nature.”289 According to Huxley, this different “frame of things in which their lives are set” has led them to apply the concept of knowledge to the other world.290   [I]t is obvious that, on this theory of the Universe, the successful conduct of life must demand careful attention to both worlds; and, if either is to be neglected, it may be safer that it should be Nature. In any given contingency, it must doubtless be desirable to know what may be expected to happen in the ordinary course of things; but it must be quite as necessary to have some inkling of the line likely to be taken by supernatural agencies able, and possibly willing, to suspend or reverse that course. Indeed, logically developed, the dualistic theory must needs end in almost exclusive attention to Supernature, and in trust that its over-ruling strength will be exerted in favour of those who stand well with its denizens.291  According to supernaturalism, we live in a universe that consists of natural and supernatural worlds, and knowledge about the supernatural world is not just necessary but more important. The seekers of knowledge about the supernatural world seem to have relied on a different kind of method than the method of verification. Huxley examined their method and its qualifications.   Huxley pointed out that history shows that “the field of the supernatural has awarded its cultivators with a harvest”: “an almost infinite diversity of Religions.”292 Religions which mainly consist of “information about Supernature” tell us about “the attributes of supernatural beings, of their relations with Nature, and of the operations by which their interference with the ordinary course of events can be secured or averted.”293 The proponents of supernaturalism, notably theologians and clerics, have preached their own knowledge about the supernatural world, which                                                           289 Ibid., 3-4.  290 Ibid., 3 291 Ibid., 4-5.  292 Ibid., 6.  293 Ibid.   110  they regard as the most important for our life. How did they arrive at such knowledge? Huxley found them appealing to faith.       Huxley noted that, to religious thinkers, having faith does not simply mean that one has a strong conviction that a given proposition is true. It also involves the notion that the proposition, or the content of faith, is a kind of special knowledge revealed or mediated by a supernatural entity such as a god. To them, having faith can serve as evidence for the truth of the content of faith, because having faith is understood to imply that one has received knowledge from a god or one has exercised divinely given faculties. We can find this conception of faith as knowledge from one’s blurring the distinction between “subjective and objective verities” or “the region of speculation and that of fact.”294 For example, in his discussion of George Fox’s “inner light,” Huxley observed the following: “When an ordinary person would say ‘I thought so and so,’ or ‘I made up my mind to do so and so,’ George Fox says, ‘It was opened to me,’ or ‘at the command of God I did so and so.’”295    Huxley argued that having or not having faith is an arbitrary criterion for identifying propositions that deserve the title of knowledge. One’s having faith is supposed to help us to determine which propositions are divine knowledge, but the relationship between one’s having faith in a proposition and a proposition being divine knowledge has been established on the basis of the fact that one has feeling that a proposition at issue is certain. Theologians and clerics wish to claim that they have some divine source for their faith, but Huxley concluded that they just appeal to faith only for their favorite dogmas. In other words, holding a particular dogma comes first. For example, the Bible, agreements among Church Fathers, inner-lights, and the notion of                                                           294 Huxley 1887b, 189;1894a, 302. 295 Huxley 1887b, 189. Huxley also made a brief criticism regarding a kind of religious inspiration because it “leads to the promulgation of a fable as divine truth” (1886a).     111  antiquity have been mentioned as a kind of intermediary that connects one’s holding faith with divine knowledge.296 However, according to Huxley, those intermediaries, in the end, only show which belief one wishes to endorse and reject. As for the Bible, all issues raised throughout history regarding which books should be included or excluded and how a given text should be interpreted attest the arbitrary use of having faith. In a similar manner, Huxley discussed “Declaration on the Truth of Holy Scripture” signed by thirty-eight clergymen in 1891. The declaration used “antiquity” to identify “traditionary testimony of the Church” which is supposed to show infallibility of the books of the Old and New Testaments.297 Huxley pointed out that the notion is conveniently indecisive and they had not looked further what is underneath their “tortoise,” and thus, their appeal to antiquity amounts to that “whoso defines the canon defines the creed.”298 According to Huxley, pointing to having faith is not adequate as a way of identifying knowledge unless its proponents establish a non-arbitrary connection between one’s strong conviction and one’s having (divine) knowledge.     The arbitrariness of using faith as an indication of knowledge leads to another concern: if we adopt that method, we cannot have grounds for favoring one’s faith over others’. Huxley considered this point to show that having faith cannot serve the role of the method of justification.299 According to Huxley, religious faith has generated mutually exclusive religions, and “their adherents delight in charging each other, not merely with error, but with criminality,                                                           296 Huxley 1892c.  297 Ibid., 27.  298 Ibid., 29, 28.   299 Huxley 1889e; 1892c.   112  deserving and ensuing punishment of infinite severity.”300 For example, Huxley quoted Henry Wace:   What made the Mahommedan world? Trust and faith in the declarations and assurances of Mahommed. And what made the Christian world? Trust and faith in the declarations and assurances of Jesus Christ and His Apostles” (l.c. p. 253).301  Wace would want to regard “Mahommed as an unbeliever” or “infidel,” but Huxley pointed out that Wace ended up confessing that faith has no value as an indicator of truth.302 There is no difference between the faith that has made “the Mahommedan world” and the faith that has made “the Christian world.” 303 Huxley raised the same issue repeatedly. “If the Eastern branch of the Church had a right to reject the Apocalypse and accept the Epistle to the Hebrews, and the Western an equal right to accept the Apocalypse and reject the Epistle,” Huxley wrote, “any other branch would have an equal right, ….”304 Similarly, we cannot arrive at an agreement over how to read a given religious text, because every interpretation would be equally well based on its adherents’ faith.  Adopting having faith as a way of identifying knowledge has not done what it is supposed to do; instead it has produced, history shows, the variety of religions and its denominations. These reflect, according to Huxley, the inadequacy of faith as a criterion for knowledge. An appeal to faith has been made arbitrarily and leads to incompatible “knowledge.”                                                           300 Huxley 1892c, 6.  301 As quoted in Huxley 1889e, 213. The passage is from Henry Wace’s “On Agnosticism,” which were read at the Manchester Church Congress in 1888. 302 Huxley 1889e, 213. 303 Ibid., 213-4.  304 Huxley 1892c, 29.   113  We might live in a dualistic world, but the proponents of supernaturalism have failed to offer an adequate justificatory process that can show that their faith deserves the title of knowledge.     4.3. Physical thinking as a constituent of a guide to knowledge I have discussed Huxley’s analysis of mind to show his view of a way of identifying knowledge. Among all groups of perceptions or mental contents, memory has been considered to have default credibility because it contains an idea of the past existence of a corresponding set of impressions. This constitutive idea is present in our consciousness only when we experience a set of impressions. From the analysis, Huxley found an epistemological lesson: by trying to have an experience of a set of impressions, we can check whether a given perception has the constitutive idea, and thereby determining whether it is trustworthy or not. This is the method of verification. Thus, we need to test an explanation at issue against an observation or experiment.  Huxley further argued that the method of verification is the only way of identifying knowledge. The traditional way of searching for necessity turns out to be a search for perceptions that have been repeatedly verified. The religious way of identifying knowledge, appeal to faith, is not qualified as a method of justification because it tends to be applied arbitrarily and it cannot function as a discriminator.  Given that the method of verification is the only legitimate way of identifying knowledge that is available to us, Huxley can provide us with a reason for adopting physical thinking, not non-physical thinking. In the previous chapter, we saw that Huxley thought that we may understand phenomena, the contents of perceptions, either physically or non-physically. If we adopt physical thinking, we are to understand a given phenomenon in material terminologies, ultimately matter and motion; we aim to offer a materialistic-mechanistic account even when a  114  given phenomenon appears to us as a spiritual or holistic phenomenon. If we adopt non-physical thinking, we are to understand a given phenomenon in immaterial terminologies and accordingly aim to offer an account in terms of spirit, vital force, divine will, or some immaterial entity. According to Huxley, the method of verification works better with physical thinking, and this suggests that a physical account can be shown to be justified.   To accept an offered account as knowledge, we should justify it. In other words, both physical and non-physical accounts of a given phenomenon, which are present to us as complex ideas, should be tested against having a series of impressions. This means to Huxley that we need to have a translation or derivation of an offered account in terms of impressions. This kind of work which is necessary for us to go through a verification process would be easier for physical thinkers than for non-physical thinkers, because physical thinkers tend to understand a given phenomenon in terms of objects, properties or events that can be observed or measured. Physical thinkers will be able to indicate what to be observed or experimented to test their account. Their account allows us to use our experience of having complex impressions.305   On the other hand, if we adopt non-physical thinking, the method of verification would not work very well, because immaterial objects and events which interest non-physical thinkers are assumed to lack properties about which information of impressions concerns. Some non-physical thinkers may have objections to using the method of verification, and thus they may claim that their account cannot be tested by trying experiencing of complex impressions. This, for Huxley, amounts to giving up the title of knowledge, since verification is the only way by which we can determine what to count as knowledge. Other non-physical thinkers may be willing to                                                           305 If we had only sensations of smell and hearing, according to Huxley, we “might have a conception of time, but could have none of extension, or of resistance, or of motion” (Huxley 1879a, 306). In this case, even if we somehow manage to come up with a notion of matter, physical thinking would not be then effective in our attempt to use the method of verification to test a physical account.     115  connect their account with other ideas that can be eventually tested against a set of impressions. However, Huxley seems not to have encountered such efforts. Refusing to participate in investigation of spirituality, he conjectured that the inquiry “would involve much trouble and (unless it were unlike all inquiries of that kind I have known) much annoyance.”306 If the proponents of spiritualism showed that they can derive, from their account, a detailed and definite causal or structural explanation for, say, a table’s moving two inches away, Huxley would find it worth examining. Non-physical accounts tend to be simply an invocation of immaterial agency, for which verification is almost useless, and in most cases, Huxley pointed out that there is an alternative physical account that has been already verified.  If non-physical thinking offers an account that can be hardly justified by verification, it should not be considered a way to knowledge. If one seeks knowledge, Huxley urged, it is better to adopt physical thinking in the first place and to hold the view that a mechanical account in material terms is universally applicable, because a physical account is explanatory and can be shown to be justified. This is just another way of saying that an explanans (as opposed to an explanandum) should better be a statement concerning a material phenomenon to be counted as a candidate for knowledge.307 Huxley’s exclusion may raise one worry. If physical thinking is not universally applicable, it could not be regarded as an approach that is proper to be included in a guide to knowledge; if a non-physical account, although unlikely, can be shown to be explanatory and susceptible of justification, we do not have to regard physical thinking as the only approach for                                                           306 Huxley 1871d, 144. 307 This also means that only material events will count as events that cause other events. Huxley has been known, in the literature of philosophy of mind, for the view that mental events are effects, not causes of other events. His view of the mind-body problem has attracted attention from several philosophers, and they have disputed over how to understand his view. I will briefly introduce the interpretative issue and my understanding in footnote 537.     116  our guide knowledge. This worry would arise because of metaphysical concerns, but Huxley found no grounds to think that there exists some metaphysical barrier against the universal application of physical thinking. Once the path of legitimate Idealism is taken, the type of content of perception – either material or immaterial – is irrelevant. What matters instead, according to Huxley, is which type of terminology we should let monopolize the explanatory task because we cannot adopt both at the same time and we are trying to achieve our epistemic goal by using a more effective way.308 Against the claim that a mechanistic-materialistic approach is not applicable to vital and mental phenomena, Huxley said, he “can discover no logical halting-place.”309  Thus, for example, Huxley applied the same approach to the generation of perception. It does not matter whether an explanandum is the statement about material (e.g., the movement of blood) or immaterial (e.g., the generation of perception) phenomena, because both are after all contents of our perception; it does not matter whether an explanandum concerns our perception, because we also have perceptions about perceptions. As we explain the perception of blood movement by utilizing contents of other perceptions, Huxley thought, we explain the perception of perception generation by utilizing contents of other perceptions. He summarized his position: “we have as much right to believe that the sensation is an effect of the molecular change, as we have to believe that motion is an effect of impact; and there is as much propriety in saying that the brain evolves sensation, as there is in saying that an iron rod, when hammered, evolves heat.”310                                                           308 Huxley 1868a.  309 Ibid., 154. 310 Huxley 1874a, 239.   117  According to Huxley, Hume had made the same point. In Hume’s Treatise of Human Nature where the immortality of the soul is discussed, Huxley found Hume arguing against the claim that a cause of thought or perception cannot be motion of a body.311 Hume had pointed out that, judging from our experience, the connection between motion and motion is no more secure than the connection between motion and thinking. The main idea is that we should employ the same reasoning, if we want to employ one at all. That is, if we want to say that our experience shows that motion of one body causes motion of another body, we also have to say that our experience shows that motion of a body causes thinking. After introducing a long quotation from Hume, Huxley developed Hume’s point and summarized as follows:   The upshot of all this is, that the “collection of perceptions,” which constitutes the mind, is really a system of effects, the causes of which are to be sought in antecedent changes of the matter of the brain, just as the “collection of motions,” which we call flying, is a system of effects, the causes of which are to be sought in the modes of motion of the matter of the muscles of the wings.312   As Huxley understood him, Hume had overcome the reluctance to entertain the notion of the interaction between the material and the mental and had “grasped the fundamental truth, that the key to the comprehension of mental operations lies in the study of the molecular changes of the nervous apparatus by which they are originated.”313 Huxley found no difficulty in applying physical thinking to vital and mental phenomena, but he was well aware that his approach had been loosely called “materialism,” according to which “there is nothing in the world but matter, force, and necessity.”314 It has been said that he                                                           311 Huxley 1878h.    312 Ibid., 92. 