In response to: “Imperfect Ignorance” (Vol. 2, No. 4).
To the editors:
There is a lot to like in Arthur Cody’s essay: some things on which we clearly agree, some things on which we disagree, and some things where I cannot tell if we agree or disagree. Of course, Cody mentions a wealth of matters and I cannot address them all and will not try. I will gesture in the direction of some overall themes that I think his essay raises.
Cody maintains that the topic of cognition is a big or bigger problem than what Chalmers calls the hard problem of consciousness—Chalmers calls it the hard problem, rather than the big problem.1 I am not sure about bigger, and am unsure whether size matters here, but I agree whole-heartedly that cognition is a problem.
Cognitive science as a discipline began in the 1980s and we still have no agreed understanding of what counts as a cognitive process—i.e., what makes something cognitive/cognition. Cody gives a list of processes and asks whether it is exhaustive. I’d ask of each item on the list: why it is there? What makes any process a cognitive one? I think this is a problem because I have discovered very smart people telling me and the world things that I do not believe to be true or justified—such as that plants and bacteria enjoy cognition (the same kind of cognition as humans and animals).2
But that is not Cody’s worry. He is more concerned with our understanding of folk psychology—the terms we use to understand one another as persons. Cody asks why, when we discuss intentional actions, we separate the belief from the intention? The answer, of course, is because they are different states with different functions (causal roles). Beliefs are there to represent the way the world is (or is not) and intentions are there to do something about it (change it, or make it better). Beliefs are different states from intentions because I can believe it is raining but have no intention (not to go out, not to take an umbrella, not to stay in, not to plan to do anything at all because it is raining). Intentions involve beliefs, it is true, but they are more than beliefs because intentions involve plans for action (even if you do not put them into action, as Cody rightly acknowledges).
I take it that Cody thinks the way folk psychology carves out psychological kinds is not adequately justified. But I do not see why not. It is true that cognitive science has only been around since about 1980, but cognitive states (beliefs, desires, intentions, memory) have been around for hundreds of years and humans have learned to recognize them and refer to them in our cognitive and linguistic interactions.
Cody, as far as I can tell, gives us no reason to think folk psychology (or the kinds of states Donald Davidson refers to in his theory of action) do not exist or do not cause things (such as behavior).3 It is one thing to wonder how it works. It is another to explain why the whole conceptual framework is mistaken. Newton did the latter for Aristotle and Einstein returned the favor for Newton, but I do not see a replacement framework from Cody for Davidson.
I surely agree with Cody that we do not yet understand how the mind works: how intentions, beliefs, and desires conspire to move us, to get us acting. I suspect that this is a major theme in Cody’s essay, and upon this I surely agree. There is much to be done to understand how things work. But if that is the main point, the essay could have been much shorter. If that is not the main point, then I am still unsure what replacement Cody has in mind for folk psychology and cognitive science. And if he does not have a replacement in mind, then these are still the only game in town for understanding purposive behavior of cognitive agents.
More agreement: Cody says many intentions may be inert. Agreed. We sometimes intend to do and yet do not. I intend to mow the lawn on Saturday (forming the intention on Friday), but do not mow on Saturday. He says: “I cannot intend to jump ten feet in the air.” Well, I cannot rationally so intend when I believe this is beyond my physical ability unassisted. But this is because rational intentions have belief constraints. One cannot rationally intend to do what is known by the same person to be impossible. Of course, on a trampoline I can jump ten feet.
There is also an air of what I’ll call holism in Cody’s musings. The issue is whether intentions or beliefs can have unique content and be distinct from other cognitive states and their contents. So, can George Washington intend to chop down the cherry tree if he does not know, or believe that it is a cherry tree, that it is a living organism—that chopping it down will kill it and require work, etc.? Now the rational intention requires believing such chopping to be possible—so there is an interaction between the intention and the belief. And the content of one set of intentions and beliefs are related to other intentions and beliefs. But this does not mean they are not distinct states just connected together. The car will not run without spark plugs, gas, and fuel injection of some kind. They all work together, but this does not mean the parts are not distinct. If Cody has an argument that the parts of the mind are not really distinct, the argument was not in his essay.
By the way, I also disagree with Cody that belief is conduct. Beliefs may or may not lead to conduct, but they are not themselves conduct. When I believe there are black holes, this may or may not lead to conduct. I may express the belief or keep it to myself. Conduct may flow from beliefs, but there is no identity here and I am not sure why Cody or anyone would think otherwise. It is also true that beliefs are not propositions, but they are related to propositions and, largely, identifiable via the propositions which are their contents.
