Mind is like a box of chocolate

I have no possible way of supporting the claim I am about to make, but I think Searle’s style as an arguer/philosopher is “thought-experimental” (can we say that for “full of thought-experiments”?) because of all the Wittgenstein he has read. I revel in not being able to imagine Searle arguing otherwise.

Anyway, here’s a good example. After having reduced the more audacious divisions of cognitive science to guff, Searle concludes:

“Suppose no one knew how clocks worked. Suppose it was frightfully difficult to figure out how they worked, because, though there were plenty around, no one knew how to build one, and efforts to figure out how they worked tended to destroy the clock. Now suppose a group of researchers said, ‘We will understand how clocks work if we design a machine that is functionally the equivalent of a clock, that keeps time just as well as a clock.’ So they designed an hour glass and claimed: ‘Now we understand how clocks work,’ or perhaps: ‘If only we could get the hour glass to be just as accurate as a clock we would at last understand how clocks work.’ Substitute ‘brain’ for ‘clock’ in this parable, and substitute ‘digital computer program’ for ‘hour glass’ and the notion of intelligence for the notion of keeping time and you have the contemporary situation in much (not all!) of artificial intelligence and cognitive science.” (Searle, 1984, p. 56)

The question which interests me here is: how do thought experiments work from argumentative viewpoint? Are they just a complicated set of hypothetical statements? Are they a story (i.e. a finely tuned string of analogies and metaphor and whatever else goes)? Are they a reductio ad absurdum?

In our case, quite intuitively and regardless of whether one already has an opinion about the subject at hand, one might think of a vast number of possible replies to Searle’s imaginary clock-less world. However, would not one miss the point if one started throwing rebuttals like: “Aha! But we do understand how the clock works if we can make something ‘functionally equivalent’! It’s just we do not understand it up to its more minute details”?

An interesting related article I might have something to say about one of these days:

Yourgau. W. (1964). On the logical status of so-called thought experiments. Proceedings of the Xth International Congress of the History of Science, Paris: Herman, 359–362.

***

Scarcely related, here’s another funny paragraph:

Because we do not understand the brain very well we are constantly tempted to use the latest technology
as a model for trying to understand it. In my childhood we were always assured that the brain was a telephone switchboard. ('What else could it be?') I was amused to see that Sherrington, the great British neuroscientist, thought that the brain worked like a telegraph system. Freud often compared the brain to hydraulic and electro-magnetic systems. Leibniz compared it to a mill, and I am told that some of the ancient Greeks thought the brain functions like a catapult. At present, obviously, the metaphor is the digital computer.

Searle is, I’d say, one of them “charming” ones.

Can computers think?

The label “thought experiment” may be applied fairly permissively to any kind of argument that involves a hypothetical situation. Often, however, imagine-that scenarios don’t rise to being called thought experiments unless they have famously proven a point, explained an idea or clarified certain conceptions. Schrödinger's cat is arguably one of the most famous (and famously spoofed) thought experiments ever devised.

	What I don't understand is just why he Can't be one or other, unquestionably. My future now hangs in between eigenstates. In one I'm enlightened, the other I ain't. If you understand, Cecil, then show me the way And rescue my psyche from quantum decay
	By the most knowledgeable Cecil Adams (here)

Searle’s Chinese Room Argument, which I will read in a bit, has the same flavor as Schrödinger's – at least in that it involves irresponsibly locking a human being in a room for some sick philosophical purposes. Unlike Schrödinger's, Searle’s thought experiment belongs to the philosophy of mind. Its conclusions, however, and the debate around, touch upon many other related fields such as cognitive sciences or philosophy of language. For instance, Stanford Encyclopedia credits Pat Hayes with having defined cognitive science as “the ongoing research project of refuting Searle's argument”.

What is interesting to follow, from an argumentative perspective, is how these hypothetical scenarios are devised and brought to bear on some conclusion or another. In principle – though this would mean simplifying things – they should be somehow reducible to certain strings of if’s and then’s. To go back to that bloke’s cat, the whole point of the experiment is to give weight to the following thesis: even though at some atomic level quantum superposition may be an edible, albeit bizarre, idea, if we transpose it to normal-sized objects it is downright outlandish. There’s no way we can conceive of it; the cat is either dead or alive despite any state of mind we might be able to put ourselves into. Searle’s thesis, as we will see, is even simpler. (The more rhetorically oriented might ask: why put everything into a thought experiment if your idea is that straightforward? or is that the actual idea? etc.)

Here’s the summary of the Chinese Room thought experiment:

Imagine a native English speaker who knows no Chinese locked in a room full of boxes of Chinese symbols (a data base) together with a book of instructions for manipulating the symbols (the program). Imagine that people outside the room send in other Chinese symbols which, unknown to the person in the room, are questions in Chinese (the input). And imagine that by following the instructions in the program the man in the room is able to pass out Chinese symbols which are correct answers to the questions (the output). The program enables the person in the room to pass the Turing Test[1] for understanding Chinese but he does not understand a word of Chinese.