313 Ibid., 94. 314 Huxley 1874a; 1878h, 94; 1868a, 154, 162.   118  should not have been so surprised when his critics labeled him as a “materialist.”315 What surprised him, however, was probably that his efforts to highlight the “union of materialistic terminology with the repudiation of materialistic philosophy” had not been properly understood or simply had been neglected, rather than the charge of being a materialist.316    As discussed in the previous chapter, material terminology should not be read as an ontological statement. For example, consider his position on the generation of sensation: he said, it had been demonstrated by observations and experiments that “the immediate antecedents of sensations are changes in the nervous system.”317 It should be understood to mean that one type of perception (about sensation) follows another type of perception (about the changes in the nervous system) in our mind. Thus, his view that has been accused of materialism, Huxley explained, makes a claim about what is going on in our mind, not in reality: a perception whose content concerns brain activity antecedes a perception whose content concerns sensation. We should not forget that all things that explain or need to be explained are contents taken from our perception; at the end of the day, our attempt to offer an account amounts to saying which state of consciousness is followed by what. This should be remembered regardless of whether we are to formulate an account in material or immaterial languages. It is in this context that Huxley further argued for adopting a materialistic-mechanical approach, because this approach offers an account with explanatory value and allows us to proceed the process of justification, verification. In this sense, Huxley claimed that only the material terminology enables us to “practically” interpret the facts of consciousness.318                                                             315 Lyons 2012, 92. 316 Huxley 1868a, 155. 317 Huxley 1878h, 89. This point has been discussed by Huxley in detail in other essays (1874a; 1879a). 318 Huxley 1870c, 194.    119  Huxley’s endorsement of the materialistic-mechanistic approach should be understood in the connection with his starting point that perceptions certainly exist and are only things given to us. This consideration is needed not only to see the difference between metaphysical materialism and what he called “legitimate materialism.” But this also helps us to see how Huxley wanted to combine Descartes’s two paths. We do not follow one path or the other because a metaphysics compels us. Adopting the materialistic-mechanistic approach involves the recognition that we are dealing with mental contents. In this manner we should follow Descartes’s second path; we should hold legitimate materialism. It is, according to Huxley, “neither more nor less than a sort of shorthand Idealism; and Descartes’ two paths meet at the summit of the mountain, though they set out on opposite sides of it.”319 As we will see in Chapter 6, Huxley’s understanding of science and its progress shows that he genuinely held the combination of legitimate Idealism and Materialism, not metaphysical Idealism or Materialism.                                                               319 Ibid.  120  ▌CHAPTER 5  Agnosticism as a Guide to Knowledge      In previous chapters, I discussed Huxley’s “Cartesian project”: Huxley tried to find a way of identifying knowledge that works within our epistemic condition. His journey led him to the position that we should adopt the approach of Materialism and use the method of verification to find knowledge. In doing so, Huxley urged, we should not mistake adopting the approach of Materialism for metaphysical Materialism; that is, we should not forget about our epistemic condition as specified in legitimate Idealism.   Huxley’s Cartesian project illuminates the nature and structure of his agnosticism; also its conclusions play roles behind his agnosticism. This chapter elaborates on that suggestion. Huxley intended agnosticism to be a (quick) guide to knowledge, consisting of an account of legitimate evidence and an ethics of knowing. To support my understanding, I will also argue that Huxley’s agnosticism should not be understood as Cliffordian ethics of belief, and discuss why Huxley situated agnosticism within the tradition of modern critical philosophy.   5.1. The two paths in agnosticism: an account of legitimate evidence I have noted that Huxley’s agnosticism has been associated with either of the two theses: first, we cannot know some metaphysical and theological matters because of our conditioned faculties; second, scientific method is the only means of attaining knowledge. This situation is understandable since it is beyond doubt that he claimed both. This fact, as seen in Chapter 2, has  121  invited two interpretations of Huxley’s notion of agnosticism. Agnosticism seems to be about our ignorance, and he appears to have used both ideas to explain what we cannot know.   The conception of Huxley’s agnosticism as an account of ignorance does not help us in understanding how he defended epistemic value of scientific method. Huxley’s conception of scientific method is featured in his agnosticism, but it has not been part of the analysis of his agnosticism. Huxley’s commitment to science, or more precisely, his dogmatic commitment to science is something that needs to be pointed out regarding his promotion of agnosticism. His conception of science has been assumed to stand as independent from his agnosticism. This perspective, as discussed in Chapter 2, has led to the criticism that Huxley’s endorsement of agnosticism either poses the philosophical problem that he could not defend science as the only way to knowledge or offers no solution to that problem.   It is not a fair interpretive framework to view Huxley’s agnosticism solely as an account of ignorance or nescience, given his discussion and practice of agnosticism which will be brought up throughout this chapter. With his Cartesian project, we can have a better understanding of his notion of agnosticism, because we can appreciate what he was trying to do with agnosticism by connecting different ideas with a single position labeled “agnosticism.” More specifically, as I will show in the following, the same concern and interest led Huxley to a Cartesian project and agnosticism, namely that we need a method for identifying knowledge. Also, the result of his project played a role behind agnosticism as his epistemological position. In a sense, his agnosticism was a summary and didactic formulation of his view of how to find knowledge.  Thus, instead of viewing Huxley’s agnosticism as an account of ignorance, I suggest understanding it as an account of knowing. Huxley’s discussion of the two paths and his effort to combine them seem to have been presented by commentators as two alternative accounts of  122  ignorance: we cannot know about matters that go beyond phenomena; we cannot know matters to which scientific method is inapplicable. However, I will argue, his discussion and effort are presented in his agnosticism as an account of legitimate evidence. Huxley’s agnostic principle has been understood by some commentators as an ethics of belief and by others as a humble stance toward our ignorance or the unknowable. I will argue that the agnostic principle was an ethics of knowing that concerns a stance toward evidential valuation.    First note that the fact that Huxley appears to have suggested two different ideas in connection with agnosticism is compatible with my claim that his agnosticism developed from his discussions of Descartes’s two paths. In fact, we can see more than compatibility. It is not surprising that his discussions of agnosticism involve “two components.” Combining the two components which appear to invite the alternative interpretations is exactly what Huxley intended to do, as he wanted to “reconcile” the physical and metaphysical followers of Descartes. Thus, if we ground his agnosticism on only one of the two components, we end up neglecting his intention. Regardless of whether he had, philosophically speaking, succeeded in combining the two paths, we need to try to understand first what he was up to.    Why did Huxley highlight the two paths found in Descartes’s writings? If we stick to the conception of agnosticism as an account of ignorance, we may answer that the two paths somehow show us what we cannot know. That is, Huxley’s intention was to set the boundary of knowledge. The proponents of a partial approach would note once again that Huxley drew two lines for the boundary. Thus, we are still left with the two ways of interpreting agnosticism even after taking Huxley’s discussions of Descartes into consideration.  However, this answer does not correctly represent Huxley’s intention behind his discussions of Descartes. As seen, his Cartesian project and subsequent journey were not geared  123  toward a conclusion about the realm of ignorance, non-knowledge. Rather, his primary aim was to find a guide to knowledge that can inform us how we can arrive at something certain within our limits. If we can admit that there is some parallel between his discussions of Descartes and of agnosticism, which I highlighted in Chapter 3, it can be misleading to frame his agnosticism as a pointer to the realm of our ignorance, because this perspective directs attention to what he said about our ignorance, not to what he was at pains to convey.   Huxley’s various essays show that his major concern was knowledge, something about which we are entitled to be certain.320 For example, in the prologue of Essays upon some Controverted Questions in which the three essays on agnosticism were republished, he wrote:    But, however the polemical concomitants of these discussions may be regarded – or better, disregarded – there is no doubt either about the importance of the topics of which they treat, or as to the public interest in the “Controverted Questions” with which they deal. Or rather, the Controverted Question; for disconnected as these pieces may, perhaps, appear to be, they are, in fact, concerned only with different aspects of a single problem, with which thinking men have been occupied, ever since they began seriously to consider the wonderful frame of things in which their lives are set, and to seek for trustworthy guidance among its intricacies.321  The main issue of Victorian society which he shared or wanted to raise was to identify “the knowledge essential to the right guidance of life”; the “Controverted Question,” according to Huxley, involves adopting a mode of inquiry for such knowledge.322 In addition, he did not consider that being a truth is sufficient for a proposition to be knowledge and he was concerned                                                           320 For example, Huxley 1868a; 1870c; 1878h; 1892c.  321 Huxley 1982c, 3.  322 Ibid., 22, 7. In this essay, Huxley contrasted two modes, naturalism and supernaturalism, and argued for the former. He formulated the “controverted question” more precisely as follows: “The question – how far is this process [of eliminating the supernatural] to go? – is, in my apprehension, the Controverted Question of our time” (Ibid., 7).   124  more with justification. Thus, he often made the following points: a given belief may be true, but not subject to verification; it may be true, but not evidence.323 As we will see shortly, Huxley’s discussions of agnosticism always involve the issue of evidence or justification conferring factor.  My suggestion was that Huxley’s agnosticism conveys his views on one of our epistemic activities, claiming to have knowledge. Accordingly, we can see that his notion of agnosticism includes an account of legitimate evidence, which determines whether or not one is entitled to make a claim to knowledge. His discussions of Descartes’s paths were meant to show what to count and not to count as having evidential value.   First consider Descartes’s path of Idealism. The point that we cannot go beyond our consciousness indicates our limited condition, which in turns shows the nature of knowledge. Our knowledge is knowledge of consciousness, based on what appeared and appear in consciousness. This is what Huxley took as the most important message of Hamilton’s and Mansel’s thoughts: our knowledge can only be about “the relative and finite.”324 Consciousness presents us something as appeared, not as it really is, and in succession, not in an infinite or eternal manner. Of course, there are some perceptions whose contents concern something beyond our consciousness. For example, people have entertained ideas about an entity that is thought to comprise a real world or a being that is assumed to exist independently from our consciousness. We have perceptions about such an entity, and yet, it has been claimed, these perceptions represent such an entity as it stands without any relation to our consciousness, or the                                                           323 For example, Huxley 1863d; 1876c.  324 Huxley 1860c, 315; 1889e; 1890a; 1895a. However, this was the only point that Huxley agreed with Hamilton and Mansel. In his early days, Huxley wrote in a letter: “I believe in Hamilton, Mansell [sic] and Herbert Spencer so long as they are destructive, and I laugh at their beards as soon as they try to spin their own cobwebs” (1863c, 349). The same point is also found in Huxley’s later letter and last essay (1889i; 1895b).  125  “Absolute.”325 However, according to Huxley, we are not in a position to claim whether or not these perceptions correctly capture a thing in itself.326 As noted, this has been formulated as the thesis of metaphysical and theological ignorance.      Huxley’s further point is that the realm of our ignorance should be excluded from our activity of knowing. We cannot use what we are incapable of knowing in order to find knowledge, because a perception that cannot be found trustworthy cannot show whether or not other perceptions are trustworthy. In short, what is not justifiable is not justificatory. For example, consider Huxley’s discussion of miraculous events such as Jesus walking on water or transferring evil spirit from people to pigs.327 He acknowledged that Christian beliefs in the miracle had been disputed in terms of whether or not such a miraculous event can happen, but he disagreed with the rationales of both believers and disbelievers. According to Huxley, due to “the limitations of our faculties,” “we never can be in a position to set bounds to the possibilities of nature.”328 The believers’ reason that the order of nature can be violated by a being like Jesus, and the disbelievers’ reason that the present science determines what is possible in nature are equally grounded in something that we are not capable of knowing. If they think that such an appeal can support their claim, they are claiming that a speculation or imagination has justificatory power. However, propositions regarding issues that require us to exceed our                                                           325 Huxley 1890a; 1895a.   326 Huxley accepted the concept of appearance and of thing-in-itself, but he did not think that we know that appearance differs from thing-in-itself (1878h).  327 Huxley 1878h; 1889e; 1892a.  328 Huxley 1892a, 198. In 1866, Huxley made the same point in a letter to an editor of Spectator, part of which, interestingly, Alfred Russel Wallace quoted, along with a passage from Herschel, in the front page of his book, The Scientific Aspect of the Supernatural: Indicating the Desirableness of an Experimental Enquiry by Men of Science into the Alleged Powers of Clairvoyants and Mediums: “With regard to the miracle question, I can only say that the word ‘impossible’ is not to my mind applicable to matters of philosophy. That the possibilities of nature are infinite is an aphorism with which I am wont to worry [weary] my friends” (quoted in Wallace 1866; Huxley 1866a; also see Prasch 2015, 34, n.6).    126  limitations cannot serve as evidence. People may speculate about such issues, but Huxley argued, they must admit that their speculation can have no evidential value. Neither does embellishing their speculation by capitalizing the first letter of words like “Force,” “Absolute,” or “Reason” render evidential power. This is, Huxley sarcastically remarked, giving “a Grenadier a bearskin cap, to make him look more formidable than he is by nature.”329 Huxley’s agnostics say not only that we are incapable of knowing whether God exists or not, but also that we are incapable of using God’s existence or inexistence as evidence for other claims.330 The lesson of Idealism tells us that no realm outside consciousness is a place to look for evidential value.  