Metaphor, History, Consciousness: From Locke to Dennett (original) (raw)

METAPHOR, HISTORY, CONSCIOUSNESS: FROM LOCKE TO DENNETT

LYNN HOLT

Abstract

Our depths who fathoms, or our shallows finds, Quick whirls, and shifting eddies, of our minds? Life’s stream for Observation will not stay, It hurries all too fast to mark their way. —Alexander Pope, Letter to Cobham

All I have done, really, is to replace one family of metaphors with another. . . . Metaphors are the tools of thought. No one can think about consciousness without them.
—Daniel Dennett, Consciousness Explained

As Joseph Margolis recently noted in this journal, contemporary analytic philosophy, particularly philosophy of mind, is all too often conducted in a cultural and historical vacuum. As he put it, the major figures in philosophy of mind are "methodological solipsists who ignore the historicized formation of the human mind and its inseparability from a historicized picture of whatever we take reality to be."1 Ingenious defenders of a variety of -isms parlay thought experiments back and forth, while the admissible evidence for interpreting these experiments is restricted to introspection, experimental psychology, and computer construction and programming. But no one is reading Pope or Wordsworth. And I admit that the mere suggestion strikes the serious ontologist as ludicrous; surely the ontological fact of consciousness is not going to be illuminated by history or literature, since these enterprises presuppose the ontology of consciousness.

I want to argue, however, that the consequences of taking seriously Dennett’s remarks on metaphor involve widening the scope of evidence and imagination much further than is ordinarily allowed, in the direction of history and literature. But caveat lector: such is the implausible nature of my suggestion that I will proceed by indirection and illustration. The argument will emerge only toward the end.

1. DENNETT AND HIS CRITICS: BACK TO THE FUTURE

Just this sort of skeptical response to an implausible suggestion was evinced by several contributors to print symposia that followed the publication of Daniel Dennett’s Consciousness Explained a few years back. 2{ }^{2} In his book, Dennett had gone some ways toward expanding the admissible evidence for consciousness, having suggested that cultural evolution plays a central role in structuring consciousness. He was attacked by those who claim that he has simply avoided the ontological issue: “Culture, learning, language, and the rest have much to do with the specific contents of consciousness, in normal mature human beings. But they do nothing to explain, they merely presuppose, the fact of consciousness.” 3{ }^{3} One diagnosis of Dennett’s alleged failure to address the real philosophical issue is that he takes his metaphors too seriously. In so doing, he is led to focus his attention in the wrong places. “Dennett’s replacement metaphors leave out something which is essential to us as human beings.” 4{ }^{4} That something, presumably, is the bare subjectivity of consciousness apart from its contents.

Dennett’s own response to criticisms like this is, perhaps, extraordinary. Rather than trying to show that his critics are wrong in claiming that he leaves out an analysis of consciousness as the support for its contents, Dennett says, “I do not shrink from the apparently embarrassing implication. . . . The alternative hypothesis, which looks pretty good, I think . . . is that, first appearances to the contrary, consciousness itself is a content-system, not a medium.” 5{ }^{5} This assertion, of course, is of a piece with his book. For both in that work and in this response he claims that his view is so new that his critics have a hard time even imagining it (“some serious thinkers find it impossible even to entertain my hypotheses”), and that when they cannot imagine it, they are mired in a bogus intellectual tradition. 6{ }^{6}

Dennett’s claims about the lack of philosophical imagination are right, but not because Dennett’s views are so very new, or because his critics are mired in an older tradition. For what his critics are objecting to, though they do not seem to know it, is a three-centuries-old metaphor into which Dennett breathes new life. But this suggests that neither Dennett’s critics nor Dennett fully understand the nature of their disagreement. The debate between Dennett and his critics needs to be recast by introducing evidence from intellectual history. In brief, my claims will be threefold: that a metaphorical shift occurs in the early eighteenth century which provides Dennett with much of the grist for his philosophical mill; that Dennett succeeds (without recognizing it) in restoring the late seventeenth-century metaphorical status quo, albeit with a few twists of his own; and that in so doing, he inadvertently points the way for intellectual history and literary/cultural criticism to supplement cognitive science in our efforts to understand consciousness. 7{ }^{7}