A good point to start is to ask whom Searle is arguing against. So, okay okay, the person does not understand Chinese, but did anyone claim the contrary? In a sense, yes, although obviously before the CR argument no one worded it in this way. What Searle calls “strong AI” (strong artificial intelligence) is the conception that “any physical system whatever that had the right program with the right inputs and outputs would have a mind in exactly the same sense that you and I have mind” (1984, p. 28). Searle goes on to quote several strong-AI supporters, including one that said that the next generation of computers would be so intelligent that we would be lucky if they are willing to keep us around the house as household pets. To us now this is important because it adds a particular ingredient: it is not that such-and-such is not the case, but rather that such-and-such cannot be the case, despite of the formidable technical achievements the future might hold. Note that Searle often refers to his thought experiment as a “refutation”, a refutation, that is, of “strong AI”.

Now let us move outside the CR scenario for a few moments. The point would be this. Computer programs are, as the name tells, programs; they are sets of rules. We tell them what to do when they are fed, we fed them with what we have told them we would, and they do what we have told them to do. The fact that this food is 1’s and 0’s is a meaningless historical accident. They could have been !@$ and 00= since the two symbols don’t stand for anything, they are just opposite values (call them “true” or “false” if you like, but they don’t “describe truthfully”). Now, as with any set of such strings (as with any syntax or grammar, or program, or automata, have your pick), the symbols which make up the strings don’t mean anything, they don’t tell us anything about the world. It is in this sense that Searle repeatedly says that “minds are semantical”. We interpret symbols, whereas computers – just as our poor man moving Chinese pictograms to and fro – do not, need not and can not. We have content, computers are just formal.

Does this mean we cannot duplicate the brain? It does not, Searle is anxious to point out. There’s no absurdity involved: “if you can duplicate the causes, you can presumably duplicate the effects”. It is just that computer programs won’t achieve this. However good they’ll be at simulating the brain, they will never manage to duplicate it – since software cannot reach the “aboutness feature”.

Searle then invites us to check the validity of his line of reasoning. Pulling out (1) from the last lecture, he goes:

1. Brains cause minds
2. Syntax is not (sufficient for) semantics
3. Computer programs are entirely defined by their formal, or syntactical, structure.
4. Minds have contents; specifically, they have semantic contents.

Conclusion: No computer program by itself is sufficient to give a system a mind. Programs, in short, are not minds, and they are not by themselves sufficient for having minds.

Should we agree? To be continued…

---

[1] The “Turing Test” Searle mentions is an imaginary test given to any computer program in order to see if it is “human”. The test consists of tricking a person into thinking it is a he or a she. Some trace this idea back to Descartes’s Discourse on Method.

Brain picking

The purpose of this review is to follow some arguments advanced by John R. Searle in his 1984 Reith Lectures (Searle, 1984). It is not to criticize or analyze them, though a few comments here and there might leap out. I have chosen these texts because they are fairly straightforward as arguments and have been presented to an unspecialized audience. This kind of situation, so they say, should make standpoints and arguments more apparent.

[As a fun fact more than anything else, here’s a passage from the introduction: “The overriding theme of the series concerns the relationship of human beings to the rest of the universe”. Well… having remarked the audacity of this announcement, let’s start.]

The mind-body problem

The puzzle known mostly as “the mind-body problem”, but formulated over time as a clash between different –isms, can be summed up as follows: how do we put together the kind of (scientific) view we have of Nature with the kind of (philosophical) view we have of the Self? How is consciousness possible – to put it à la Kant – in a worldview where science sees motion and particles and many other unconscious phenomena?

This puzzle still puzzles - but it ought not to, Searle adds. We do not have to account for the relationship between two completely different kinds of things (mind/matter, thoughts/objects etc.) because the things are not that completely different. However we choose to philosophize about this relationship, we should produce something that also explains the following elements of our mental life: (1) consciousness – first and foremost, (2) intentionality – the feature of our mental states being about something, (3) subjectivity – each one’s access to his own mental states & each one’s lack of access to the mental states of the others, (4) mental causation – how do the two realms interact? “How can something mental make a physical difference?”

Searle’s answer starts with the following thesis: all our mental life is caused by processes going on in the brain (or, to be more precise, in the nervous system). In other words, brains cause minds. This point, without any further clarification, can hardly be doubted although it can hardly be used also; it is what ideas one further derives from it that make up its philosophical potential. For instance, Searle continues like this: the causation between the two realms (“mental life” vs. “processes in the brain”) has nothing contradictory or spectacular (and here it comes) because the two are different levels of representation. When we describe something as “neurons firing up and down” and “me feeling pain”, there’s no opposition because – although we’re describing the same thing – we’re describing it at different levels. The levels should not be conjoined but nor should they be kept apart. Just as we can’t “reach into this glass of water, pull out a molecule and say: ‘This one is wet’”, we should not say liquidity is something apart from the molecules that make up water. The “grey and white oatmeal-textured substance” causes all the conscious and unconscious elements of mental life – in short, brain cause minds and there’s nothing problematic about it. At micro-level, what is happening is describable in terms of processes, at macro-level, in terms of states of mind.

Now, granted, this was anticlimactic. What Searle did was to dissolve not to solve the puzzle. He basically said: “Wow, wow, wow, slow down. Aren’t we missing something?” And, according to him, we are. The mind and the body interact despite of being two different types of things, and we can describe them separately or as a whole, so long as we are very careful as to what we are doing. None of them is more or less real than the other; although a certain view of Mme Science, occupying herself solely with “objective” facts and not “subjective” states might prompt one to a certain division of labor.

If by naive physicalism we mean “the view that all that exists in the world are physical particles with their properties and relations” and by naive mentalism we mean “mental phenomena really exist”, then Searle’s thesis is this: the two are consistent with each other. And true.