Cody also seems to suggest that the states of folk psychology do not fit nicely with what we know about the brain. He seems to suggest that the cognitive states of intention, belief, desire, memory, and so on, may not be implementable by the brain. It is here where I just do not know why he thinks this. So I do not know if we agree or disagree. He says too little to know. It is true that intentions can be a messy lot, but what in cognitive science (or science, generally) is not? Nothing in science worth finding out is easy to learn or discover.
At one point, Cody raises the question of whether we intend every little thing we do. Do I intend to walk to the car? Do I intend each step on the walk to the car? Do I intend each left, right, each motor movement? These are excellent questions and should be asked—and have been asked. There is a growing literature on motor intentions and motor routines and how these interact with higher level cognitive plans (such as the conscious thought to go to the car).4 One need not consciously intend or will each motor movement along the way, but somehow the motor routines do depend upon the higher-level intention to go somewhere. Must there be triggers for the intention to form? Yes. Must there be triggers for the intention to interact with motor routines that get you to your car? Yes. Just how this all works and is implemented in the brain is important and is under investigation.5
So, Cody is right to point the finger at work to be done, but I cannot shake the suspicion that he thinks at bottom there will be a fly in the ointment and that we should not expect the folk psychological explanations of intentions and the neuroscience of the brain to make a smooth fit. Still, he does not explain why and, though there is more to be discovered, I cannot tell if we agree or disagree on this matter because he is not more forthcoming about why there is a problem (if there is). The closest I can get to an announced problem is when Cody seems to agree with Nagel that folk psychology involves norms (things one should and should not think) while the brain may not (if it involves things happening in the brain because of natural laws, not norms).6 Now I do not think there is a mismatch here any more than the contents of representations in a computer program may have norms in their contents (say, about ethics or law), but the program that runs on the computer conforms to the law-like programming of the computer system and language. These are compatible in your laptop. I do not see why they cannot be compatible in your mind/brain as well. Neither Nagel nor Cody persuade me otherwise (and the same can be said of many others in cognitive science).
Lastly, what I know (a big dog is growling at me) can be what I fear (a big dog is growling at me). But that does not mean knowing = fearing. It just means that these two cognitive states can have the same object. None of this is news to Cody, to be sure, so I am still at a loss as to why he thinks pointing out such matters is news or problematic. There are causal connections and influences all across one’s cognitive economy of states. And there are enough connections in the brain to implement all such causal and contentful connections. I have no idea what he means by saying these connections and divisions are not explicated. Philosophers of action and mind have been so explicating for hundreds of years and cognitive sciences have also been involved in the pursuit since about 1980. Is the progress perfect? By no means! But, as Cody says, our remaining ignorance is not perfect either.
Arthur Cody replies:
It is to his credit that Fred Adams, whose publications are many, accepted the invitation to respond to my article. I thank him for taking the time to do so when his teaching and his own writing obligations must leave him very little time. Finding twenty minutes or half an hour is not easy in a busy career.
I had thought of defending my contentions against his charges, but I realized that he did not so much make charges against my contentions as pit his opinions, which might better be called his worldview, against mine. In that Adams reflects the predominant views among American philosophers, it is for him I wrote the essay he finds more puzzling than deficient. I had hoped to persuade him that his unexamined and doctrinaire assumptions are not unquestionable. An explicit defense of them would be welcome and salutary. Therefore, rather than restate my recommendations, which my essay advanced, I shall contrast his with my worldview.
Adams and I agree on one thing, and I fear on nothing else: human cognition is poorly understood. On diagnosis, we disagree, and that is because our interpretations of the problem stem from different worldviews. His, roughly, is this: science as we receive it today dictates to us the categories in which human cognition is to be understood. Mine reverses the order. The categories in which we presently comprehend human cognition dictate what science is obliged to investigate if it can, even if it must find new categories of explanation to do so.
What worldviews are these? His accepts cause and effect, mine, cause and effect and something else where warranted. I do not advocate supernatural explanations not only because they are not warranted but because they are not explanations. I recommend that we face facts as we discern them, he recommends we re-discern the facts so as to face them. I say answers in science and philosophy ought to be guided by the problem; Adams says the problem ought to be framed in terms science and philosophy are able to answer. He will get answers that may be fallacious. Abhorring fallacy, I reach no answers at all.
Adams makes his position clear by stating it. He does not argue it, nor does he examine it. His position is held from a certainty for which serve no better words than “perfectly obvious.”
“Folk psychology,” he tells us comprehends the terms we use to understand one another. The terms to which he refers that we use are ordinary ones, “intend,” “belief,” “remember” and so forth, by which “we understand one another as persons.”7 But Adams implicitly identifies another “we,” those who use the term “folk psychology.” I do not use that term, so I am not included in that “we.” It is a term philosophers use for derision or for evasion; it is not one that has an ordinary use. I do not like it.