2. LOCKE AND HIS CRITICS: A METAPHORICAL SHIFT OF CONSCIOUSNESS

Christopher Fox has argued that consciousness, as an explicit subject of philosophical/psychological investigation, arose quite late in the Western tradition, near the end of the seventeenth century. 8{ }^{8} Specifically, it appeared in the last decade of that century with the publication of the second edition of Locke’s Essay, in a new chapter ( 27 of part 2) written at the suggestion of William Molyneux of Dublin. 9{ }^{9} Molyneux had asked that Locke devote attention to a new topic: the principium individuationis, the principle of individuation or identity. Locke responded with a new chapter which introduced a new thing, personal identity, which was distinct from identity of substance and identity of individual man. 10{ }^{10} He further proposed that the criterion of personal identity is consciousness, by which he meant "the perception of what passes in a Man’s own mind. 11{ }^{11} In tying consciousness to an issue that already exercised philosophical thought, Locke made consciousness itself available for theorizing.

Fox notes that the years between 1695 and 1740 saw the first series of exchanges between scholars explicitly directed toward the notion of consciousness. 12{ }^{12} This debate reached its apex when in 1728 Charles Mayne published the first work ever exclusively devoted to consciousness, the Essay on Consciousness, in which he noted that there had heretofore been “no Account whatever of Consciousness . . . in the most elaborate Discourses concerning the Mind . . . unless occasionally, or where the doing of it was in a manner unavoidable.” 13{ }^{13}

To say that consciousness as a philosophical problem arose in 1694 does not mean that philosophically astute thinkers were totally unprepared to think about it. They simply had not had reason to do so. But Locke unwittingly gave them reason, for the universal hue and cry of Locke’s critics is that his criterion for personal identity will not support the weight it must bear. For any criterion of personal identity must be able, on their view, to show how a person can be held legally responsible for her actions, and to show moreover that it will be the same person who, if fortunate enough, will be resurrected to join with Christ in the end times. Ample evidence of the breadth and temporal length of the outcry against Locke can be found not only in the works of Samuel Clarke, James Stillingfleet, Joseph Butler, and later George Berkeley, Alexander Pope, and still later Thomas Reid (1785), but it also can be found in numerous sermons preached at Oxford over some forty years, with titles such as Henry Felton’s The Resurrection of the same Numerical Body . . . in which Mr. Lock’s [sic] Notions of Personality and Identity are confuted (1725 Easter at Oxford). 14{ }^{14}

But what is the conception of consciousness that gives rise to these criticisms? It can be gleaned not only from Locke’s detractors but also from supporters like Anthony Collins: "no Man has the same . . . Consciousness to Day that he had

Yesterday"; consciousness is the matter “of the Brain in constant flux.” 15{ }^{15} In the words of Collins’s critic Clarke, consciousness is a “mode or power, like the Roundness or the Mode of Motion of Circles upon the Face of a running Stream.” 16{ }^{16} Pope uses the same flux and stream metaphors: the “depths . . . whirls . . . shifting eddies of our minds/ Life’s stream . . . hurries.” 17{ }^{17} And Reid says that consciousness is “flowing . . . like the water of a river,” is "transient and momentary, and has no continued existence."18

What had Locke’s new criterion supplanted? In the words of Edward Stillingfleet, Locke had “almost discarded Substance out of the reasonable part of the world.” 19{ }^{19} Berkeley’s Philonous (mind-lover) in Third Dialogue responds to Hylas’s assertion that he may be “only a system of floating ideas, without any substance to support them” with the claim that there must be “one individual principle” uniting diverse ideas. “I myself am not my ideas, but somewhat else, a thinking, active principle” that supports and entertains ideas. 20{ }^{20}

It seems clear from the remarks of Locke’s objectors that they perceived his consciousness criterion of personal identity to be an attack on the still prevalent notion of the self as immaterial (or, in some cases) material substance. This notion is, of course, vaguely Aristotelian, and was given to Scholastic thought by Boethius: a person is an individual substance with a rational nature, an enduring self that could be held responsible in the next life for acts in this. 21{ }^{21} Thinking, as Locke’s critics did, that consciousness itself is too fluid and shifting to establish self-identity, they thought it necessary to ground consciousness in the notion of individual substance. But as we all know, Locke’s critics gradually died out, and the quasi-Aristotelian substantial self got the boot. Indeed, by 1740, Hume could say that “most philosophers . . . think . . . that personal identity arises from consciousness.” 22{ }^{22}

This shift of thinking about consciousness, from a discontinuous stream in flux that needs support from a substantial self to consciousness itself as the support, is the crucial move that supplies Dennett with his whipping boy. So perhaps we need to move to a more explicit discussion of metaphor as well as Dennett’s own metaphors. 23{ }^{23} But pause briefly here to recognize that a widely held metaphor for consciousness in the late seventeenth and early eighteenth centuries is the stream-in-flux metaphor, which is often thought to arise in William James’s work at the turn of the twentieth century, and which Dennett frequently makes use of.

