Tag: philosophy of science

Kuhnian and Conceptual Reflections on Dennett’s Critique of the Hard Problem

Disclaimer: A draft of a paper written for an analytic philosophy course. Propety of David M. Price. 2007.  No use, reuse, or sale of any or all of the following or any other content on this blog unless otherwise authorized by David M. Price.   Feel free to discuss and refer to my blog, but do not cast the content on this blog as your own.  Thanks very much!


Kuhnian and Conceptual Reflections on Dennett’s Critique of the Hard Problem


Daniel Dennett, in Facing Backwards on the Problem of Consciousness, opposes Chalmers’ strategy of ‘divide and conquer’-the proposal that explanatory problems of consciousness can be divided into categories of easy and hard-on the basis that the strategy is misguided and empirically distractive. The main thrust of Dennett’s argument is motivated by a certain kind of analogy he makes (the kind of analogy is presented in two examples, but I will focus mainly on the example of the vitalist here since they share a common and essential thread).  In particular, Dennett attempts to show that Chalmers’ position is analogous in its defectiveness to the following sort of claim: x is a hard problem for discipline y because the functional breakdown of x does not necessarily explain x itself.  Dennett charges Chalmers with the “[need] to defend his claim that his counterpart is not a conceptual mistake as well.”  Finally, Dennett maintains that in fact the hard problem of consciousness, accounting for the experience of consciousness and/or the qualia one experiences in consciousness, is wholly accounted for via recourse to the functions associated with consciousness explicated by cognitive theories.  “If you carefully dissociate all these remarkable functions from consciousness—in your own, first person case—there is nothing left for you to wonder about.” (Dennett, 5)

In what follows, I will argue both that (a) the sort of analogy Dennett has in mind falls prey or at least is possibly susceptible to Kuhnian-type criticisms and (b) on a conceptual level, Chalmers’ introduction of the hard problem is not without justification; the conceivability argument against a materialist identity theory is at least partially predicated on the empirical possibility of spectrum-inversion.  This possibility is what Chalmers later uses to develop his argument from conceivability against materialist frameworks of mind, but all that requires discussion here is the conceptual possibility of spectrum-inversion within theories of cognitive science. It shows that, on a conceptual level, the concepts do little to explain certain instances of seeing such and such.  I will first represent the major points of Dennett’s thesis and the sort of analogy they rely on.  I will then expose those points to a Kuhnian perspective of the philosophy of science.  Lastly, I will show that because theories of cognitive science allow for spectrum-inversion, Dennett’s claim that there is a relation of identity between the functional components of conscious states and consciousness itself may not be completely justified. 

            Dennett attempts to typify the problematic kind of proposal Chalmers makes via an analogy between, on the one hand, the hard problem of consciousness and explanations of the cognitive/functional components of consciousness, and on the other hand, the relation of the hard problem of life and the biological explanation of the processes necessary for it.  Dennett attributes the creation of “artifactual” (Dennett, 4) problems to the position typified by the vitalist position (a position which attempts, in all cases, to separate hard problems of being x itself from the easy problems of empirically breaking down x into functional components). The hypothetical vitalist is “…somehow under the impression that being alive is something over and above all these subsidiary component phenomena [including reproduction, development, growth, metabolism, et cetera].” (Dennett, 4)  Essentially, Dennett’s analogy is motivated by the claim that in each case, there really is nothing, or rather any real mystery, over and above the breakdown of the problem into the functional components explained by each respective branch of science.  To insist that there is something else going is to “misdescribe what is going on.” (Dennet, 4) 