What can be then considered to be evidence? Evidence should be found among what appear in our consciousness. To Huxley, evidence is, above all, a perception that can support one claim (perception), and not others. Not all contents of mind, however, can play a role of evidence. Our consciousness sometimes presents to us inconsistent perceptions at different times or places; it also presents to us perceptions contradictory to other perceptions. We need to find trustworthy perceptions to have evidence. If unjustifiable statements cannot function as evidence, neither can unjustified statements. Huxley applied the lesson of Materialism. To turn perceptions into perceptions with evidential power, physical thinking works better. For a mental content at issue should be expressed in terms of what-have-been-experienced and what-will-be-experienced so that it can be shown to be justified and thus can be used as evidence.    Huxley’s suggestion amounts to the idea that we should only take a perception that has been verified as evidence, and we can see whether a given perception has been verified only                                                           329 Huxley 1870c, 179. 330 Relatedly, Huxley wrote, regarding “syllogisms” that concern “Substance,” that “the premises of [the syllogisms] convey no meaning, while the conclusions carry no conviction” (1878h, 211).   127  when it is expressed as an effect of interactions of material objects about which we can ultimately have a complex impression. This idea is folded in his illustrations like the following: “It [agnosticism] simply means that a man shall not say he knows or believes that which he has no scientific grounds for professing to know or believe.”331 It is not surprising that Huxley’s agnosticism has been understood to suggest that science is the only way to knowledge.  However, a logically prior point is Huxley’s view of legitimate evidence.332 This view, according to Huxley, is not peculiar to (narrowly understood) scientific disciplines; other areas like history and law share this account of evidence.333 Moreover, when Huxley discussed an agnostic’s position on controversial issues like miracles, he did not simply dismiss beliefs in miracles by pointing out that miraculous events are discordant with scientific discoveries up to his days. Instead, Huxley’s focus is on believers’ evidence for their claim; he went through the trouble of examining their evidence and judged whether offered evidence indeed has evidential value or not.334 I will return to this point with specific examples when I discuss what his agnostics are supposed to do.        Huxley’s epistemological position that he presented in his discussions of the modern philosophers such as Descartes, Berkeley and Hume is layered behind his agnosticism. Rehashing it as qualification for evidence would effectively serve Huxley who wanted to deal with a claim to knowledge. From Huxley’s perspective, Victorian intellectual contemporaries were suffering                                                           331 Huxley 1884a, 5.  332 Huxley did not express interests in the concept of evidence only in his later years. In 1853, he anonymously published “The Valuation of Evidence” which shows his interests in this issue. In fact, around mid-19th century, due to the popularity of spiritualism and séances, there were discussions of evidence and testimony (See Lamont 2004). In the anonymous essay, Huxley mentioned spirit rapping and table moving (1853).   333 Huxley 1886b; 1889f; 1890c.   334 Huxley 1889e; 1892a.  128  from loose and incorrect conceptions of knowledge or knowing. Denying the evidential power of the unknowable and unknown and affirming the evidential power of experience were his fundamental theme found in his Cartesian project and practice of agnosticism. Saying that we cannot know is surely a confession of our ignorance, but it is also a declaration that we cannot find any evidential value.335 Thus, the identity of agnostic would be better captured with the remark “Tell me your reason” or “There is no evidence,” rather than “I don’t know” per se. Huxley described “a true agnostic” as follows:  … if you were to meet with such a phœnix and to tell him [an agnostic] that you had discovered that two and two make five, he would patiently ask you to state your reasons for that conviction, and express his readiness to agree with you if he found them satisfactory. The apostolic injunction to “suffer fools gladly” should be the rule of life of a true agnostic. I am deeply conscious how far I myself fall short of this ideal, but it is my personal conception of what agnostics ought to be.336    We can also see that Huxley himself did not merely say “I don’t know,” but further moved on to discuss the issue of evidential value. For example, when discussing the doctrine of the immortality of the soul, Huxley said: “Give me such evidence as would justify me in believing anything else, and I will believe that.”337 With respect to the claim that Jesus had said and done what is written in the Bible, according to Huxley, an agnostic would try to determine the value of evidence and then say, “I cannot find good evidence that so and so is true.”338                                                            335 As I understand Huxley, he tended to highlight confession of ignorance when a discussion involves his criticism of idolization of the unknowable. For example, in a letter, Huxley contrasted a confession of ignorance to “the apotheosis of ignorance under the name of the ‘Absolute’ or its equivalent” (Huxley 1889h in Lightman 1987, 13). I will return to this issue in Section 4 of this chapter.    336 Huxley 1889e, 246-7. Also Huxley 1889f; 1892c.  337 Huxley 1860c, 314; see also 1870b. In a similar manner, Huxley replied to his disputant: “It may be so, or it may not be so; but where is the evidence which would justify any one in making a positive assertion on the subject [the order of creation]?” (1886e, 177). 338 Huxley 1889e, 212. I cannot agree more with Asprem’s following statement:   129   If we pay due attention to Huxley’s persistent concern over when one can say that we attain knowledge, which is also the concern expressed in his discussion and practice of agnosticism, we need not see him as suggesting two alternative accounts of ignorance. Huxley claimed that we are not capable of knowing something beyond our consciousness; he also claimed that science is the only means of attaining knowledge. These two ideas are traces of his Cartesian project which was to find a guide to knowledge that can work regardless of our conditioned faculties. In agnosticism, they are intended to work together to provide an account of legitimate evidence, which Huxley considered to be crucial and urgent information for Victorians who aimed to seek knowledge.   5.2. Another component of agnosticism: the agnostic principle  In addition to the combined instruction regarding evidence, Huxley’s guide has another instruction. The additional component concerns what an epistemic agent should do. According Huxley, Descartes had followed a rule which Huxley called “Descartes’ maxim” or “golden rule” – “assent to no proposition the matter of which is not so clear and distinct that it cannot be doubted.”339 This maxim had regulated Descartes’s “assent” regarding the truth value of a given proposition and required him not to “lie” to himself about the certainty of a given proposition.340   Before Huxley began to discuss agnosticism publicly, he had introduced Descartes’s maxim when discussing how to reconcile metaphysical thinkers and physical thinkers who                                                           The valuing of evidence is thus a crucial element of Huxley’s agnosticism that often goes overlooked or misunderstood – both in his own days and in later conception of agnosticism. The disbelief associated with agnosticism is not of a purely a priori character, but deals rather with the justification of claims and the coherence of knowledge in general (2014, 295-6, her italics). 339 Huxley 1870c, 169, 194. 340 Ibid., 171.   130  exclusively followed only one of Descartes’s two paths. Huxley illustrated this maxim as the condition for their reconciliation.341 They have their own view on reality and their metaphysical views are incompatible. We can mediate both parties while preserving their own mode of thinking, Huxley thought, by having them admit that their different views concern something that we cannot be certain about. If they complied with Descartes’s maxim, they would not have assented to a proposition regarding a real world. Huxley wanted to reconcile the two parties, because he could neither take one side nor give up either type of thinking; each type of thinking has something to say to us about how to attain knowledge. Moreover, Huxley had an example – Descartes who had somehow managed to employ both ways of thinking.   As Descartes had had a maxim, Huxley also included a similar rule in his guide to knowledge, the agnostic principle.342 Yet, the agnostic principle should not be understood as a merely re-named Cartesian maxim, because Huxley focused on an idea underlying the maxim rather than on unpacking the meanings of Descartes’s words such as “clear” and “distinct.” Huxley introduced other thinkers whom he considered to share Descartes’s maxim such as Socrates.343 For example, consider the following series of quotes.                                                             341 “The reconciliation of physics and metaphysics lies … in the observance by both metaphysical and physical thinkers of Descartes’ maxim–assent to no proposition the matter of which is not so clear and distinct that it cannot be doubted” (Ibid., 194). 342 Huxley mentioned this principle in many places, but the exact expression “agnostic principle” or “Agnostic principle” appears in the following: Huxley 1889e, 246, 249, 253; 1889g, 310, 313, 317, 327 (this list is not exhaustive).  343 Huxley 1871c; 1878h; 1889e; 1893-4 (Vol. VI); 1895a. Although Huxley was willing to credit Hamilton and Mansel for holding the doctrine that we are ignorant of what is beyond phenomena and further he called them “agnostics,” Huxley did not introduce them as those who held the maxim. Even before Huxley mentioned Hamilton in connection with agnosticism, Huxley had considered Hamilton to overstep his own boundary of ignorance and thus Huxley could not see him as a follower of the maxim (1860c, 315; 1895a; 1895b). Huxley appears to have considered Mansel to be one of like-minded persons in an enemy camp: when Huxley “came across the Limits of Religious Thought,” Huxley found “the thrill of pleasure” because “I [Huxley] was as orthodox as a dignitary of the Church, …” (1985a, 534).    131  It is the Cartesian doubt – the maxim that assent may properly be given to no propositions but such as are perfectly clear and distinct – which, becoming incarnate, so to speak, in the Englishmen, Anthony Collins, [John] Toland, [Matthew] Tindal, [Thomas] Woolston, and in the wonderful Frenchman, Pierre Bayle, reached its final term in Hume.344   Berkeley and Locke, each in his way, applied philosophical criticism in other directions; but they always, at any rate professedly, followed the Cartesian maxim of admitting no propositions to be true but such as are clear, distinct, and evident …345  That [Agnostic] principle is of great antiquity; it is as old as Socrates; as old as the writer who said, “Try all things, hold fast by that which is good”; it is the foundation of the Reformation, which simply illustrated the axiom that every man should be able to give a reason for the faith that is in him; it is the great principle of Descartes …346   He [Socrates] also persisted in demanding that no man should “take anything for truth without a clear knowledge that it is such.”347   Huxley appears to have seen an idea shared by them: one should assent to a proposition only when one has grounds. We might be tempted to compare this idea with the position now called “evidentialism,” according to which one ought to believe something based on evidential reasons as opposed to non-evidential reasons such as prudential reasons.348 From one aspect, the shared idea seems to restrict grounds for assent to a particular type and, as I have discussed, Huxley had a specific view of what can or cannot serve as evidence. Thus, in a sense, it sounds similar to evidentialism. Yet, from another aspect, the shared idea is not identical with evidentialism,                                                           344 Huxley 1871c, 245. 345 Huxley 1878h, 66. 346 Huxley 1889e, 245-6. The old aphorism is from Thessalonians (5:21) and he also mentioned it in his earlier essay (1887d, 18).  347 Huxley 1893-4 (Vol. VI), viii. 348 Aikin 2008; Reisner 2008; Wood 2008; Chignell 2016.   132  because it seems not to concern a way of forming a belief and it does not explicitly convey the notion that there is a right way of forming a belief. My point here is that, at this moment, we do not have information enough to formulate the shared idea as a clear thesis. What is clearer seems to me that Huxley wanted to introduce a rule regarding our attitude toward the certainty of a given proposition, and that the rule has to do with a particular understanding of grounds. We can tentatively say that Huxley saw a sort of “evidentialistic attitude” shared by the past knowledge or truth seekers, but we need to see how he illustrated it in his own terms.   Descartes’s maxim, or the shared attitude that Huxley wanted to introduce, was later formulated as the agnostic principle. Huxley illustrated his principle in different ways and practiced it in many places, but the following quotes are where he explained it most explicitly.    Positively the principle may be expressed: In matters of the intellect, follow your reason as far as it will take you, without regard to any other consideration. And negatively: In matters of the intellect, do not pretend that conclusions are certain which are not demonstrated or demonstrable.349    This principle may be stated in various ways, but they all amount to this: that it is wrong for a man to say that he is certain of the objective truth of any proposition unless he can produce evidence which logically justifies that certainty.350   Huxley’s descriptions of the agnostic principle show how he understood the shared attitude. As we see in the first quote, he expressed the agnostic principle “positively” and “negatively.” First consider the negative characterization concerning what his principle says to be wrong. The principle regulates an act of “saying,” “professing,” or “pretending” rather than believing. It also                                                           349 Huxley 1889e, 246. 350 Huxley 1889g, 310.  133  concerns the certainty of the truth of a proposition as opposed to the truth of a proposition. Neither simply holding a belief nor professing a belief about the plausibility of a proposition is subject to the agnostic principle. Notice that the above illustrations include “certainty,” and remind that Huxley understood knowledge in terms of certainty. In short, his agnostic principle targets a claim to knowledge.   Huxley’s agnostic principle also specifies conditions under which it is wrong to claim a proposition to be certain or knowledge: we are allowed to do so only when a given proposition is demonstrable and has been demonstrated, or only when our certainty can be justified by evidence. The expressions like “not demonstrated or demonstrable” or “evidence” are not neutral, of course. To understand the conditions under which a claim to knowledge should be made, we need to take his view of evidence into consideration, which I discussed in the previous section. Accordingly, the agnostic principle prohibits a particular type of action. If one claims one’s belief to be certain without offering any evidence or with something that in fact lacks evidential value, the person is doing something wrong according to the agnostic principle.    Next, consider the other illustration of the agnostic principle: “follow our reason as far as it will take us, without regard to any other consideration.” We can see here something “evidentialistic,” if we understand “follow our reason” to mean taking evidential reasons into consideration, and “any other consideration” to mean any non-evidential reasons. Yet the agnostic principle also sounds similar to Hume’s “wise man” rule that concerns the degree of assurance or certainty, according to which one should proportion one’s confidence in a belief to given evidence.351 We can then understand the principle as follows: take into consideration an                                                           351 Although Huxley did not quote Hume to introduce the rule, there seems to be no doubt that Huxley as an author of Hume was well aware that the rule had been associated with Hume.    134  evidential reason only and apportion a given proposition’s certainty to its evidence. Again, to fill out what he meant by “reason” and “any other consideration,” we need to consult his account of legitimate evidence.    