3. METAPHORS FOR CONSCIOUSNESS AND SOME CONSEQUENCES OF MISATTRIBUTION

The mind is not what it once was. Most will agree if all that I mean is that thought patterns vary over time. 24{ }^{24} Consciousness, on this view, is a stable element

of human nature that underlies changes of its contents over time. One implication of understanding the mind in this way is that, whatever philosophical problems there are about consciousness, they are not to be found in ferreting out merely historical differences in the content of thought. Such differences may well interest historians of ideas, but they cannot touch on “the problem of consciousness,” because that is what underlies changes in content and what indeed makes such content possible.

There are two basic structural metaphors at work in this view of consciousness: consciousness supports its contents (the support metaphor) and consciousness is a container for its contents (the container metaphor). 25{ }^{25} Exactly what kind of support and containment necessitates a further trope. For example, Hume suggested in the Treatise that “the mind is a kind of theatre, where several perceptions successively make their appearance; pass, repass, glide away in an infinite variety of postures and situations.” 26{ }^{26}

Dennett attributes this metaphor to Descartes, but it is instructive to see why Dennett is wrong in his attribution of this as a metaphor for consciousness. 27{ }^{27} Dennett describes the Cartesian Theatre metaphor for consciousness as a single central stage (material or immaterial) in which the objects we are aware of pass in chronological review before the mind’s audience. According to this metaphor, consciousness is the central place where it all comes together and is unified. But Descartes was sufficiently a Schoolman to believe in and argue for a res cogitans, a thinking substance that supports the activity of consciousness and gives consciousness its unity; thinking requires a thinker. 28{ }^{28} In these metaphorical respects, Descartes was a conservative, a Scholastic holdover.

By contrast, though in part unintentionally, Locke initiated a move that culminated in a view of consciousness as a self-unified entity, itself a support for thoughts and ideas. This move was in part unintentional since, even though Locke thought consciousness to be stable enough to ground identity claims, he continued to believe in a mental substance that supported consciousness. Unfortunately, this substance (just like matter) is opaque to the empirical consciousness, a “something I know not what.” Locke was a novel figure not only because he grounded personal identity in consciousness, but because he thought of consciousness as a stable container of or support for all that passes in a person’s mind. Not that such a support or container metaphor was itself new; what was new was identifying the support as consciousness, not substance. This view gained widespread acceptance only after some fifty-odd years, and even then it had significant opponents, such as Thomas Reid. But it is this quasi-Lockean view of consciousness which Dennett attacks, not Descartes’ view.

So Dennett gets the metaphorical history of consciousness in the seventeenth and eighteenth centuries wrong. What bearing does this have on our understanding and assessment of his view?

Let us suppose that my phrase “the mind is not what it once was” makes a more radical claim, and that it is also Dennett’s claim. Dennett says that he rejects the underlying structural metaphors that themselves give plausibility, aid, and comfort to those who conceive of consciousness as a unitary phenomenon. Dennett’s explicit metaphors, the “stream” of consciousness with multiple currents, “multiple drafts” text metaphor, and the “pandemonium” of subprocessors all attempt to model a consciousness that is not unified and is not a support or containing structure. In fact, his model for consciousness appears to reject the support/supported and container/contained dichotomies by jettisoning the support and container and retaining the supported and contained; that is, consciousness just is its content. This is very close to another of Hume’s metaphors: the self as a bundle of impressions. 29{ }^{29} Couple the claim that the contents of thought have changed over time with Dennett’s metaphors for consciousness, and you get a more radical claim, since if there is nothing to consciousness but its shifting contents, then you can never step into the same state of mind twice. 30{ }^{30}

We must be more careful, though; to say that consciousness is nothing but its contents invokes the support/container metaphors again. Worse, the invocation then begs for an account of what the container is like, emptied of its contents. If Dennett really is abandoning the container and support metaphors, then no account of the bare fact of containment (the bare fact of consciousness) that makes the contained possible will be forthcoming. That no such account is forthcoming is easily seen by returning to Dennett’s critics. The telling title to one article is “Consciousness Avoided,” and another claims that "Dennett’s book is not a case of consciousness explained, but of consciousness . . . ignored. 331{ }^{331} Such slogans pinpoint what exercises Dennett’s critics on this matter: that Dennett has failed to explain the bare existence of consciousness in a world populated also by nonconscious entities.