Dennett next prescribes the proper solution to ‘artifactual’ problems: one must take explanatory concerns like the hard problem of consciousness (qualia or phenomenal consciousness) and life itself (though of course no one really purports vitalism anymore, Dennett simply uses it for comparative purposes) and break them down into functional components.  “Is it similarly a mistake, following Chalmers, to think that he can make progress on the easy questions without in the process answering the hard problems? I think so…I make the parallel claim about the purported subjective qualities…of experience: if you don’t begin breaking them down to their functional components from the outset…you create a monster…” (Dennett, 5)  Dennett draws parallels from the example of the vitalist to Chalmers’ hard problem of consciousness, but also oddly states “Chalmers has not yet fallen in either of these traps—not quite.” And while “[Chalmers] understand that he must show how his strategic proposal differs from these [the case of the purported vitalist and the hypothetical neuroscientist Crock]” it is also clear that Dennett thinks Chalmers either cannot at all, or at the least hasn’t yet: “Chalmers says this would be a conceptual mistake on the part of the vitalist, and I agree, but he needs to defend his claim that his counterpart is not a conceptual mistake as well.” (Dennett, 5)

In what follows, Dennett attempts to take Chalmers’ cue for first-person explanation: Dennett purports that, contrary to Chalmers’ claim (that there is nothing necessary about phenomenal consciousness implied by the concepts of cognitive functionalism ), careful dissociation of the remarkable functions of consciousness leaves one with nothing left to ponder.  Essentially, Dennett argues that the functional components denoted by cognitive theories actually are identical with the referents of introspective states that call attention to consciousness itself.  “In the course of making an introspective catalogue of evidence, I wouldn’t know what I was thinking about if I couldn’t identify them for myself by these functional differentia.  Subtract them away and nothing is left beyond a weird conviction there there is some ineffable residue of qualitative content bereft of all powers to move us, delight us, annoy us, remind us of anything.” (Dennett, 5)  If Dennett is not claiming a relation of identity between phenomenal states and cognitive states (assuming both are conscious states), then I am not sure exactly what he is saying since it seems clear that: x is identical with y if and only if the conditions under which x is identified as x are exactly the same as the conditions under which y is identified as y; that is, x and y are identical if and only if x and y share identity conditions.  Clearly, that seems to be what Dennett is arguing.

What is remarkable in this is that, in addition to involving a conceptual mistake which Chalmers either has not and/or cannot defend against, Dennett additionally charges Chalmers with the introduction of a concept of experience that “does not do any explanatory work.” (Dennett, 6) Another analogy surfaces here, and Dennett is quick to draw another parallel to Chalmers’ proposal: “we can see this [that experience as a concept does no explanatory work] by comparing Chalmers’ proposal with [another] non-starter: cutism, the proposal that since some things are just plain cute…and other things aren’t cute at all…we had better postulate cuteness as a fundamental property of physics alongside mass, charge, and space-time.” (Dennett, 6)  The point here is to indicate that analogous to cutism, the introduction of experience as a fundamental property of things does absolutely nothing to develop explanations and/or invite depth to the conceptual schema already in place.  Like the cutism example, Chalmers’ introduction of the hard problem, and the proposal to take experience as a basic property of things, do not offer independent ground.  They are unjustified and unmotivated concepts that do nothing to solve puzzles. 

This is a good spot to begin my critique.  First, on the level of the philosophy of science, it is quite clear from Kuhn’s The Structure of Scientific Revolutions that the perception of anomaly (and equally of crisis-evoking anomalies), does not require that the perceiver be immediately ready to offer a new way to explain the anomaly or set of anomalies. It may not be an appropriate time to do so: alternative theories are only necessary during states of scientific crisis, whereby the number of theorists who actually see the anomaly is at some critical level and a large part or the whole of what was the paradigm of normal science becomes something dubious to the most respected theorists. Certainly, some sort of alternative explanation may come about, but these alternative explanations need not always represent full-fledged and confirmed hypotheses.  It is reasonably certain that, for Chalmers (who, in addition to being a philosopher, was also trained on the graduate level in cognitive science) the hard problem represents a sort of anomaly for cognitive science.  It is something which cannot be accounted for in cognitive science; that is, no tweaking of the methods of functional analysis, no implementation of advanced software tools in the branch of AI, could possibly account for phenomenal consciousness. Certainly, such seems to be implied when Chalmers frames his thinking in the following way: “We can think of these phenomena [senses of consciousness having to do with discrimination of stimuli, reporting of information, monitoring of mental states] as posing the easy problems of consciousness…the problems of explaining them have the character of puzzles rather than mysteries [my italics]…the hard problem of consciousness is the problem of experience. Human beings have subjective experience: there is something it is like to be them.” (Chalmers, 247)