Paul Churchland declares that “…folk psychology is a hopelessly primitive and deeply confused conception of our internal activities.”8 This is derisive.
Daniel Dennett, in books and articles over decades, makes folk psychology the “intentional stance.” There are two other stances: the design stance and the physical stance. The physical is basic. It is that stance taken by scientists (and philosophers). It refers to the objects, laws, and techniques recognized by science. One must be qualified to take this stance. Folk psychology is a stance, revealed and expressed in “the terms we use to understand one another as persons.” According to Dennett, we ordinarily employ these words for our convenience; they are embedded in the language, and they are those for which users need possess no (scientific) credentials to use.9
This is evasive. My worries, as Adams describes them, arose because there is a very great disparity between what we can say about our cognitive life as it, being cognitive, moves on and what we can say about our brain states as they, being neuronal, move on. The presupposition that they will march as one so that the latter may, with enough scientific knowledge, explain the former, is contradicted by that disparity, which my essay tried to illustrate.
The evasion is that of diverting attention from the disconnection between life and theory by creating a vocabulary that diverts attention from life to theory as though the latter properly commands the former. On Dennett’s and Adams’s theory, the mind is the brain as we know it; therefore, the realities of brain and mind are the same.10 If our experience, on which I solicit Adams to reflect, seems to contradict this presupposition, he, with Dennett, say the presupposition wins; it follows that our experience as I portray it is misconceived. But it is not misconceived. Look to the portrait I went to such pains to draw.11
Concerning how it is we make cognitional distinctions and understand one another as persons, Adams would sooth my worries by remonstrance. He has no such worries. In a manner as simple as it is abrupt, he says, “the answer, of course, is because they are different states with different functions (causal roles).” Is this a case in which to assert is to prove?
His pronouncements in which he calls beliefs and intentions “states” presumes they are states. But if they are states, it is not obvious that they are. I see no defense for presuming they are. To say beliefs and intentions are states begs the issue between us. My essay was an attempt to argue—if Adams will permit forms of argument he eschews—that our acquaintance with our own and others’ cognition is good enough to justify the conclusion that cognitive science, as a science of brain matter, barks up the wrong tree. The word “state,” if it is anything but a place holder, refers to a condition of something. States do not exist as ponds exist but as conditions ponds exist in, frozen, dried, boiling, rank. If beliefs and intentions are states, then they must be states of something, brains. But to be enlightened we must be told how brains have states that may be called “beliefs” or “intentions.” What properties do beliefs, recollections, inferences, or intentions possess by which they may be identified according to their neurological features? And once we have hardened cognitions into states, will the states perform cooperatively thereafter, that is, in the course we perceive them to take as our understanding of each other as persons requires?
Cognitions have no properties of their own. Their only properties are those of “aboutness.” My essay tried to rehearse the aboutness of beliefs and intentions, etc.; it is not a property of which physical science presently is able to admit or, save a revolutionary reconception of what is now called “cognitive science,” is ever likely to admit. The ways in which we “understand one another as persons” is not convertible to the modes in which machines, such as fMRI scanners, inform us.12
Let me go on with “states” a bit. Adams does not tell us what he takes a “state” to be. Donald Davidson, with whom he appears to agree, gave us very little, but not nothing, to go on. Davidson observed that states are not causes, events are causes. The states of belief and of intention must turn into events before they acquire causal power. To the obvious question, how do states become events? Davidson gave only one clue: it is an “onslaught” that does it.13 He proclaimed onslaughts events, but Davidson did not tell us what causes onslaught events. One is likely to think, as I do, that if states of beliefs and intentions transition from states to events, it is when a person apprehends and believes that appropriate conditions are met and that person intends to do something (or nothing) about them. That is to say, since the actuating conditions in the world cannot automatically do anything, they must at a minimum be noticed (apprehended), and they, the actuating causes, must be classed as (precedently) cognitive in order to deem them responsive actions. The alternative open to Adams, of course, is to make “human action” an empty category, or rather a conceptually altered category, one belonging to “physical movements caused by other physical movements”: muscular movements caused by brain movements.
Adams asserts, “One need not consciously intend…” Can one unconsciously intend? But how do we find out about intentions if not consciously? Unconsciously? I can argue this, but, judging by his criticisms (and articles) he cannot. Well, he will say, we do not find out about intentions, we infer them. My examples and comments on the splendid variety of intentions and beliefs were made in order to show they are not, as things stand, construable as states. Adams gratuitously (or by fiat) makes them into links of a causal chain consisting of states, some conscious, others not. Can you imagine this long chain transmogrifying as a unit, or as intelligibly remaining that chain, when observed (and possibly unobserved) conditions of heart or of circumstance change? Without evidence or logic, the logic imposed by a credible theory, Adams can construct anything in the brain, but he can explain nothing in our world.