We might at this point be tempted to draw a historical parallel between Locke’s critics and Dennett’s. And insofar as both sets of critics recognize an attack on entrenched views, the parallel exists. But it was Locke’s critics who figured consciousness as a stream in flux; Dennett’s critics figure consciousness as the support for its contents. For they assume that there must be a consciousness that is separable from its content, which supports and/or contains its content; the stream must be unified, there must be a consciousness that authors its contents. In this they accuse Dennett of being more like Locke than Descartes, but exactly the reverse is true. Because Dennett is as plain as he can be that there is a support for consciousness: the brain. Consciousness itself is not a support, nor is it itself a medium. What he recognizes is that the container and support metaphors are misapplied to consciousness if consciousness is meant as the support and container. Consciousness even at its most unified is membrane-thin (as Nietzsche had it), incapable of supporting the metaphysical weight of its

supposed contents. To apply the support and container metaphors properly, we must find a support for the flux that is consciousness, and for Dennett, the brain will serve quite well. Dennett has not abandoned the support and container metaphors after all; he has argued for a remapping of the metaphorical items onto (putative) mental structures.

Dennett’s answer to his critics at this stage could thus be two-fold. 32{ }^{32} First, the critics are led astray by a Lockean misuse of support metaphors into thinking that there must be a bare consciousness that is separable from its content. To rid themselves of this notion is to re-apply the metaphor properly, locating consciousness as the item supported. Having done so, his critics are now in a position to appreciate his evolutionary continuum story about the buildup of gradually more complex content-fixing structures, from the rudimentary and staid contents of the sea-squirt’s brain to the complex and plastic contents of ours.

4. A NEW PARSING OF AN OLD METAPHOR, WITH APPROPRIATE TEXTUAL OVERLAY

Dennett doesn’t really abandon the support/container metaphors after all; what he does is shift their application, turning consciousness back into the supported and contained. There is thus nothing to consciousness but its content. But now what better metaphor for such content than a text metaphor? Commenting on a David Lodge novel which is itself commenting on bad deconstructionism, Dennett agrees that there is no self that is not constituted by texts; the self is a text, constantly rewriting itself. 33{ }^{33} Later, narrating the multiple draft metaphor, he says: "Our tales are spun, but for the most part we don’t spin them; they spin us. Our human consciousness . . . is their product. 34He{ }^{34} \mathrm{He} goes on to say that the convictions that there cannot be “quasi-selves or sort-of selves . . . are not self-evident.” 35{ }^{35} The self is just one among many fictional entities created by brain activity. This explicit text metaphor for consciousness overlays the less explicit support and container metaphors.

How do we investigate this phenomenal self-text? We engage in heterophenomenology. This consists in constructing a third-person text of someone’s behavior, including verbal behavior, scrupulously avoiding evaluating the subject’s claims about existence and causation. “The heterophenomenological method neither challenges nor accepts as entirely true the assertions of subjects, but rather maintains a constructive and sympathetic neutrality, in the hopes of compiling a definitive description of the world according to the subjects.” 36{ }^{36} The point here is that a self and its bare awareness are themselves, heterophenomenologically described, nothing but elements in a text, albeit a definitive draft produced by an “objective” outside editor.

Now it is crucial to note that this heterophenomenology is required by Dennett’s metaphorical shift, and not the other way around. Every item in a draft of a fictional work, qua item in a work of fiction, has exactly the same ontological status: fictional entity. If we want to engage in the philistine work of saying, e.g., what Edmund Spenser’s Red Cross Knight really is, then we might begin to assign external referents and causal powers that are reflected in the text (the knight is really his father, so embodied because of his guilty desire for his mother). But Dennett wants to be a run-of-the-mill deconstructionist, so every item in the multiple channel stream of consciousness must be of a piece, none of it privileged: bits of content, struggling for air, some rising to the surface because powerful currents sweep them there, some fished out of the water by some external probe of a fisherman-psychotherapist.