Chalmers points out in Consciousness and Its Place in Nature (Chalmers, 1996) that there is nothing in theories of cognitive science which necessitates the phenomenology of consciousness. Thus, in a sense, artificial intelligence could pass the Turing Test and seem intentional to the relevant audience: us.  Chalmers concedes that cognitive theories have indicated various functions which are alongside or accompanied by consciousness (such as internal reportability, specific sorts of memory, and the executive control of other cognitive processes) but they do not require the ability to be conscious in different ways of the same object, for example, and/or the ability to be conscious of the fact that one is conscious of a particular judgment (Chalmers calls these second and third-order phenomenal judgments).  I do not intend to defend the content of these ideas here, but it is quite clear that Chalmers sees functional analysis as being unable to provide an adequate account of the experience of being conscious.  Thus, to say that Chalmers’ introduction of the hard problem is misguided and generative of illusions is really to say nothing at all (redundant from a Kuhnian perspective on the involvements of paradigm-crises and perceptions of anomalies): any instance of a perceived crisis-evoking anomaly involves the perception of a problem perceived as fundamentally at odds with the current paradigm-in this case, cognitive science.

One might say that necessarily, the perception of a crisis-evoking anomaly involves the perception of something which misdirects attention, that is, its motivation is to direct the attention of theorists in the field away from normally assumed principles or explanations.  To defend his thesis against the claim that “well, really Chalmers and Dennett are simply arguing from the perspective of competing paradigms: Chalmers can see an anomaly in a location perhaps Dennett has looked before, though Dennett can’t (yet) see it,” Dennett would have to go about tweaking the paradigm such that functional analysis did provide a sufficient explanation for phenomenal consciousness.  Although, it is entirely possible that, at the current time, there is no alternative with which to explain consciousness in its phenomenological form.  Certainly there have been philosophical attempts, but on the empirical level, phenomenology is not exactly historically constitutive of normal science.  Still, at the least, Kuhnian philosophy of science indicates that if Chalmers perceives an anomaly, there is good reason to suspect the possibility that (a) its perception is not wholly unjustified, it may really indicate something fundamentally at odds with the paradigm and (b) the offering of an alternative paradigm or at the least an alternative explanation is not absolutely necessary for the act of perceiving an anomaly; on a significant level, the perception of anomaly and the posing of an alternative theory are separate or at least possibly disassociated. 

On a conceptual level, Chalmers’ introduction of the hard problem may be more justified than Dennett gives him credit for.  Chalmers’ conceivability argument to some extent can be situated in the empirical possibility (a possibility sometimes cited in the cognitive science literature) that one can posit, without contradiction, that all the conditions of a cognitive theory of vision hold for the visual perception of redness without the subject actually experiencing what the authority or the normal-state individual means by’ seeing red.’ This phenomenon is known as color-spectrum inversion.  Essentially, it shows that while the meaning of the concept red in a cognitive theory of color vision means to be in such and such conditions, one could be in those conditions while still seeing something other than what was meant by the concept red as it occurs in the theory. The concept of ‘seeing red’ means to be in a certain network of cognitive states which are causally connected in the right sort of way: there is an activation of receptors on the retina, the information is passed to V1 and so forth, and finally the subject reports “yes I see red.” As the theory stands, one could have the right receptors activated, be in the right cognitive state, reply yes to the question “did you see read?” and still not phenomenally see the same sort of red we mean when we ask that question.  Physically speaking, their red is the same as our red, but on a phenomenal level, if we saw their red, we would not call it red.  Perhaps we’d call it green, despite the fact that the same functional and causal conditions apply. 