How can we summarize the message of Huxley’s agnostic principle? Generally speaking, the principle urges everyone to apply the concept of knowledge or knowing strictly. More specifically, by proposing the agnostic principle, Huxley intended to express, I suggest, the following three ideas. First, the degree of certainty of a given proposition, or its status of knowledge, is determined by evidential reasons alone. This idea has nothing to do with what one should do, and it basically re-states the other components of agnosticism in more abstract terms: only propositions that have been verified by experience (in addition to a logical relation) can contribute to determining whether or not a given proposition is certain or entitled to the status of knowledge. Second, to form a judgement or opinion about the certainty of a given proposition, we have a duty to find evidential reasons. It is wrong to have an opinion on whether or not a given proposition is certain without examining evidential reasons. Third, we have a duty to follow what given evidential reasons say about the certainty of a proposition at issue. It is wrong to ignore or manipulate the result of evaluation of evidential reasons by, for example, taking prudential reasons into consideration. The two duties are found in Huxley’s writings.352   Huxley’s agnosticism has normative part, and this is an interesting aspect as compared with a definitional analysis of knowledge. Why did he need that part? In a sense, a concept can function as a norm, without having an additional “principle,” and he had an account of evidence embedded in his agnosticism. To examine the implication of having the agnostic principle, the                                                           352 Huxley 1863d; 1878h; 1886e; 1887e; 1890d; 1892b; 1892c.   135  two duties briefly introduced in the previous paragraph should be explained more. In this regard, it would be helpful to contrast Huxley’s agnosticism with Clifford’s ethics of belief.   5.3. Huxley’s agnosticism and Clifford’s ethics of belief   This section examines whether or not Huxley endorsed evidentialism concerning belief formation. His agnosticism has been understood as a type of evidentialism, because scholars, especially in the literature of ethics of belief, have considered him to share the position of William K. Clifford.353 Clifford, who has been also regarded as one of influential Victorian intellectuals, made a strong claim in his famous essay, “The Ethics of Belief”:   To sum up: it is wrong always, everywhere, and for anyone to believe anything upon insufficient evidence.354     According to Clifford, whenever we believe something on the basis of insufficient evidence, we are doing something morally and intellectually wrong. This position is later called “(strong or strict) evidentialism,” according to which a legitimate reason for believing must be evidential as opposed to prudential reasons, although the contemporary version does not usually further include moral evaluation.355 Evidential reasons are reasons related to the truth value of a given proposition and considered to increase its plausibility. On the other hand, prudential reasons are reasons related to potential consequences, usually benefits that one would gain by believing a                                                           353 Scholars have noted that Huxley’s agnosticism and Clifford’s ethics of belief convey the same imperative (James 1896; Livingston 1987; Doore 1983; Aikin 2008; Yoder 2013). Van A. Harvey (2013) is a notable exception, whom I will discuss shortly.   354 Clifford 1876, 295. 355 Reisner 2008; Chignell 2016.  136  given proposition. The famous case in philosophy is Pascal’s wager. Blaise Pascal (1623-1662) suggested believing the existence of God based on prudential reasons: weigh the benefits that we would gain by believing and not believing in God. He claimed, we can gain far more benefits by believing in God; hence, believe in God. Clifford’s imperative prohibits this kind of prudential belief formation.  It is important to examine whether Huxley indeed endorsed Clifford’s imperative, because this issue is related to another issue of whether Huxley violated his own agnosticism when he expressed his faith in axioms of science. Clifford’s imperative says that we are (morally and intellectually) allowed to have a belief based on only sufficient reasons, but there seem not to be sufficient reasons to demonstrate axioms of science such as the uniformity of nature. As seen in Chapter 2, the assumption that Huxley endorsed Clifford’s imperative is one of the factors that have generated the typical criticism against Huxley, according to which Huxley closed his eyes to his own faith and yet criticized his religious targets for their faith.   It is a historically possible scenario that Huxley and Clifford in fact had an identical view. Huxley was Clifford’s academic supporter and dear friend. In 1874, Clifford became a member of the Metaphysical Society on Huxley’s recommendation.356 It was the Society where a shorter version of Clifford’s “Ethics of Belief” was presented one year before its publication. Moreover, it was the place where Huxley felt that his position differed from other members’ and was motivated to coin the term “agnostic.”357 According to Alan Brown who researched and documented the Society, Clifford’s essay was presented in April 11, 1876, but he could not find the participation log for that meeting (when Huxley presented an essay titled “The Evidence of                                                           356 Bivona 2012. 357 Huxley 1889e.  137  the Miracle of Resurrection” three month earlier, Clifford attended).358 However, there seems to be no doubt that Huxley was well aware of Clifford’s position, considering controversies generated by Clifford’s essay. Regarding Clifford’s essay, Timothy J. Madigan writes that “[w]hile exciting strong criticisms among certain of the Society’s members, Clifford also had his defenders” and mentions Huxley, FitzJames Stephens, and Leslie Stephens as the defenders.359 Madigan shows how the Stephens brothers defended Clifford’s position by introducing their writings and by explaining their position, but he does not discuss Huxley’s defense in the same manner and only comments that “colleagues such as Thomas Huxley sprang to his defense.”360 Meanwhile, Gowan Dawson reports that Clifford’s radical view concerned Huxley.361 The most plausible historical conclusion seems to be a minimal one: Huxley knew well about Clifford’s position on ethics of belief. However, there seems to be no document that shows that Huxley committed himself to Clifford’s thesis or that either of them said so. Thus it would be better to compare and contrast Huxley’s agnosticism and Clifford’s ethics of belief.   We need to clarify Clifford’s position first. Following the literature, I assume that Clifford held at least the view that has been characterized as strong evidentialism: anyone as a believer has a moral and intellectual duty to form a belief on the basis of evidential reasons only. Some scholars have argued that Clifford’s ethics of belief is not as simple as expressed in his famous thesis above.362 However, I will not go into specific issues over Clifford’s position because my main concern is how to interpret Huxley’s agnosticism in connection with Clifford, and his                                                           358 Brown 1947. 359 Madigan 2008, 91. 360 Ibid., 85. 361 Dawson 2007. 362 For example, Madigan 2008.  138  agnosticism has been associated with that understanding of Clifford’s position, namely, evidentialism centered on the famous thesis. My interpretative issue here is two-fold: whether or not Huxley endorsed Clifford’s thesis, and whether or not Huxley’s agnosticism incorporated the basic evidentialistic notion in the same manner of Clifford’s ethics of belief.      Note four features of Clifford’s imperative. First, it is moral and intellectual. Second, it is universal in the sense that it is applied to everyone who is capable of believing. Third, it is universal in another sense that it is applied to every case; it does not allow an exception that a failure to fulfill the duty is exempted from blame. Fourth, it targets an act of believing. These features are, we may say, “added on” to the basic evidentialistic notion of considering evidential reasons only.   Although Huxley’s agnosticism and Clifford’s ethics of belief have been grouped together as evidentialism over belief formation, there is a notable exception. Van A. Harvey recently argues that their views have “subtle but important differences.”363 One of differences that Harvey highlights is that Clifford’s imperative concerned anyone’s every belief, whereas Huxley’s imperative (the agnostic principle) “more narrowly directed at anyone who claims to be certain of the objective truth of any proposition but cannot produce evidence justifying that certainty.”364 Thus, an agent who believes p on prudential or inadequate reasons does not violate Huxley’s principle, because she is not subject to it in the first place. When she claims p to be certain, we can then discuss whether she is violating Huxley’s principle or not. According to Harvey, Huxley did “not argue that one requires evidence or justification for every belief that one holds.”365                                                                  363 Harvey 2013, 10. 364 Ibid. 365 Ibid.  139   I am sympathetic to Harvey, because his views on their different targets capture two points that I take as important in understanding of Huxley’s agnosticism. The first point concerns how significantly believing stands as an action or attitude that we take toward a proposition. It is not disputable that Huxley’s agnosticism has something to do with our action or attitude toward a proposition. What is disputable is whether Huxley’s agnosticism takes believing to be the most essential action or attitude that needs to be regulated as Clifford suggested. The interpretation of Huxley’s agnosticism as Clifford’s ethics of belief assumes that believing was the only significant action to Huxley, although Huxley mainly talked about “saying,” “professing,” or “pretending to believe.” According to Harvey, Huxley’s main concern was not believing.   The second point is that Harvey’s interpretation makes a distinction between considering p true and considering p certain, which is often ignored in the interpretation of Huxley’s agnosticism as Clifford’s ethics of belief. When it comes to blameworthiness, Huxley’s concern was unentitled attitudes like showing over-confidence in one’s belief, not holding a belief.366 The difference between believing p and believing the certainty of p might look to be an insignificant point to make, but not for Huxley who distinguished being true and being knowledge, as noted earlier. Evidence does not work as a truth maker of p, but renders justification for turning our feeling certain of p into our being certain of p.        The lack of attention to those two points seems to have led one to consider Huxley’s agnosticism to be identical with Clifford’s ethics of belief. Yet it is still possible that Huxley’s agnosticism implicitly includes Clifford’s thesis even though Huxley had a different focal point. As Harvey points out, Huxley’s agnostic principle focuses on claiming one’s belief to be certain without justification. We can further ask what grounds Huxley had for the principle. Huxley                                                           366 Huxley 1886e; 1889g; 1892c.  140  might have found such assertions blameworthy because, at the end of the day, he considered it to be wrong to believe something on non-evidential reasons. If so, then, despite his wording, Huxley’s view amounts to Clifford’s. Did Huxley endorse Clifford’s universal moral imperative on belief formation? Harvey’s position on this issue is not clear; yet he writes that “[Huxley and Clifford] believed that irrational beliefs had social consequences, and so it was a duty to weigh the evidence for beliefs.”367 This suggests that they had a utilitarian argument for or against a certain type of belief formation.   I have introduced the possibility that because Huxley was committed to Clifford’s moral imperative on belief formation, he would have found it blameworthy to claim a proposition to be certain without having evidential reasons. From this, we can find one line of reasoning behind the interpretation of Huxley’s agnosticism as Clifford’s ethics of belief. Huxley himself took the evidentialistic way of forming a belief as an ideal to pursue and tried not to form a belief on insufficient or “wrong” types of reasons. For example, in a letter to Charles Kingsley, which has been often cited as evidence to show that Huxley had held agnosticism before he coined “agnostic,” Huxley wrote:  I neither deny nor affirm the immortality of man. I see no reason for believing it, but, on the other hand, I have no means of disproving it. … Give me such evidence as would justify me in believing anything else, and I will believe that.368    Huxley, who cries “Oh devil! truth is better than much profit” in the same letter, appears to be committing himself to Clifford’s imperative on belief formation.369 Moreover, Huxley seems to                                                           367 Harvey 2013, 10; 2012.   368 Huxley 1860c, 314. 369 Ibid., 313.  141  have thought that a “true” agnostic is a person who strictly follows the evidentialistic way of forming a belief. In an unpublished fragment on agnosticism, Huxley penned that agnosticism is “the method [into] the cause of belief, which [sic] marks the true agnostic, not the results of the intellectual operation conducted according to that method.”370 If an agnostic is supposed to form a belief based on evidential reasons alone and Huxley also tried to be such an agnostic, he would have made a value judgment on a way of forming one’s belief, which amounts to Clifford’s thesis. Moreover, as the line of reasoning may further go, it is plausible to understand the agnostic principle in terms of an agnostic’s ideal regarding belief formation.   Including the two quotations I just introduced, there are some places where Huxley mentioned “believe” or “believing” in normative tone in connection with his agnosticism (four places, as far as I am aware).371 This might have encouraged scholars to group Huxley and Clifford together, but it seems that William James should be held responsible for the association and its spread. In his famous lecture essay “Will to Believe,” James introduced, with a comment, the last sentence of Huxley’s essay written for a symposium on “Influence upon Morality of a Decline in Religious Belief” right before the quote from Clifford’s essay.372    Huxley exclaims: “My only consolation lies in the reflection that, however bad our posterity may become, so far as they hold by the plain rule of not pretending to believe what they have no reason to believe, because it may be to their advantage so to pretend [the word ‘pretend’ is surely here redundant], they will not have reached                                                           370 Huxley n.d.1, Huxley File’s brackets. It is undated but we can conjecture that Huxley wrote this after 1889, because he referred to the essays on agnosticism published in 1889. 371 The third place is Huxley’s symposium essay that James quoted. It will be introduced shortly. The last example is Huxley’s letter to Watts, which was published, without Huxley’s permission, in the first issue of Agnostic Annual (Lightman 2002; Le Poidevin 2010). Huxley replied to Watt’s question “1. Is Agnosticism in accord with modern science?” as follows: “1. Agnosticism is of the essence of science, whether ancient or modern. It simply means that a man shall not say he knows or believes that which he has no scientific grounds for professing to know or believe” (Huxley 1884a, 5). 372 Huxley 1877d.  142  the lowest depth of immorality.” And that delicious enfant terrible Clifford writes: “Belief is desecrated when given to unproved and unquestioned statements for the solace and private pleasure of the believer [...] … If [a] belief has been accepted on insufficient evidence [even though the belief be true, as Clifford on the same page explains] the pleasure is a stolen one. [...] It is sinful because it is stolen in defiance of our duty to mankind. That duty is to guard ourselves from such beliefs as from a pestilence which may shortly master our own body and then spread to the rest of the town. [...] It is wrong always, everywhere, and for every one, to believe anything upon insufficient evidence.”373   In response to the Cliffordian unqualified moral imperative on believing, James argued that one has a right to believe despite insufficient evidence under certain circumstances.374 If we understand that one’s having a right implies others’ duty to accommodate the right, James’s claim would mean that one has a duty not to require sufficient evidence for holding a belief under some conditions, and thus that sometimes holding a belief upon insufficient evidence is exempt from Clifford’s moral valuation. When introducing one of such conditions, James used the expression “lowest kind of immorality” with which he seems to refer back to his earlier direct quote of Huxley, “the lowest depth of immorality” (in fact, Huxley wrote “the lowest depths of immorality”),375 and considered “our scientific absolutists” to share Clifford’s moral imperative:  There are, then, cases where a fact cannot come at all unless a preliminary faith exists in its coming. And where faith in a fact can help create the fact, that would be an insane logic which should say that faith running ahead of scientific evidence is the “lowest kind of immorality” into which a thinking being can fall. Yet such is the logic by which our scientific absolutists pretend to regulate our lives!376                                                            373 James 1896, 7-8, his brackets.   