Once the heterophenomenological move has been made, it facilitates Dennett’s metaphorical shift. After all, this move is itself a metaphor; it figures the self and its objects, a you and I exchange, as a report by an outside party. This is not to suggest that his not quite self-conscious neutral observer metaphor is illegitimate. Quite the contrary, such a trope is necessary for the continuance of his project. As Dennett insists, he wants to provide a radical alternative to common ways of conceiving consciousness and the mind. He rightly asserts that such rethinking involves refiguring the issue by the employment of metaphor, and thus metaphors are the necessary “tools of thought” for conceiving consciousness.

Curiously, however, though Dennett is willing to ascribe metaphorical status to other elements of his account, he doesn’t want to think of the heterophenomenological turn as a metaphor. Indeed, he says that this thirdperson business is a sine qua non of science. 37{ }^{37} Properly scientific theories, according to Dennett, must ultimately be judged by how well they account for the “data”; but only facts gathered by neutral observers, “from the outside,” can count as “data of science.” Since mental events (if they exist), by their very nature, “are not among the data of science” because they cannot be “properly verified by objective methods,” we must have some way of rendering them as suitable subjects for investigation. We can’t rely on first-person reports, since they are not properly objective; what we need is a third-person point of view, “since all science is constructed from that perspective.” Such a third-person perspective will provide “a neutral way of describing the data.” Heterophenomenology is just such a neutral method of providing data for a science of consciousness.

This may well be merely a naive empiricism coupled with an insupportable dichotomy between theory and data. But I prefer to think of it as a brilliant metaphorical strategy for figuring the inquiry, barely but effectively concealing the tropic nature of the move from those who would object to such a procedure, but

who would not so readily object to a “scientific” account of phenomenology. Wasn’t it Husserl who also wanted a science of phenomena? Dennett can later afford to use the language of metaphor and model in connection with his multiple drafts and streams of consciousness since these are plausible metaphors of consciousness from the third-person point of view, and he has already gained the high ground of objectivity with a point of view that he claims is the essence of science.

5. THE CONSEQUENCES OF METAPHOR

What Dennett has done is two-fold. He has shifted the application of the fundamental (and background) support and container metaphors back to their seventeenth-century positions. He has then further figured the consciousness supported by the brain with a variety of explicit metaphors, ranging from a resurrection of the seventeenth-century stream metaphor to text and machine metaphors (the latter an eighteenth-century creation). In this metaphorical layering resides Dennett’s unique contribution.

But there are some interesting consequences of Dennett’s metaphorical layering, consequences that he does not draw. Let us focus on the text metaphor. If the features of consciousness are just parts of texts, then a study of consciousness should properly be a textual analysis of, in part, the metaphors which structure that consciousness. After all, Dennett himself claims just this structuring efficacy for metaphor in his discussion of memes. 38{ }^{38} Of course, Dennett wants to use the quasi-scientific language of information processing when he says that memes (bits of information, following Richard Dawkins) structure our consciousness. But he fails to recognize the implications of his own work by not taking seriously the in-forming capability of metaphor, the figuring power of metaphor. In this respect, the Humean “theatre of the mind” has been a fairly successful metaphor, precisely because it succeeded in structuring a great many minds, among them some of Dennett’s critics who, when they introspect, find just such a metaphorical structure.

The claims that metaphors structure texts, and that consciousness just is a text, or is composed of parts of texts, have two important consequences. The first is that the phrase “the mind is not what it once was” is true for Dennett in a radical sense. Since the mind and consciousness are nothing but content, and since our current contents can recognize a variety of different contents with different organizing principles across history, there really are different consciousnesses. The reflexive corollary is that, with his metaphors, it’s possible Dennett has explained only contemporary and late seventeenth-century consciousness. Historians of ideas are always telling us that contemporary consciousness is much more fluid and fragmentary than that of some other ages.

Which brings me to the second consequence. Though textual interpretation must now be central to cognitive science on Dennett’s view, he seems woefully ignorant of it, and merely waves a hand at the possible problems associated with interpreting the texts generated by individual consciousnesses, in particular consciousnesses that seem to be structured in different ways than our own. For all his citations, he has not once consulted the literature in history of ideas or cross-cultural literary criticism that is directly focused on consciousness. His ignorance of these works, coupled with his metaphorical figuring of consciousness as text, evinces a great lacuna, almost a willful misdirection towards experimental science away from historical textual criticism. Though not for their reasons, I am now in partial agreement with Dennett’s critics: he has ignored the most fundamental features of consciousness on his own view by ignoring the findings and interpretive methods of textual criticism.