I might as well get to the heart of our deepest philosophical divergence rather than go on with others that merely reflect it. Adams puts it this way:
I am still unsure what replacement Cody has in mind for folk psychology and cognitive science. And if he does not have a replacement in mind, then these are still the only game in town for understanding purposive behavior of cognitive agents.
Therefore what? Are we to accept the logic, methods, concepts, and ontology prescribed by the only game in town for no better reason that that it is the only game in town? Is this what “inference to the best explanation” entails?14 I count this madness!
Frederick Adams is Professor of Cognitive Science & Philosophy at the University of Delaware.
Arthur Cody is an American philosopher.
- David Chalmers, “Facing Up to the Problem of Consciousness,” Journal of Consciousness Studies 2, no. 3 (1995): 200–19. ↩
- Fred Adams, (Forthcoming) “Cognition Wars,” in Studies in History and Philosophy of Science, A. Manafu, ed.; Fred Adams and Rebecca Garrison, “The Mark of the Cognitive,” Minds & Machines 23 (2013): 339–52. ↩
- Donald Davidson, Actions, Reasons, and Causes," in Essays on Actions and Events (Oxford: Clarendon Press, 1980), 4–13. ↩
- Marc Jeannerod, “The Representing Brain: Neural Correlates of Motor Intention and Imagery,” Behavioral and Brain Sciences 17, no. 2 (1994): 187. ↩
- Fred Adams, “The Causal Theory of Action Meets the Embodied Theory of Cognition.” in Causing Human Action: New Perspectives on the Causal Theory of Action, A. Buckareff and J. Aguilar, eds. (Cambridge, MA: MIT Press, 2010), 229–52. ↩
- Thomas Nagel, Why the Naturalist Neo-Darwinian Conception of Nature is Almost Certainly Wrong (Oxford: Oxford University Press, 2012). ↩
- Adams does not tell us what he means by “persons.” ↩
- Paul Churchland, Matter and Consciousness (New York: The MIT Press, 1998), 45. ↩
- You will get what Dennett means in this quotation chosen for its brevity:
Thinking about the postulated functions of the parts is making assumptions about the reasons for their presence, and this often permits one to make giant leaps of inference that finesse one’s ignorance of the underlying physics, or lower-level design elements of the object.Daniel Dennett, Darwin’s Dangerous Idea (New York: Simon and Shuster, 1995), 230. ↩
- The qualification “as we know it” allows for a future we cannot predict, but it is a denial that contemporary science is competent in this respect. ↩
- On where intentions are to be found, Adams, with his co-author, Kenneth Aizawa, proclaims “…your intention to pass the salt is entirely constituted by events and processes within your cranium.” And on processing they aver,
…cognitive processing involves non-derived content, that is, that cognitive states have the content they do in virtue of the satisfaction of certain naturalistic conditions that do not depend on the existence of other contentful, representational, or intentional states.It is perfectly logical on their view that a single intention may exist in the world, provided it is in the brain. Fredrick Adams and Kenneth Aisawa, “Why the Mind is Still in the Head,” in Cambridge Handbook on Situated Cognition, P. Robbins and M. Aydede, eds. (New York: Cambridge University Press, 2009), 78–95. ↩
- Adams instructively endnotes Marc Jeannerod (see his fourth endnote). Compare my remarks on “belief” imagined with two persons wired together with a similar thought experiment in Jeannerod:
When two agents socially interact with one another, representations are activated in their respective brains: each agent generates representations for his own (overt or covert) actions, and simultaneously simulates actions he observes from the other agent. In normal conditions, the existence of non-overlapping parts in the respectively activated networks allows each agent to discriminate what belongs to him from what belongs to the other. This process would thus be the basis for correctly attributing a representation (or the corresponding action) to the proper agent or, in other words, for answering the question of ‘Who’ is the author of an action.Marc Jeannerod, “Neural Simulation of Action: A Unifying Mechanism for Motor Cognition,” NeuroImage 14, S103–S109 (2001): S107. ↩
- Donald Davidson, “Actions, Reasons, and Causes” in Actions and Events (Oxford: Clarendon Press, 1980), 12. ↩
- This not uncommon idea these days is mistakenly attributed to what C. S. Lewis termed “abduction,” but I fear Lewis, a Pragmatist, would not approve. ↩