374 It has been known that James regretted having titled his essay “The Will to Believe” instead of “The Right to Believe” (Harvey 1969).  375 Huxley 1877d, 530. 376 James 1896, 25, his italics.   143  In the symposium essay on the relationship between morality and religious beliefs, Huxley noted that theology had exerted an “enormous influence” on morality (and also vice versa), but questioned the claim that the decline in theological beliefs will negatively affect morality.377 Huxley thought that morality is “strong enough to hold its own,” because the “capacity for the pleasures and pains afforded by sense, by sympathy, or by the contemplation of moral beauty and ugliness” is “obviously in no way affected” by “the abbreviation or the prolongation of his conscious life” or even by “the mere existence or non-existence of anything not included in Nature”; as long as “the constitution of man” remains similar and “he believes that actions have consequences,” Huxley conjectured, social and personal morality would not be lost.378 Thus, Huxley did not share the worry that “without this or that theological dogma the human race will lapse into bipedal cattle, more brutal than the beasts by means of their greater cleverness.”379 Yet, he admitted that he might be wrong and wanted to learn about whether the worry has some grounds. According to Huxley who distinguished religion from theology, many theological doctrines are about nature and are subject to “ordinary methods of investigation.”380 Huxley said: “I have not the slightest doubt that if mankind could be got to believe that every socially immoral act would be instantly followed by three months’ severe toothache, such acts would soon cease to be perpetrated”; “[i]t would be a faith charged with most beneficent works,” he continued, “but unfortunately this faith can so easily be shown to be disaccordant with fact that it is not worth                                                           377 Huxley 1877d. 378 Ibid., 530, 536. Huxley understood morality as follows: “[s]ocial morality relates to that course of action which tends to increase the happiness or diminish the misery of other beings; personal morality relates to that which has the like effect upon ourselves” (Ibid., 537). 379 Ibid., 530. 380 Ibid., 538.  144  while to become its prophet.”381 In the same manner in which he dealt with this “toothache” doctrine, Huxley asked if there are reasons for assuming some sort of connection between the decline in morality and disbeliefs in particular theological doctrines. His point was that if we keep following the ordinary methods of investigation, we would not reach “the lowest depths of immorality,” even if it were true that the loss of the faith in particular theological doctrines weakens our morality.   Huxley’s expression “the lowest depths of immorality” appears to have caught James’s special attention. In an essay written before “The Will to Believe,” James wrote:   With regard to all other possible truths, however, a number of our most influential contemporaries think that an attitude of faith is not only illogical but shameful. Faith in a religious dogma for which there is no outward proof, but which we are tempted to postulate for our emotional interests, just as we postulate the uniformity of nature for our intellectual interests, is branded by Professor Huxley as “the lowest depth of immorality.”382  Here we can also see the typical criticism of Huxley: Huxley was blind to his own faith in axioms of science and yet condemned religious faith. Huxley is here portrayed as one who endorsed the Cliffordian imperative but gave himself an exception.    It would be incorrect to say that Huxley and Clifford shared nothing. Like Clifford, Huxley valued the evidential way of forming a belief. There could be several ways that one can adopt as a policy for forming one’s belief, for instance, relying on the Bible, people in authority, rolling a die, and so on. Huxley himself tried to follow the evidentialistic policy and believe                                                           381 Ibid., 530 382 James 1879, 92-3.   145  accordingly.383 They also shared the social context. They lived in Victorian society where they thought their evidentialistic policy was neither secured nor welcomed but yet needed; they applied their policy to theological matters (and non-theological matters) and as a result, they were blamed and tagged as “infidels” or “atheists.”384 In other words, they had to argue that forming a belief solely on evidence is not something that should be condemned. They shared the same aim.     However, Huxley and Clifford did not further share a reactive strategy for defending their cherished policy. Here we can see how they differently formulated their evidentialistic criterion. Against those who asserted that it is morally wrong not to believe in Christian doctrines, Clifford’s position would be that on the contrary what is actually morally wrong is to believe something on non-evidential reasons. According to Clifford, believing is not private act, but a public act that has social repercussions.385 In this sense, believing can be an act that requires regulation. In other words, Clifford’s strategy amounts to immoralizing his critics’ habit of believing. In this way, the evidentialistic policy would have been shaped to be a duty to believe and framed as a universal moral duty.  On the other hand, Huxley was defensive, as opposed to proactive.386 Huxley’s focus was on de-immoralizing the evidentialistic way of forming a belief, rather than on moralizing a certain way as the only right way of believing. This point should not be regarded as an overly interpretative distinction, fooled by his rhetorical skill. If we examine why Huxley reacted in such                                                           383 Huxley 1860c, 314-5; 1866a, 158; n.d.1.  384 Huxley 1866a; 1889e; 1892c; 1892b; Madigan 2008. 385 Harvey 1969; Madigan 2008. 386 Harvey makes a similar point: “He [Huxley] does claim that his principle is both ethical and intellectual, but interestingly enough, his invocation of the terms ‘reprobation’ and ‘abomination’ is applied principally to those clerics who themselves assert that it is morally wrong not to believe certain propositions (about God, Christ, etc.)” (2013, 10, his italics).   146  a defensive manner, I think, we can conclude that he was not willing to endorse the evidentialistic policy in the form of universal moral duty for any believer.  Let’s see first how Huxley defended his application of the evidentialistic criterion. Although James complained about Huxley’s use of word “pretend,” it is important because it tells us how Huxley understood what he was not doing and what his religious opponents were doing. We need to see contexts regarding what Huxley argued against to see in what sense pretending can be a blameworthy act.   Huxley had to deal with the claim that it is condemnable not to believe Christian doctrines, which may sound an unpopular claim to make in the current secularized society or at least in the literature of epistemology or of ethics of belief. For example, Huxley introduced the claim made by John Henry Newman:   The Cleric [Newman] asserts that it is morally wrong not to believe certain propositions, whatever the results of a strict scientific investigation of the evidence of these propositions. He tells us “that religious error is, in itself, of an immoral nature.” He declares that he has prejudged certain conclusions, and looks upon those who show cause for arrest of judgment as emissaries of Satan.387  Similarly, Huxley reported the assertion of Henry Wace who blamed non-believers:   “It is, and it ought to be,” authoritatively declares this official representative of Christian ethics, “an unpleasant thing for a man to have to say plainly that he does not believe in Jesus Christ.”388                                                             387 Huxley 1889g, 313. 388 Huxley 1889e, 240.  147  Huxley acknowledged that the evidentialistic way of forming a belief was not properly appreciated, and especially by clerics who appeared to care only about its outcome, that is, whether one believes Christian doctrines or not. In “Agnosticism and Christianity,” Huxley clarified the ideas underlying the moral accusation brought against agnostics’ practice:   That which Agnostics deny and repudiate, as immoral, is the contrary doctrine, that there are propositions which men ought to believe, without logically satisfactory evidence; and that reprobation ought to attach to the profession of disbelief in such inadequately supported propositions.389   We can see here what Huxley argued against. He did not argue against the theses that there are propositions which a person may believe without logically satisfactory evidence and that reprobation ought not to attach to the profession of belief in such inadequately supported propositions, which James wanted to claim while setting up Huxley as a foil.   Huxley argued against the doctrine that there are propositions that one ought to believe without or regardless of evidence, because it is not compatible with the practice of the evidentialistic way of forming a belief, which agnostics would take as an ideal life to lead. Moreover, if it is morally wrong to believe in accordance with the result of one’s evidentialistic examination of a given proposition, it would be hardly meaningful to conduct such an examination in the first place, which he thought was critical for scientific practice. It was an important and necessary task for Huxley to eliminate morally negative connotations that had been associated with a group of people who requested reasons, decided to follow reasons only and cast doubts or believed accordingly.                                                            389 Huxley 1889g, 310.  148   The importance and necessity of the task seems to have led Huxley to a particular focal point of illustrating the history of human inquiry.390 People following the evidentialistic way of forming a belief had been oppressed. Despite intellectual persecution, evidentialistic thinkers had continued to appear and yet again, to struggle. It is worth noting how he portrayed them: “[i]t had long been difficult for reasonably honest people even to pretend to believe in the mythological fables held sacred by their forefathers.”391 They are not moral offenders who deserve a shameful label, but innocent and intellectual victims (or at least powerless minority). To Huxley, Victorian society was society where a non-evidentialistic way was regarded as a morally default way of forming one’s belief.392  How, then, did Huxley argue against the claim that there are propositions that one ought to believe without evidence? To eliminate the moral blame placed on the evidentialistic way of forming a belief, which an agnostic pursues, Huxley could have turned an agnostic’s ideal to any believer’s moral duty as Clifford did. Yet it is hard to find Huxley arguing in this way.   The first reason why Huxley thought it not morally wrong to practice the evidentialistic way of belief formation can be seen in Huxley’s reply to Wace (whose moral accusation was                                                           390 Huxley 1892c; 1895b. 391 Huxley 1895a, 527.  392 Recollecting his childhood, Huxley wrote as follows:  From dark allusions to “sceptics” and “infidels,” I became aware of the existence of people who trusted in carnal reason; who audaciously doubted that the world was made in six natural days, or that the deluge was universal; perhaps even went so far as to question the literal accuracy of the story of Eve’s temptation …. At the same time, [preachers] imagined they were discharging that most sacred duty by impressing upon my childish mind the necessity, on pain of reprobation in this world and damnation in the next, of accepting, in the strict and literal sense, every statement contained in the Protestant Bible. I was told to believe, and I did believe, that doubt about any of them was a sin, not less reprehensible than a moral delict. I suppose that, out of a thousand of my contemporaries, nine hundred, at least, had their minds systematically warped and poisoned, in the name of the God of truth, by like discipline. I am sure that, even a score of years later, those who ventured to question the exact historical accuracy of any part of the Old Testament and a fortiori of the Gospels, had to expect a pitiless shower of verbal missiles, to say nothing of the other disagreeable consequences which visit those who, in any way, run counter to that chaos of prejudices called public opinion (Huxley 1892c, 21-2).   149  quoted above). Huxley said, “A thousand times, no! It ought not to be unpleasant to say that which one honestly believes or disbelieves.”393 To Huxley, agnostics are those who only follow evidential reasons when examining the certainty of a given proposition and accordingly form a belief. Thus they would refuse to say that they believe what they find lacking evidential reasons. Considering shared moral virtues like honesty or sincerity, Huxley thought, there is nothing morally wrong about their profession of disbelief and their refusal to pretend by saying that they believe what they do not actually believe.394 Yet, as Huxley acknowledged, the moral blame ultimately targets agnostics’ very practice of the evidentialistic way of believing.   It is also found in his early writings that Huxley acknowledged the negative connotations attached to disbelievers and tried to get rid of them. For example, in 1847, Huxley wrote a letter to his future wife Henrietta Heathorn (1825-1914). Huxley appears to have worried that his views on religious doctrines might have negative impacts on their relationship which had just started.395  I have thought much of our afternoon conversation, and I am ill at ease as to the impression I may have left on your mind regarding my sentiments. If there be one fact in a man’s character rather than another, which may be taken as a key to the whole, it is the tendency of his religious speculations. … Opinion is the result of evidence. From a given amount and strength of evidence, as cause, a certain belief must, in all minds, always follow as effect. The intellect here acts passively, and is as irresponsible for its conclusion as a jury, who convict a man on the strength of certain evidence are irresponsible for their conclusion should that evidence turn out to have been unworthy of trust. … The opinion a man has, once more, neither is nor can be a matter of moral responsibility. The extent to which he deserves approbation or reprobation depends on the mode in which he has founded his opinion – and of this the Almighty search of hearts can alone be the efficient judge. May his fellowmen                                                           393 Huxley 1889e, 241, his italics. 394 Huxley 1860c; 1889e. 395 Huxley first met Heathorn in 1847, and this letter seems to be the third letter sent to her, judging from the information at website “Thomas Henry Huxley Collection” provided by Imperial College London, Records and Archives (http://www.imperial.ac.uk/recordsandarchives/huxleypapers/).    150  then form no judgment upon the point? Surely they must and will do so, .... But let them not judge him by his agreement or disagreement with their own ideas however venerable and raised the latter may appear to them – let them rather inquire whether he be truthful and earnest – or vain and talkative – whether he be one of those who would spend years of silent investigation in the faint hope of at length finding truth, or one of those who conscious of capability would rather gratify a selfish ambition by adopting and defending the first fashionable error suited to his purpose.396   Here, we can see two ideas regarding how Huxley defended evidentialistic thinkers. Like the previously introduced defense, the first idea also appeals to moral virtues, but differently. The earlier defense was that it is not morally wrong to express one’s genuine opinion or belief. In this letter, the defense is that there is nothing morally wrong about the evidentialistic way of forming one’s belief. If we are to make a moral judgment regarding one’s opinion or belief, he suggested attending not to one’s opinion but to one’s “mode in which [one] has founded [one’s] opinion.” He then implied that similar minded persons like him, evidentialistic thinkers, are “trustful and earnest” and “spend years of silent investigation.” There is nothing morally wrong about their mode of arriving at an opinion.   