If consciousness really is a text-that is, if the best metaphor around for consciousness really is a text metaphor-why should philosophy of mind only learn from cognitive psychology? I am not saying that philosophers of mind should sever the ties which have bound them to cognitive scientists for some thirty years. But if there is nothing to consciousness but its contents, and these are historically constituted, then the direct study of consciousness by philosophers stands to be enriched by learning what historians of ideas and literary critics have to say. And though you should not judge the possible impact which such learning might have on philosophy by my meagre attempts here, nevertheless my offering is in part an attempt to enrich the philosophical debate by broadening its scope to include the relevant extramural literature.

What is the character of this enrichment? Consider just one example. Suppose Dennett had read Fox’s book on Locke, identity, and consciousness. He would, of course, recognize the immediate affinities of his view with the prevailing seventeenth- and eighteenth-century mindset in Britain (and thus would be spared from committing the barbarism that his view is radically new). But more importantly, he would have gained powerful evidence for arguing against those who claim that their phenomenal worlds are unified by a single self. For as we noted, most critics of Locke, and some of his supporters, found disunity upon introspection. Their phenomenal world was (on the best textual evidence around) very different from Dennett’s critics. Dennett could turn the table on his critics by noting that their introspective results are heavily conditioned by the culture in which they live, and that is to be expected on his view. Indeed, a general lack of historical continuity of consciousness would not be surprising on Dennett’s view, though it would be for those who claim that introspection yields the notion of a unified support/container structure.

I close by returning to a possibility suggested a moment ago. In recognizing that there are different consciousnesses, we may also recognize that some are

better than others. And it is possible that, as texts writing ourselves, we also recognize that the type of mind which we have is an achievement; change may not be only by chance. As Stephen Clark suggests, a unified self is hard work, but worth it. So one final consequence of introducing intellectual history and textual criticism into the philosophical debate may be a therapy of the imagination. But that is a metaphor which needs exploring.

Mississippi State University

NOTES

1 Joseph Margolis, “The Growing Philosophical Neglect of History and Culture,” Philosophical Forum 29 (1997), 291. His claims about the conceptual impoverishment of late analytic philosophy extend well beyond philosophy of mind, but he thinks it a prime case in point.
2 Daniel Dennett, Consciousness Explained (Boston: Little, Brown and Co., 1991). Philosophy and Phenomenological Research 53 (December 1993); Inquiry 36 (March 1993). Not all of the contributors to these issues respond this way. Those contributors who make the claim that Dennett has somehow missed the point about consciousness by not addressing what they see as the fundamental issue(s) about consciousness, how the thing itself could arise in an otherwise non-conscious world, are as follows. Inquiry: John Foster, “Dennett’s Rejection of Dualism,” 17-31; Michael Lockwood, “Dennett’s Mind,” 59-72; Stephen R. L. Clark, “Minds, Memes, and Rhetoric,” 3-16; Roger Fellows and Anthony O’Hear, “Consciousness Avoided,” 73-92. Philosophy and Phenomenological Research: Frank Jackson, “Appendix A (For Philosophers).” Judging by Dennett’s responses in each issue (“Living on the Edge,” Inquiry, 135-60; “The Message Is: There Is No Medium,” Philosophy and Phenomenological Research, 919-31), T. L. S. Sprigge, “Is Dennett a Disillusioned Zimbo?” Inquiry, 33-58, William Seager, “Verificationism, Scepticism, and Consciousness,” Inquiry, 113-34, and Sydney Shoemaker, “Lovely and Suspect Ideas,” Philosophy and Phenomenological Researach, 905-10, make the same charge, though I do not find it explicitly in their articles.
3 Lockwood, “Dennett’s Mind,” 67.
4 Fellows and O’Hear, “Consciousness Avoided,” 75. Note also Jackson, “Appendix A (For Philosophers),” 899.
5 Dennett, “Living on the Edge,” 155.
6 For instance, in his response in Philosophy and Phenomenological Research, Dennett writes: “I view the standard philosophical terminology as worse than useless-a major obstacle to progress since it consists of so many errors trapped in the seductively lucid amber of tradition: ‘obvious truths’ that are simply false, broken-backed distinctions, and other cognitive illusions.” (“The Message Is: There Is No Medium,” 921). See also his response in Inquiry, 135.
7 Thus the explicit presupposition animating this paper is that many contemporary debates may be both enriched and advanced by actively engaging their history.
8 Christopher Fox, Locke and the Scriblerians: Identity and Consciousness in Early EighteenthCentury Britain (Berkeley: University of California Press, 1988), chaps. 1 and 2. Fox provides an admirable account of the seventeenth- and eighteenth-century reception of Locke’s views on personal identity in the second edition, and the subsequent satire of the debates on personal identity found in the Memoirs of Scriblerus. My claims about shifting metaphor and the metaphorical character of consciousness in the seventeenth and eighteenth centuries (and in the twentieth century) should not be attributed to Fox, however. Any faults in those claims are mine. But I am

