Part 3b. Non-representationalists Are From…

In Part 3a, I mentioned that both representationalists and non-representationalists address HOW questions about our mental life. Both give causal explanations for how we come to think, believe, and understand things about the outside world. And when they’re feeling ambitious, they’ll also both try to explain how feelings, sensations, and other conscious experiences come about. But there are distinct differences in the way the two groups answer those questions. What follows is a broad, somewhat stereotyped overview of these differences. In this post I'll explain how the two accounts differ with regard to what is being caused; in the next one I'll explain how they differ with regard to what counts as a cause.

Note that here I’m just generalizing; in Part 4 I’ll get into a lot more specific detail to back up what I’m saying. I’ll be sure to include examples of what Andy Clark calls “representation-hungry problems” — specifically, “cases that involve reasoning about absent, nonexistent, or counterfactual states of affairs” (Being There, page 167). For now I'm steering clear of feelings and sensations… apologies to qualiaphiles for that. I promise I'll get there eventually.

What is being caused:
In Representationalist accounts (R): context-independent, free-standing capacities that many people may have in common

Examples:

1. Knowing (or understanding) that a robin is a type of bird (note that I can know this whether or not a robin is right in front of me, and many different individuals may share this same piece of knowledge)
2. Thinking about a party that has not happened yet
3. Believing that there is milk in the refrigerator
   
   And, moving toward the more generalized, abstract end of the spectrum:
4. Having the concept “dog”
5. Reasoning about absent (or nonexistent or counterfactual) states of affairs
   

In Non-representationalist accounts (N): specific behaviors and experiences that actually occur, in real time, in response to particular internal or external stimuli

Examples:

1. Upon being asked the question “What type of animal is a robin?” answering “A bird”; upon seeing a bird with a red chest, pointing to it and saying “Oh, look — a robin”; and so forth
2. Upon being asked, “What are you doing next Saturday,” answering “Going to my friend’s birthday party”; in the process of writing a to-do list, having the thought (i.e., internally hearing the words) “I should get a present for Karen,” then writing down “Karen present” on the list; and so forth
3. In the process of writing a grocery list based on a cookbook recipe, seeing “1 cup milk” in the ingredients and not writing “milk” on the grocery list; after pouring a bowl of cereal, opening the refrigerator door, moving items around, experiencing a sense of surprise and irritation, and then asking one’s spouse, “Didn’t we have some milk in here?”
4. All one’s behaviors (including verbal behaviors) that relate to dogs, including using and responding to the word “dog,” pointing to dogs, talking about specific breeds of dogs, and so forth. (I might also mention here what does not happen; e.g., upon being told “Jim and I just adopted a dog,” one does not respond, “You adopted a what? What’s that?”)
5. An enormous range of specific behaviors and experiences, the great majority involving verbal behavior — either external (in speech) or internal (in thoughts).
   

Now, at first glance, the distinction outlined above may seem to be trivial. If so, I encourage you to pause and give it some more serious thought. Can you think of ways in which an analysis that attempts to explain general capacities (as in R) might tend to proceed differently from one that limits its focus strictly to specific behaviors and experiences (as in N)? I hope everyone can agree that in the end, a satisfactory account of our mental life need only explain specific things that actually happen. That is, if we came upon an analysis that could explain the full range of human experiences and behaviors (and I mean the FULL range, with no omissions), there’s nothing further we could ask of it. To take one isolated piece, imagine that someone came up with an exhaustive causal analysis explaining every single dog-related behavior and experience — all possible talk about dogs, thoughts about dogs, interactions with dogs and pictures of dogs, etc. Everyone agreed that there was no dog-related experience or behavior that was unaccounted for. If that analysis did not include any references to concepts, would anyone be entitled to complain, “Well, sure you explained everything I could ever do or say or feel related to dogs — including everything I could say or think about ‘dog’ being a concept — but you still haven’t explained how I have the concept DOG!” I think not. If you think so, please argue with me here, because this is a critical issue to resolve.

More likely, I think, is that people are making two assumptions:

1. Once we have adequate representationalist accounts, explaining more general capacities, those accounts will serve to explain all the relevant specific behaviors and experiences.
2. Any account that explains all the relevant specific behaviors and experiences will of necessity include references to mental representations.

Obvious though these ideas may seem, they are assumptions and not proven facts. Moreover, I will argue that they are insidious, dangerous assumptions that get philosophers of mind (and others) off on the wrong track before they even begin.