Huxley appears to think that if people do not judge his opinion by checking whether it is the same as that of clerics or that of infidels, but instead they judge it by considering what kind of person he has been and what kind of mode his inquiry has taken, they could not find something immoral about the practice leading him to an opinion that he has. He brought in one’s moral character and intentions that would indicate one’s mode of arriving at an opinion, because, we are not usually in position of being an “efficient judge” who can directly evaluate one’s mode. He noted that we should make a moral judgment regarding one’s opinion in this way. This suggestion is based on another point that he made to defend the evidentialistic thinking: a moral                                                           396 Huxley 1847. See also Huxley 1860c; 1880b; 1892c.  151  evaluation should be made concerning one’s mode of arriving at an opinion, not one’s holding an opinion. This point deserves further attention.    According to Huxley, one “neither is nor can be” held morally responsible for holding a certain opinion, because holding an opinion is not something that one can control but a result caused by learning evidence.397 This point is particularly interesting because he appealed to an idea that is similar to the view now labeled “doxastic involuntarism” which is often invoked, ironically, to criticize Clifford’s ethics of belief. This type of involuntarism is a position on believing, according to which we cannot make ourselves believe. As William P. Alston puts it, “beliefs are items we find ourselves with, not items we choose to have.”398 Critics of Clifford argue, based on the so-called “‘ought’ implies ‘can’ principle,” that the very idea of ethics of belief makes no sense because it prescribes a moral rule over what we cannot control, that is, what to believe.399   Huxley’s rationale for the claim that one is not morally responsible for holding one’s opinion per se is that one comes to hold it involuntarily or, as he puts it, the intellect “acts passively” in the presence of evidence. He also held a view that captures the ought-implies-can principle. For example, regarding an idea about “a prohibitory duty upon philosophical speculation,” he wrote:   In this case, however, as in some others, those who lay down the law seem to forget that a wise legislator will consider, not merely whether his proposed enactment is                                                           397 Huxley expressed a similar idea as follows: “Every belief is the product of two factors: the first is the state of the mind to which the evidence in favour of that belief is presented; and the second is the logical cogency of the evidence itself” (1880b, 230).  398 Alston 1996, 7, his italics.  399 Chignell 2016.  152  desirable, but whether obedience to it is possible. For, if the latter question is answered negatively, the former is surely hardly worth debate.400  Similarly, he wrote fifteen years later:  Suppose there is an immutable eternal moral law for the angels: what is that to us who are not angels and do not live under heavenly conditions? Surely a farmer, who laid down rules for his horses and expected his pigs to obey them, would be a little unreasonable!401   Huxley thought that “morality is in its very essence a rule for the guidance of his [man’s] conduct,” and regarding the suggestion of a moral rule that prescribes something an agent cannot do, he confessed that he was “at a loss to understand.”402     Huxley’s version of doxastic involuntarism can be then understood as follows. We cannot entirely control our believing, but as he suggested, we can indirectly influence our believing by exposing or not exposing ourselves to evidence. Thus, for this reason, we cannot be held responsible for holding an opinion but we can be held responsible for our manner with which we come to believe something. Based on this idea, he argued that there is nothing morally blameworthy in the manner in which agnostics come to have a belief or opinion. Although each agnostic would lead a different life and have various characters, agnostics’ manner in principle (or “true” agnostics’ manner) is to adopt the evidentialistic policy for arriving at their belief. In hopes of finding knowledge, taking into consideration only factors that indicate whether a given proposition can count as knowledge is far from being morally condemnable.                                                            400 Huxley 1879a, 288.  401 Huxley 1895b in Peterson 1932, 320-1.   402 Ibid., 320.   153   Yet Huxley’s version has an interesting qualification: only believing on evidential reasons is involuntary.403 The idea is that when presented with evidential reasons, an opinion or belief emerges in our mind. Here he is pointing to the force of evidence and logic which is not something that is easily resistible. On the other hand, our attending to prudential reasons does not get us to believe proposition p involuntarily. Acknowledging the usefulness of believing p does not generate belief p in our mind whereas acknowledging the amount and strength of evidence for p does. In this sense, he said that we pretend that p when attending to prudential reasons while disregarding evidential reasons, whereas he did not say that we pretend that ~p when attending to evidential reasons while disregarding prudential reasons. Thus, to Huxley, holding a belief based on evidential reasons is the state in which we have been placed involuntarily, and is not subject to moral condemnation; our act of exposing ourselves to evidential reasons is a voluntary act, but it is hardly morally reproachable. On the other hand, holding a belief based on prudential reasons is the state in which we have chosen to be, often disregarding a belief generated by the lack or presence of evidential reasons, and in this sense, the state involves pretension; our act of paying attention to prudential reasons is a voluntary act, and he found it intellectually blameworthy when it is done to make a claim to knowledge. In short, the restricted version of doxastic involuntarism enabled Huxley to de-immoralize one’s evidentialistic way of belief formation and make room for epistemically blaming one’s non-evidentialistic way of belief formation.    It is then more plausible to conclude that Huxley did not hold the strong thesis that has been credited to Clifford. Huxley’s defensive stance shows that he focused on de-immoralizing the evidentialistic way of forming a belief, and his rationale behind his defense undermines the interpretation that Huxley and Clifford endorsed the same moral imperative on believing. Huxley                                                           403 Huxley 1869a.   154  could not see any moral wrongdoing in those who adopt the evidentialistic policy and believe accordingly: they are just people who have earnestly struggled to find knowledge and expressed their genuine beliefs. Huxley’s views discussed so far do not establish the case that he made the claim that the evidentialistic way of belief formation is the only morally right way.   We can find another set of reasons for concluding that Huxley’s position is not the same as Clifford’s imperative from Huxley’s stance on ordinary beliefs and on having a faith. Clifford’s imperative does not allow exceptions, but Huxley was not willing to claim that it is always morally wrong to believe anything on non-evidential reasons or on faith. Huxley appears to have granted multiple manners of arriving at a belief to lead one’s own life.   First of all, Huxley admitted that a belief without evidential reasons can have practical values in ordinary life (of course, pernicious effects as well).404 It can give people some emotional comforts such as encouragement or consolation, and in this sense, he noted, “even the worst form of Christianity” can be said to have given a “great practical advantage to them [believers].”405 However, one might point out, admitting practical values does not necessarily mean that he rejected Clifford’s thesis. Clifford can also admit practical values of believing on non-evidential reasons while thinking it morally and intellectually wrong. This is a fair point, but Huxley did not make a moral judgement on one’s having a groundless belief as long as context concerns one’s way of leading a life, not one’s claiming such a belief to be certain or knowledge.   Second, Huxley was well aware that some beliefs that we hold lack evidential reasons such as a belief that we rely on as a “starting point” for reasoning.406 He would have agreed on                                                           404 Huxley 1860c; 1889e. 405 Huxley 1889e, 241. 406 Huxley 1878h;1889e.    155  James’s remark that “[t]here are, then, cases where a fact cannot come at all unless a preliminary faith exists in its coming,” although James assumed that he would not.407 Huxley also noted that we are “often obliged, by the pressure of events, to act upon very bad evidence.”408 This shows that Huxley permitted exceptions to Clifford’s imperative.   Last, Huxley did not claim that having faith in itself is morally wrong. He had no quarrel with individuals living by faith, and often described himself as having faith. In fact, this has led critics to argue that he had his own article of faith, that is, faith in science. Yet interestingly, he openly and shamelessly expressed his faith.409 His stance on faith has been unappreciated and I will return to it in next chapter where I examine the criticism that Huxley’s faith in science is on par with his religious targets’ faith. For now, we just need to note that Clifford’s imperative is too strong for Huxley to fully endorse.  To sum up, Huxley observed that some people (whom he called “agnostics”) adopt the evidentialistic policy for forming their belief and tried to be one of them, but he did not formulate this policy as a universal moral imperative that any believer ought to obey in every case. Huxley would not be happy with the interpretation that his agnosticism amounts to evidentialism concerning what to believe; he tried to argue against the notion that there is something that we ought to believe. Now we can focus on my suggestion about the imperatives that he wanted to have in his agnosticism. In the previous section, I briefly introduced a duty to examine evidential reasons for making a claim to be certain or knowledge, and a duty to accept the result of an examination. In the following, I will discuss them in detail.                                                            407 James 1896, 25.  408 Huxley 1889e, 243. Also Huxley 1888b.  409 Huxley 1887e; 1887g, 553; 1889e; 1892c.  156  5.4. How to follow the agnostic principle   The agnostic principle sounds like an instruction only for those who are willing to be an agnostic. Yet it also sounds like a universal principle, especially when we consider that it was the basis that Huxley had for criticizing clerics. If his principle was only meant to be a rule for insiders, his agnosticism would not work as a criticism against those who are not interested in becoming an agnostic such as Victorian clerics, materialists, or all sorts of “-ists.” Huxley’s writings show that he intended his agnosticism to be universally applicable. I suggest that the agnostic principle was his ethics of knowing. The aim of this section is to clarify duties implied in his agnosticism. Clifford and Huxley promoted the evidentialistic policy – taking into consideration evidential reasons alone. Clifford urged people to adopt this policy whenever they form a belief or whenever they judge what to believe. It was turned into a universal and unqualified imperative for any believer. In contrast, I argue, Huxley urged people to adopt this policy whenever they intend to have a belief about the certainty of a proposition or whenever they judge what to accept as certain or knowledge. Huxley turned the policy into a universal and unqualified imperative not for any believer. Instead of claiming that his opponents believed Christian doctrines in a morally wrong way, Huxley claimed that their cherished doctrines were not certain as they insisted or that their assurance was unfounded.410 Clifford characterized the evidentialistic policy as an ethics of believing, whereas Huxley characterized it as an ethics of knowing.   What would it mean to have the evidentialistic policy in a guide to knowledge as a principle? I believe that Huxley wanted to highlight the following universal duties by proposing his agnostic principle. First, to learn about the knowledge status of a given proposition, one ought                                                           410 Huxley 1878h; 1886e; 1889e; 1889g; 1890d.  157  to attend to evidential reasons and evidential reasons only. Afterwards, second, one ought to accept the result of an evaluation: in Huxley’s expressions, “professing to accept for proof evidence which we are well aware is inadequate” or “willfully shutting our eyes and our ears to facts which militate against this or that comfortable hypothesis” should not be done.411 I regard the second duty as more important in understanding his agnostic principle, because he had a quite specific picture about what we should do once we expose ourselves to evidential reasons, as I will show shortly.   Let’s start with the first duty. It requires us to take into consideration evidential reasons and evidential reasons only when we judge the certainty or knowledge status of a given proposition. This duty might be seen as uncontroversial because it appears to prescribe that we should strictly employ the concept of knowing or knowledge. However, as shown, Huxley’s agnosticism includes his view of evidential reasons, which might be contentious. This duty, then, turns out to be that we should commit ourselves to his account of legitimate evidence.    According to Huxley, evidence is a trustworthy perception, and a trustworthy perception can function as evidence when it supports the occurrence of a particular event or phenomenon at issue, and not others. He grouped evidence into three categories: direct observational or experimental evidence, (human) testimonial evidence, and circumstantial evidence.412 The observational evidence is evidence that can verify a given statement by showing that we have experienced a (expected) set of impressions, but it is not always available. Especially when we deal with the past, we have to rely on the last two types of evidence. Using Huxley’s example, a                                                           411 Huxley 1892c, 54; 1886e; 1890d. 412 Huxley 1876c. Huxley considered an experiment to be a special kind of observation where we know conditions related to an event to be observed; thus he called experiment “artificial observation” (1854a, 52; also 1880a).    158  witness’s report on a murder is testimonial evidence whereas the shape or character of injury is circumstantial evidence.413 Although “[w]e are very much in the habit of considering circumstantial evidence as of less value than testimonial evidence,” Huxley argued, “it must not be forgotten that, in many cases, circumstantial is quite as conclusive as testimonial evidence, and that not unfrequently, it is a great deal weightier than testimonial evidence,” because circumstantial evidence is less subject to falsification and various kinds of doubt.414 We may understand this point to mean that circumstantial evidence is less subject to, in contemporary terms, respectively, rebuttal defeaters and undercutting defeaters.   As implied in the quote above, we also need to evaluate the quality of alleged evidence. Evidence is after all a perception. Perceptions are the only materials available to us, but not all perceptions are trustworthy. Huxley wrote:   “I tell you I saw it myself,” is the so-thought conclusive assertion with which many a controversy is abruptly ended. Commonly those who make this assertion think that after it nothing remains to be urged; and they are astonished at the unreasonableness of those who still withhold their belief. … [Y]et they cannot imagine that their own perceptions have been vitiated by influences like those which vitiated the perception of others. Or, to put the thing more charitably and perhaps more truly, they forget that such vitiations are constantly occurring.415  According to Huxley, they do have a perception, but not thereby having evidence. If we want evidence, we need to check, for example by trying experiencing again, whether there is any defeater that “vitiates” the quality of a perception at issue, because it can count as evidence only                                                           413 Huxley 1876c. 414 Ibid., 57. 415 Huxley [Anon.] 1853, 162. See also Huxley 1889e.   159  when it is indeed trustworthy. That is, in some cases, alleged evidence itself becomes a claim that calls for independent evidence.  