in his debt for the sources to which he points, and the character of this indebtedness, as well as its implications for philosophy of mind, are discussed in the last section of this paper.
9 My thanks to the archives of the University of the South for allowing me access to a well-preserved copy of the 1694 second edition. The chapter title in the second edition was “Of Identity and Diversity.” Molyneux precipitated a lengthy correspondence with Locke when he praised Locke above all others in the dedicatory epistle to his Dioptrica Nova (London, 1692). Molyneux made many suggestions for improvement, which were incorporated in the new edition: marginal summaries, typographical corrections, rectifying small errors of fact. In his letter to Locke dated 2 March 1693, Molyneux suggested that Locke devote space to what he calls the principium individuationis. Locke wrote back on 23 August 1693, to send a pre-print copy of a new chapter, which was “writ only at your instance.” See The Correspondence of John Locke, ed. E. S. DeBeer (Oxford: Clarendon Press, 1976), v. 4. This exchange is also in The Works of John Locke, 9 vols. (London: C & J Rivington, 1824), v. 9.
10 Section 9 in the second edition. The story of Prince Maurice’s rational parrot, an attempt to sever the notion of person and intelligence from an essential connection with the human body, does not appear in the second edition, though there is a brief comparison between a hypothetical “very intelligent rational Parret [sic]” and a “dull irrational man.”
11 This definition of consciousness is found in Book 2, chapter 1, section 19. In section 23 of chapter 27, Locke writes: “Nothing but consciousness can unite remote existences into the same person. . . . So that self is not determined by identity or diversity of substance . . . but only by identity of consciousness.” He goes on to say in section 26 that “Person, as I take it, is the name for this self.”
12 A list of contributors to the debate can be found in Fox, Locke and the Scriblerians, 38. Beyond noting, as Mayne does, that no one had written anything substantial directed toward some notion of consciousness prior to the late seventeenth century, we may note some linguistic evidence. Fox himself (12-13) claims that the received view (see An Essay Concerning Human Understanding, ed. A. C. Fraser [Oxford: Clarendon Press, 1894], notes to chapter 27, section 11) that Locke imported “consciousness” from post-Cartesian French is wrong, since Locke’s first French translator, Pierre Coste (Locke’s assistant for several years), felt obliged to provide apologies for rendering the term as conscience, since it was such a stretch of the usual meaning. The groundwork for Locke’s usage of “consciousness” is accomplished when late fourteenthcentury English borrows the Latin term conscientia to produce “conscience,” and sixteenthcentury English borrows conscius to produce “conscious.” The Oxford English Dictionary cites Locke’s Essay as the first usage of “consciousness” in his philosophical sense. Whether “consciousness” is an English coinage, or whether it is an import from France, consciousness had not been a serious subject of inquiry until Locke. And even then, discussions of consciousness in the next thirty years, while much more numerous and extensive, were nevertheless subordinated to questions of identity over time.
13 Charles Mayne, Two Dissertations Concerning Sense and the Imagination. With an Essay on Consciousness (London, 1728), as quoted in Fox, Locke and the Scriblerians, 11. The authorship of this text is disputed.
14 Fox, Locke and the Scriblerians, 149n9.
15 Anthony Collins, An Answer to Mr. Clarke’s Third Defence, in The Works of Samuel Clarke (London: J. and P. Knapton, 1738), v. 3, 870, as quoted in Fox, Locke and the Scriblerians, 18−1918-19.
16 Samuel Clarke, A Third Defence, in The Works of Samuel Clarke, v. 3, 889, as quoted in Fox, Locke and the Scriblerians, 51.

17 Alexander Pope, Epistles to Several Persons, ed. F. W. Bateson, v. 3 of The Twickenham Edition of the Works of Alexander Pope, ed. John Butt (New Haven: Yale University Press, 2d ed., 1961), 17, lines 29-32.