If Dennett was right in the assertion that the functional contributions of consciousness are identical to consciousness itself, that there is nothing about consciousness which is somehow over and above the functional conditions consciousness realizes on the cognitive level, then it would not or should not be possible for a subject to identify the color red when he or she was experiencing what the theory says should be red but what would actually be green to us in our consciousness.  That is, the functional conditions necessary for the attribution of “I am aware of redness” or “Yes, what I see is red, or appears red to me” does not exhaust the possible meanings of the concept of red.  It is possible to be causally conditioned to see red, but to not really see red in the sense that we want red to be taken in.  I think that this empirical possibility lends conceptual support for Chalmers’ proposal to introduce the hard problems of consciousness.  It is clear in cases like these that it is not a matter of functionally analyzing the relevant causal and cognitive states, doing so cannot exhaust the meaning of being conscious of a certain mental state, so there may be or there must be something else to the nature of consciousness which is left out by the current paradigm of cognitive science.





























Works Cited



Chalmers, David (1996).  Consciousness and Its Place in Nature. Philosophy of Mind: Classical and Contemporary Readings. New York: Oxford University Press


Dennett, Daniel C. (1996) Commentary on Chalmers “Facing Backwards on the Problem of Consciousness”.  Journal of Consciousness Studies 3 (1) (Special Issue- Part 2) 4-6



Problems with Simplicity and Occam’s Razor: Helen Longino’s “Feminist Epistemology as a Local Epistemology”

 Occam’s Razor-courtesy of www.savagechickens.com and creator Doug Savage

I’ve been reading and rereading Longino’s “Feminist Epistemology as a Local Epistemology“.  I originally came to it because, in missing philosophy, and in particular epistemology, I wanted to go back and read influential works that curriculum or ideological influences made impossible.   Of course, neither of those constitute something intentionally enacted.

I think my favorite part of the essay is where Longino points out problems with the a central ontological commitment that drives empirical research: the idea that, in general, it is best be as simple as possible when it comes to a theory’s commitment to the existence of types of entities.  If the data can be explained in a simpler way, then it should be.  This idea, one that Longino takes as a standard virtue in epistemology, and often presumed in the natural science, is frequently characterized as “what closes the gap between evidence and theory”.

I understood these virtues also as justification for, as well as motivation to, invoke the use of “Occam’s Razor”.  Occam’s Razor is the normative view that more often than not, the simplest theory is the best choice.  While this description seems equivalent with the description issued above, it is not the same as Occam’s Razor belongs squarely to the area of theory selection/comparison.  Thus is rears its head only in the context of a philosophical or scientific dispute over which one of two or more possible theories a particular scientific community should invoke and/or “project”–to use Nelson Goodman’s terminology.

Longino’s arguments against both the standard epistemic value (simplicity) and the standard method of theory selection (Occam’s Razor) bring up important inconsistencies that many academic philosophers have failed to acknowledge and/or listen to–as Longino also points out.

In any event, here are her most concise arguments levied against these normative constraints:

i. This formulation begs the question what counts as an adequate explanation.  Is an adequate explanation an account sufficient to generate predictions or an account of underlying processes, and, if explanation is just retrospective prediction, then must it be successful at individual or population levels?  Either the meaning of simplicity will be relative to one’s account of explanation, thus undermining the capacity of simplicity to function as an independent epistemic value, or the insistence on simplicity will dictate what gets explained and how.

ii. We have no a priori reason to think the universe simple, i.e. composed of very few kinds of thing (as few as the kinds of elementary particles, for example) rather than of many different kinds of thing.  Nor is there or could there be empirical evidence for such a view.

iii. The degree of simplicity or variety in one’s theoretical ontology may be dependent on the degree of variety one admits into one’s description of the phenomena.  If one imposes uniformity on the data by rejecting anomalies, then one is making a choice for a certain kind of account. If the view that the boundaries of our descriptive categories are conventional is correct, then there is no epistemological fault in this, but neither is there virtue.

I will discuss these and analyze them in greater detail in a (near) future post.  In the meantime, you can see Longino’s piece Louis P. Pojman’s The Theory of Knowledge-Classical and Contemporary Readings (Third Edition)