A perception suggested as evidence needs to be examined from another aspect. According to Huxley, our perception may be either distorted or incomplete, because our perceptions are generated in various degrees with the aid of a “pre-condition.”416 Huxley noticed that people tended to consider their “evidence” to report what they were sensing, but sometimes their evidence turned out to be one of interpretations of what they were sensing. In this case, alleged evidence may not have evidential power as intended. We should keep in mind that a “sensible phenomenon” can be reported in different ways, and thus we need to examine whether an interpretative framework incorporated into given evidence is credible.417 In this regard, Huxley introduced Johann Wolfgang von Goethe’s aphorism on facts as theories. Huxley understood it to imply that “[o]ur faculties are liable to report falsely from two opposite causes – the presence of hypothesis, and the absence of hypothesis.”418 The first cause tends to lead to a distorted perception and the second to an incomplete perception.419 From this perspective, he approached witnesses in miracles or some magical powers like psychokinesis. They may be sincere and well-intended, but they may not know that other ways may have produced the event.420    When are we advised to examine whether a given perception indeed has evidential power? Huxley introduced a “canon,” which he considered to capture Hume’s principle and common sense reasoning: “the more a statement of fact conflicts with previous experience, the                                                           416 Huxley [Anon.]1853, 167. 417 Huxley 1889e, 262, n. 9. 418 Huxley [Anon.] 1853, 162. He introduced Goethe’s “Alles factische ist schon Theorie” in his 1889e (262, n. 9).    419 Huxley [Anon.] 1853, 167. 420 Huxley [Anon.] 1853; 1878h; 1889e; 1889a.  160  more complete must be the evidence which is to justify us in believing it.”421 He illustrated the cannon as follows:   If a man tells me he saw a piebald horse in Piccadilly, I believe him without hesitation. The thing itself is likely enough, and there is no imaginable motive for his deceiving me. But if the same person tells me he observed a zebra there, I might hesitate a little about accepting his testimony, unless I were well satisfied, not only as to his previous acquaintance with zebras, but as to his powers and opportunities of observation in the present case. If, however, my informant assured me that he beheld a centaur trotting down that famous thoroughfare, I should emphatically decline to credit his statement; and this even if he were the most saintly of men and ready to suffer martyrdom in support of his belief. In such a case, I could, of course, entertain no doubt of the good faith of the witness; it would be only his competency, which unfortunately has very little to do with good faith, or intensity of conviction, which I should presume to call in question.422  Huxley explained that his strong hesitation to credit the testimony on a centaur was not meant to point to an impossibility. As he noted, if a centaur is shown to exist, anatomists and physiologists would have something to work on and might need to revise established generalizations. The hesitation means “to cast the entire burthen of proof, that centaurs exist, on the shoulders of those who ask [one] to believe the statement.”423 The burden can be placed on them, because to Huxley, a less conflict with previous experience and repeated experience suggests a wider range of and higher degree of being verified.424    In many places, we can see Huxley applying this canon. Many of his critical discussions of Christian doctrines developed in this manner.425 Some clerics claimed that their belief was                                                           421 Huxley 1878h, 158. Also, Huxley 1887c; 1889e, 226.  422 Huxley 1878h, 158-9. 423 Ibid., 160. 424 Ibid.; 1887c.  425 Huxley 1878h; 1886e; 1889e; 1890d; 1892c.  161  certain; based on the canon, Huxley “hesitated” to accept their beliefs as certain and instead asked for evidence; he examined and found that alleged evidence had in fact no evidential value; accordingly, he concluded that their claim was not certain. This process shows that Huxley did not dismiss their claims in an a priori manner. He might be seen as biased against clerics’ claims, but he was applying the canon, which he considered to have epistemological grounds. In order for the testimonial evidence that an adult man walked on water to count as evidence, more demanding scrutiny is required than the report, say, that an adult man went down.   Complying with the agnostic principle involves taking into account evidential reasons and not other types of reasons, when our concern or context is the certainty of a given proposition. However, Huxley did not stop there. Following the agnostic principle also means admitting the result of an evaluation. This, to Huxley, involves exhibiting two kinds of attitude. First, we should be honest. Second, we should take a proper epistemic attitude toward a given proposition in accordance with a given outcome. Each point will be clarified in the following.     Consider that we examine evidential reasons offered for a given proposition, and suppose that the reasons do not show the certainty of the proposition. According to the agnostic principle, we should accept this result as turned out. Yet what does honesty have to do with this acceptance? According to Huxley, a competent judge will be able to see the value of given evidence, and a knowledge seeker will note that whether a proposition at issue counts as certain is determined in the light of given evidence and independently from one’s wish. Once evaluation is done, admitting the result is not an act that requires voluntary efforts or will, as discussed in the previous section. When we expose ourselves to evidential reasons, through simple or complex valuation of evidence depending on a case, we are led (or “compelled,” as Huxley sometimes  162  said) to have a belief or opinion concerning whether a proposition at issue is certain or not.426 What we need to do is to be honest about a generated belief or opinion and not to forge one. Thus, Huxley’s agnostic principle says: do not pretend that one’s belief is certain when it turns out not to be certain; do not pretend that others’ belief is uncertain when it turns out to be certain. In this way, Huxley connected his agnosticism with the virtue of honesty.   Honesty has been associated with agnosticism, but as far as Huxley’s agnosticism is concerned, it is more precise to understand honesty as being honest with logical and evidential forces that compel us to have a belief or opinion, rather than as being honest about the fact that there is something that we are not capable of knowing. We have this kind of honest attitude when we are in a state of the lack of pretension or intention to misrepresent, as opposed to a humble state with feelings of inferiority in front of something of which we are ignorant, or, say, the “Unknowable.” To Huxley, “intellectual honesty means absolute submission to evidence.”427  In addition to the attitude of being honest, admitting the result of an evidential evaluation also involves an epistemic attitude. There are three modes of being honest in accordance with possible results of an examination on evidence. First, a proposition at issue falls under the boundary of knowledge, and it has been demonstrated. Second, a proposition at issue falls under the boundary, but neither it nor its opposite has been demonstrated. Third, a proposition at issue does not fall under the boundary; it is not demonstrable because no evidence can be offered.   Consider the first two cases that concern our attitude toward propositions within the boundary of knowledge. Huxley was more generous about counting a proposition as being within the boundary of knowledge than his opponents would like to assume. All we can know is, in the                                                           426 Huxley 1893-4 (Vol. V), xvi.  427 Huxley 1876b, 7.  163  end, facts of phenomena, and this sets the boundary of our knowledge. In other words, except statements that purport to be about the “real” world as opposed to the phenomenal world, any statement was counted by him as being placed within the boundary of knowledge. For example, proponents of materialism according to which reality consists of matter and force transgress the boundary of knowledge, because they wish to deal with absolute and eternal entities. On the other hand, if one intends to make a claim about material objects as constituents of the phenomenal world, this claim falls under the boundary of knowledge; he treated such a claim as a hypothesis. Thus, for the same reason, he did not dismiss statements about immaterial or spiritual phenomena as transgressing the boundary of knowledge.428 Many of Christian doctrines, he found, concern the phenomenal world.429    With respect to any statement within the boundary of knowledge, according to Huxley, we have to request evidence and evaluate the value of offered evidence, as explained above. When either p or ~p has been shown to be justified by evidence, we accept it as knowledge or certain. He noted that not all his contemporaries, including clerics and even scientists, exhibited this attitude when it was required.430   An interesting case is when neither p nor ~p is certain, because Huxley recommended a specific stance that has become a signature stance of agnosticism. As mentioned, to conclude that proposition p is certain and deserves the status of knowledge, there must be evidential reasons. In other words, justification is necessary. A possibility of p being true or the lack of evidence for ~p is not sufficient enough to count p as knowledge.431 If we have this case, according to Huxley, we                                                           428 However, Huxley strongly refused to engaged in an inquiry on spiritualism (for example, see 1871d).    429 Huxley 1878h; 1886e; 1888b; 1889e; 1892a.  430 Huxley 1863d; 1878h;1887c; 1887g; 1892a. 431 Huxley 1876c; 1878h; 1870e.  164  should say “I don’t know.” This profession of ignorance differs from the profession of ignorance regarding something beyond our boundary of knowledge. In the case where we do not have positive evidence for either p or ~p, which may be found later, the profession of ignorance means the suspension of judgment. We neither affirm nor negate p, but we can assume that either p or ~p can be supported by evidence. In short, we do not know whether p or ~p is the case. Huxley tended not to take or recommend the suspension of judgment regarding statements outside the knowledge boundary; there is no judgement to be suspended. Huxley appears to have reserved this stance only for statements within the boundary. Examples of such statements that Huxley mentioned were ones about whether evolution occurred, whether evil spirit was moved to pigs, whether a large amount of people were fed with a small amount of food, the story about Regal Rome, or legendary figures like King Author.432 This stance, suspension of judgement, deserves further attention.   Some might think that Huxley’s conception of the suspension of judgment is similar to that of the stance recommended by ancient skeptics for cases that competing claims are equally convincing. However, Huxley’s conception involves the situation of claims being equally not conclusive. Since his agnostic principle requires the presence of positive evidence to make a claim to knowledge, when a position has not been supported by evidence, the suspension of judgement is required. As he said, when “none is worthy of belief,” “our condition of mind should be that suspension of judgment which is so difficult to all but trained intellects.”433 His standard for not suspending judgment or for withdrawing the suspension of judgement would be comparable to the standard of proof used in criminal cases, beyond reasonable doubt, not the one                                                           432 Huxley 1866a; 1889e; 1890d. 433 Huxley 1876c, 56; 1870e.  165  in civil cases. In this manner, he mentioned a “verdict of ‘not proven’” with respect to historical clams of Christianity.434 To Huxley, the suspension of judgment is not, strictly speaking, a stance led by relative comparison of how compelling each position is.   Another point that distinguishes Huxley’s conception of the suspension of judgment from that of ancient skeptics is that to him the suspension of judgment is not the final stage to stay.435 Ancient skeptics claimed that by suspending our judgment, we would be able to be in tranquil mind.436 Although Huxley expressed worries about people’s impatience with the “most wholesome state of mind – suspended judgment,” it is a temporal and intermediate stage.437  Suspending judgement about a given proposition means to him that if we continue to pursue knowledge and search for evidence, we would be able to arrive at our judgment about which one is the case. Thus, for example, he explained why some people had been “unwilling to accept evolution” before Darwin as follows: there had been no better grounds than those offered by Jean-Baptiste Lamarck and Robert Chambers, and due to the lack of evidence, those people “therefore [had] preferred to suspend their judgment on the question.”438      The point that the suspension of judgment is not the final stage leads to another point that it involves “active doubt,” which was Huxley’s translation of Goethe’s “Thätige Skepsis.” “It is,” Huxley explained, “doubt which so loves truth that it neither dares rest in doubting, nor extinguish itself by unjustified belief.”439 Huxley’s notion of doubt does not necessarily refer to a                                                           434 Huxley 1889g, 311; 1893-4 (Vol. V), xxxi; also 1884a, 5.  435 Huxley 1893-4 (Vol. V).   436 Vogt 2014. 437 Huxley 1890g, 272. Also Huxley 1870e. 438 Huxley 1878d, 222. 439 Huxley 1859f, 20. Also Huxley 1887g; 1892b.   166  certain psychological state, although being doubtful or skeptical is often characterized as a psychological state. To illustrate what Huxley had in mind, it would be useful to introduce David Fate Norton’s discussion of doubt. According to Norton, doubt is not always accompanied with a particular psychological state or disposition (e.g. being uneasy or hesitating) and it may also refer to “intellectual activity” like challenging and cautionary behaviors.440 Thus Norton distinguishes two kinds of doubt: “active doubt” understood in terms of activity and “affective doubt” in terms of the state of mind.441 Huxley’s notion of doubt, especially when he introduced Goethe’s “Thätige Skepsis,” is closer to Norton’s “active doubt” than to “affective doubt.”   Huxley’s notion that the suspension of judgment and active doubt are two sides of the same coin helps us to understand how he could think that suspending judgment and favoring one position over others are compatible. While suspending judgment with respect to the certainty of each of competing propositions, one may prefer one to others based on, for example, relative probabilities.442      For instance, in that way, we can understand Huxley’s stance toward Darwin’s theory of natural selection.443 Huxley regarded Darwin’s theory as the most compelling and “ingenious” hypothesis, yet never accepted it as an established fact, knowledge.444 He appreciated the explanatory power of Darwin’s theory, but its relative superiority regarding explanatory breadth does not establish its status as knowledge. There is a conceptual gap between “hypothesis” and “fact”; Huxley thought that “the admission of a state of mind intermediate between knowledge                                                           440 Norton 1982, 288. 441 Ibid., 284. 442 Huxley 1889g.  443 For the discussions of Huxley’s defense of Darwin, see Bartholomew 1975; Bowler 1997; Ruse 1997; di Gregorio 1981; 1984; Lyons 1999, 2009. 444 Huxley 1859f, 19; 1860a; 1860b, 391; 1862a; 1862b; 1863d; 1891d; 1892d.    167  and no-knowledge is fatal to all clear thought.”445 Without “experimental proof,” he could not rank Darwin’s hypothesis of natural selection as a fact. In 1892, three years before his death, Huxley wrote in a letter as follows:   In a review of Darwin’s Origin of Species published in the Westminster for 1860 (Lay Sermons, pp. 323-24), you will see that I insisted on the logical incompleteness of the theory so long as it was not backed by experimental proof that the cause assumed was competent to produce all the effects required. … In fact, Darwin used to reproach me sometimes for my pertinacious insistence on the need of experimental verification.446  Mainly around early 1860s, Huxley and Darwin had arguments over natural selection. Huxley’s concern was that all aspects of the causal sufficiency of selective breeding – “whether artificial or natural” – had not been verified.447 Since natural selection was proposed by Darwin as the cause of the generation of a new species, Huxley expected natural selection to account for characteristics that mark different species (as opposed to varieties). One is morphological divergence among species, and Huxley thought that it had been verified that natural selection is the cause of such morphological divergence.  