18 Thomas Reid, An Essay on the Intellectual Powers of Man, ed. A. D. Woozley (London: MacMillan and Co., 1941), 214.
19 Edward L. Stillingfleet, Bishop of Worcester, Discourse in vindication of the Doctrine of the Trinity (London, 1696), 234. Locke replies in Letter to the right reverend Edward L., Bishop of Worcester (London: Clark, 1697) that “I have done nothing towards the discarding Substance out of the Reasonable Part of the World . . . because all Simple Ideas . . . carry with them a supposition of a Substratum to exist in” (9-10).
20 George Berkeley, Three Dialogues Between Hylas and Philonous, in The Works of George Berkeley, ed. A. C. Fraser, v. 1 (Oxford: Clarendon Press, 1901), 450-51.
21 “Persona . . . est naturae rationabilis indiuidua substantia.” Boethius, The Theological Tractates, trans. H. F. Stewart and E. K. Rand (Harvard: Loeb Classical Library), 92.
22 David Hume, A Treatise of Human Nature, ed. L. A. Selby-Bigge, rev. P. H. Nidditch (2d ed., Oxford: Clarendon Press, 1978), 635.
23 For two very different accounts of Dennett’s metaphors, see Robert Kirk, “The Best Set of Tools? Dennett’s Metaphors and the Mind-Body Problem” Philosophical Quarterly 43 (1993) 335-43; Teed Rockwell, “Global Workspace or Pandemonium?” Journal of Consciousness Studies 4 (1997), 334-37.
24 For example, there is widespread agreement among historians of science that Newton would not have had the thoughts he did about the relationship of analytical geometry to natural philosophy had not Galileo provided demonstrations about nature which relied solely (or so Galileo maintained) on geometrical principles. This latter thesis is unobjectionable (though not trivial) in part because it makes no claims about changes in thinkers that are not accountable for in terms of changes of thought.
25 I do not mean to suggest that these metaphors are employed by their users with full awareness. On the contrary, such fundamental structural metaphors typically provide the tacit background against which explicit inquiries are carried out. For discussions of this structuring and figuring function of metaphor, see George Lakoff and Mark Johnson, Metaphors We Live By (Chicago: University of Chicago Press, 1980); Hayden White, Tropics of Discourse (Baltimore: Johns Hopkins University Press, 1978). The support and container metaphors are not mutually exclusive, and often are mixed by the same thinker.
26 Hume, Treatise, 253. Hume himself is quick to point out that we should not be misled by the theatre comparison: “They are the successive perceptions only, that constitute the mind. Nor have we the most distant notion of the place, where these scenes are represented, or of the materials, of which it is compos’d.”
27 I learned this from Lockwood’s contribution to the Dennett symposium, “Dennett’s Mind,” 60.
28 Descartes, Principles of Philosophy, trans. John Cottingham, Robert Stoothoff, and Dugald Murdoch (Cambridge: Cambridge University Press, 1985), 195, 203, 210-11. The term Descartes uses for “thought” in the widest sense possible, “everything we are aware (conscius) of as happening within us,” is cogitationes. Conscientia is the awareness we have of this, and to be clearly and distinctly aware we must intuit (intueri) our thoughts. See John Cottingham, Descartes (London: Basil Blackwell, 1986), 39ff.
29 Hume, Treatise, 252. Dennett offers this passage as an epigraph to chap. 13 of Consciousness Explained. However, Hume’s bundle is too tightly drawn for Dennett; a loose and unstable assortment, a melange of overlapping and competing fragments, a shifting and unstable “pile” would be better.
30 With apologies to Heraclitus.

31 Fellows and O’Hear, “Conciousness Avoided,” 73-92; Foster, “Dennett’s Rejection of Dualism,” 30 .
32 This is, of course, not his response.
33 Dennett, Consciousness Explained, 410-11. See also Daniel Dennett, “The Interpretation of Texts, People and Other Artifacts,” Philosophy and Phenomenological Research, v. 50 supplement, (Fall 1990) 177-194.
34 Dennett, Consciousness Explained, 418.
35 Dennett, Consciousness Explained, 422.
36 Dennett, Consciousness Explained, 83.
37 The following phrases are all from Dennett, Consciousness Explained, 70-71, emphasis in original.
38 Chap. 8: “How Words Do Things with Us.”