Another divergence among species that Huxley brought to Darwin as an issue concerns hybrid infertility: two distinctive species cannot breed with each other or their offspring is sterile. Huxley’s point was that it had not been verified that natural selection is also the cause of such physiological divergence. Huxley insisted that Darwin needed experimental proof, because it was “the weak point of his case from the point of view of scientific logic.”448 In other words, Huxley                                                           445 Huxley 1895a, 535; 1879c. 446 Huxley 1892b, 204, his brackets.  447 Huxley 1860a, 74.  448 Huxley 1891d, 203.  168  claimed, we need to have a memory of natural or artificial selection producing that effect. As discussed, a complex idea is either an imagination, an expectation or a memory, and we can turn our imagination or expectation into our memory by experiencing a corresponding complex impression. To complete “scientific logic,” Huxley found it necessary to have experience of an (scientifically well) expected causal process, and one way was to get “virtually infertile breeds from a common stock.”449 Darwin complained to Joseph Dalton Hooker that Huxley wanted to see natural selection “in action.”450 In addition, Darwin initially (and finally) saw hybrid infertility as a by-product of morphological differences, and pointed out that infertility varies in degree.451 In other words, hybrid infertility is not a significant characteristic that marks species as different. On the other hand, as Huxley saw it, it constitutes “physiological species” and any theory of the origin of species, to be perfect, must be able to account for physiological divergence.452 In short, for Huxley, physiological divergence was an important phenomenon to be explained, and no one had experienced, in some way, selective breeding leading to hybrid infertility. Huxley could not but say that Darwin’s natural selection was a hypothesis about the origin of species.453  However, Huxley expected that experimental proof for the physiological effects of natural selection will be done in future and was supportive of Darwin and others who engaged in the project of producing the proof.454 Thus Huxley could consider Darwin’s natural selection to be                                                           449 Ibid. 450 Darwin 1860; also quoted in Lyons 1999, 241. See also Darwin 1862; Huxley 1864a, 92l. 451 Darwin 1975[1856-8], Ch. IX; Lyons 1999, Ch. 7; Harvey 2003. 452 Huxley 1860a; 1863a; 1891b. 453 Huxley’s view of natural selection can be best summarized by his own sentence: “He [Darwin] has shown that selective breeding is a vera causa for morphological species; he has not yet shown it a vera causa for physiological species” (1863a, 344, his italics). 454 Huxley 1860a; 1863a; 1891d.  169  the most promising hypothesis and urge scientists to take it as a research hypothesis.455 In other words, he suspended his judgment and actively doubted. This stance would not be taken if one thinks that natural selection cannot possibly be the cause of the origin of species.   Let’s move on to the case of statements outside the boundary of knowledge. This kind of statement is not demonstrable because the boundary has been drawn in accordance with human limits. We have no way of justifying our feeling certain and thus turning it into being certain. Even employing Descartes’s method of doubt with the highest standard cannot settle an issue of this kind. If we understand an idea about a non-phenomenal world as its proponents intend, the idea is, to Huxley, neither a memory nor an expectation, but an imagination. What should we do about such an idea?  The agnostic principle, first of all, demands the profession of ignorance regarding the certainty of a proposition outside the boundary of knowledge. The state of being ignorant in this case differs from the state of suspending judgment. An agnostic may say “I don’t know,” but strictly speaking, it has the connotation of “I can’t know, and neither can you.” As the two cases of the profession of ignorance are different in meaning, accompanied attitudes are different. In the case of non-demonstrated proposition within the boundary of knowledge, we are temporarily ignorant and we are capable of knowing. We thus should suspend our judgement for a while and practice active doubt to move forward to our state of knowing from our state of being ignorant. On the other hand, in the case of non-demonstrable proposition, we are not capable of knowing. Intellectual activity like active doubt is pointless in this case. According to Huxley, although                                                           455 Huxley 1860a, 74; 1860b; 1887g; 1894b, 3.   170  some “short-sighted” people forget about the boundary of knowledge, “they must be content with imagination, with hope, and with ignorance.”456 To clarify the attitude associated with the confession of ignorance in this case, first note examples that Huxley had in mind. Although it seems to have been expected that he would have considered his contemporary clerics’ claims to be outside the boundary of knowledge, he did not. What Huxley counted as going beyond our boundary of knowledge was claims mainly put forward by “a priori philosophers,” “a priori speculators” or “pure metaphysicians.”457 For example, one of such issues is whether our mental contents correspond to reality.458 The mental picture may resemble or may not resemble what we call “reality” or what are regarded as the cause of the picture, as metaphysical thinkers have variously claimed, but we are not capable of knowing about this issue.459     We can also see an attitude that an agnostic would take toward non-demonstrable propositions by considering Huxley’s own attitude toward unknowable issues and his criticism of a priori philosophers. Huxley did not attempt to examine relative plausibility of each of competing propositions as he did with disputed propositions within the boundary of knowledge; he simply listed different views as if he introduced diverse stories or “imaginations” about one issue. Regarding the issue of our mental picture, Huxley wrote as follows:                                                                456 Huxley 1885b, 161.  457 Huxley singled out Plato who had used imagination to far extent (1860c; 1863c; 1878h; 1888a, 1893-4 (Vol. VI); 1895a; 1895b).  458 Huxley 1870c; 1878h. Another “insoluble” problem that Huxley briefly mentioned is whether a frog has consciousness, because we can attain “no positive evidence” for and against it; his reason seems to be that we cannot be a frog (1870a, 7).      459 Huxley does not appear to have considered non-demonstrable propositions to lack a truth value.     171  This picture may be a true likeness – though how this can be is inconceivable; or it may have no more resemblance to its cause than one of Bach’s fugues has to the person who is playing it; or than a piece of poetry has to the mouth and lips of a reciter. It is enough for all the practical purposes of human existence if we find that our trust in the representations of consciousness is verified by results; ….460   Huxley seems to suggest no epistemic stance toward such imaginations.461 Yet, sometimes he expressed more than epistemic indifference: “Why trouble ourselves about matters of which, however important they may be, we do know nothing, and can know nothing?”462   The agnostic principle demands something more than the profession of ignorance, as some people have attached an additional connotation to the notion of the unknowable. For example, it has been sometimes told that something “beyond” or “above” (not simply “outside,” as Huxley loved to point out) our knowledge is more important or meaningful to our life than what we know and can know; or, the “fact” that we cannot know something proves the unknowable to be a being. It seems that the young Huxley had been rather tolerant of such a connotation attached to the notion of our limits or ignorance, if it had no explicit epistemological import. But the later Huxley seems not. He declared that he did not care about the “unknowable” and he did not even want to waste the capital U.463 Mystification or idolization, which is usually intended to make a further point, is done thanks to an idea of the certain existence of a thing; it is not free from making a claim to know. We identify the realm of our ignorance to learn what we can know and what we can use as evidence, not to affirm the existence of the unknowable and further to invoke some mystic access to it. According to Huxley, the term “agnostic” was for him                                                           460 Huxley 1870c, 178. 461 Huxley 1870c; 1892c. 462 Huxley 1868a, 162-3. 463 Huxley 1889g; 1895a. See also Huxley 1890b.    172  “a fit antithesis” to “the gnostics being those ancient heretics.”464 Huxley found himself strongly disagreeing with Spencer over how to understand what we are incapable of knowing, although Huxley could not express the difference in his early days.465 Regarding the origin of the term “agnostic,” in the letter part of which I just quoted, Huxley wrote:  The term “agnostic” was not suggested by the paragraph in the Acts of the Apostles in which Paul speaks of an inscription to the unknown God (agnostic theo). It is obvious that the author of this inscription was a theist – I may say an anxious theist – who desired not to offend any God not known to him by ignoring the existence of such a deity. The person who erected the altar was therefore in the same position as those philosophers who in modern times have brought about the apotheosis of ignorance under the name of the “Absolute” or its equivalent.466   To Huxley, the confession of ignorance with an “anxious” state (of desiring not to miss any truth unknown, in this case) is not the confession of ignorance that his agnostics would make. Huxley’s focus was always put on what we can know and how we can know, rather than what we cannot know.   How can we then formulate a required attitude toward statements outside the boundary of knowledge? “[A] consistent agnostic,” according to Huxley, “might let his imagination wander freely among such possibilities and remain perfectly true to his principles, so long as he did not mistake his dreams for knowledge, or abuse other people because they dreamed dreams of other kind or refused to dream at all.”467 Huxley made a similar point repeatedly, but did not present it in the form of thesis. I suggest, by saying that we cannot know p, Huxley would mean to declare                                                           464 Huxley 1889h in Lightman 1987, 13 465 Huxley 1863c; 1889i; 1895b. See also Lightman’s account which I briefly introduced in footnote 84 of Chapter 2.  466 Huxley 1889h in Lightman 1987, 13 467 Huxley 1895b in Paterson 1932, 319.  173  that p is neither justified nor justificatory. Dreaming has nothing to do with justification. Thus, we should refrain from thinking that a claim transgressing the boundary can justify our beliefs about a phenomenal world. By removing epistemological relevance, Huxley could undercut any intention or motivation to invoke the unknowable. Huxley noticed that many of his contemporaries tended to regard the realm of ignorance as a source for justification as if their assumption about the realm is certain, although not all made their intention explicit. To Huxley, “[building] castles in the air” makes no contribution to our activity of justification.468 By proposing agnosticism, i.e., a guide to knowledge consisting of an account of legitimate evidence and an ethics of knowing, I suggest, Huxley was depriving alleged evidential power of the unknowable or unknown and empowering evidential value of the known by experience.      5.5. Agnosticism in the tradition of modern critical philosophy  Another place to see the nature and structure of agnosticism is Huxley’s discussions of critical philosophy or “philosophical criticism,” as he sometimes called it, because he located agnosticism within this tradition.469 His understanding of critical philosophy as an epistemological struggle to overcome skepticism on one hand and dogmatism on the other hand helps us to see why a guide to knowledge is a better frame to understand his agnosticism because we can see what he intended to do with it. Furthermore, Huxley’s discussions of characteristics shared by modern critical philosophies illuminate what he wanted to incorporate into his version of critical philosophy, agnosticism.                                                            468 Huxley 1893-4 (Vol. VI), viii, x. Also Huxley 1894g. 469 Huxley 1878h, 65.   174  Huxley valued critical philosophy because he considered it to be a philosophical attempt to avoid both (Pyrrhonian or excessive) skepticism and (Platonic or metaphysical) dogmatism. Skeptics made a clever or “shallow” destructive point to claim that we can have no knowledge; dogmatists insisted that we can arrive at knowledge only by employing “Reason” (whatever it means), while discrediting experience.470 Skepticism left us with no knowledge for us who cannot but live based on some knowledge, whereas dogmatism left us with various speculations or imaginations that confuse us who had learned from experience in everyday life.  Huxley told us that he had wanted to distinguish his position from other sorts of “-ists” and thus had come up with the label, “agnostics,” but philosophical foes that he set up in a rather general level were skeptics and dogmatists. His discussions of agnosticism are often accompanied with expressing dissatisfaction at skepticism and dogmatism, which varies from a sarcastic caricature to a serious criticism. This set up, agnosticism contrasted to skepticism and dogmatism, was his recurrent theme.471     Huxley found that modern philosophers had shared his epistemological concern. They had examined a way to find something certain or knowledge, while trying to overcome both skepticism and dogmatism. Huxley wrote as follows:     The modern spirit is not the spirit “which always denies,” delighting only in destruction; still less is it that which builds castles in the air rather than not construct; it is that spirit which works and will work “without haste and without rest,” gathering harvest after harvest of truth into its barns and devouring error with unquenchable fire.472                                                             470 Huxley 1878h; 1888a; 1895b in Paterson 1932, 316-7.   471 Huxley 1887g; 1985b; 1893-4 (Vol. VI). 472 Huxley 1893-4 (Vol. VI), x.  175  According to Huxley, this “spirit” had been exhibited in “philosophical criticism” put forward by modern philosophers such as Descartes, Locke, Berkeley, Hume, and Kant.473 Although, as he noted, the term “critical philosophy” had become famous because of Kant, he understood modern philosophy as a segment of history of philosophy during which critical philosophy had been embarked on and developed.474  It was Huxley’s view that the modern critical philosophers’ approach had been on the right track. He discussed shared characteristics of their approach and considered them to be desiderata that an epistemological inquiry needs to fulfill. In other words, Huxley not only shared philosophical concerns with those whom he regarded as modern critical philosophers, but also took critical philosophy as a model of philosophy.   First of all, as well known from Kant’s expression “critical,” critical philosophy involves an examination of the capability and incapability of human beings as epistemic agents.475 To Huxley, any philosophical thinker who recognizes the importance of what human beings can know and cannot know counts as a critical philosopher. Thus, Socrates, despite living in ancient Greek and having “no true disciples,” counts as a critical philosopher because he saw that knowing both what we know and what we cannot know is important.476 Although Socrates was not considered to offer a critical philosophy, as Huxley understood him, he held “a kind of inverse agnosticism.”477 Critical philosophy examines the power and conditions of human                                                           473 Huxley 1870c; 1871c; 1878h; 1879a. 474 Huxley 1878h; 1893-4 (Vol. V).  475 Huxley 1878h. 476 Huxley 1893-4 (Vol. VI), viii. 477 Huxley 1893a, 70. According to Huxley, Socrates had a basic sprit of critical philosophy but reversely assigned the realm of knowledge and of ignorance: he “set the fashion of a kind of inverse agnosticism, by teaching that the problems of physics lie beyond the reach of the human intellect; that the attempt to solve them is essentially vain; that the one worthy object of investigation is the problem of ethical life” (ibid.).  176  faculties in order to understand the nature, source and limits of human knowledge. Huxley’s conception of critical philosophy seems to be broader than usually thought. Descartes’s inquiry had shown that raw materials for knowledge are the contents of mind and that we can be certain about their existence. Locke, Berkeley, Hume and Kant also examined and wrote a monograph on human understanding and knowledge. Their specific accounts are different from each other, but according to Huxley, we should note the general point that they took the condition and nature of human faculties into consideration.   An inquiry into human faculties stands as imp