L1.1 Lottie’s first lesson on concepts
(The casual reader of the thrilling adventures of the JugginsVerse will probably wish to skip ahead to here. Lottie herself almost certainly would do, too, to save herself 11 rather dry undergraduate-level lessons on the philosophy of thought and language.)
***
Lottie arrives for her morning lesson, or tutorial, as it is grandly called with her teacher, or tutor, as he calls himself, Hipparchus, the hippo.
“This week’s lessons are all going to be about concepts,” he says proudly. “I have 10 scintillating tutorials planned.” [In fact, there are 11 lessons, ending here, Ed.]
Lottie groans. “That sounds boring.”
“Well you shouldn’t have said ‘Concepts, shmoncepts’ last week when we were discussing the concept of justice. In saying that, you let me down, you let the rest of the Potting Shed community down, and you let yourself down.”
Lottie makes the sort of noise that only a teenager expressing complete contempt but not quite putting it into words can.
“Right, so let’s begin.”
Concepts are abstractions from whole thoughts or judgements.
Hipparchus plunges in with startling suddenness and Lottie struggles to follow his train of thought.
“It’s important that we start from the idea that a concept is an abstraction from a whole thought or judgement. For example, the judgement that teenagers lack concentration. This says that something is the case or is true. What is that? That teenagers lack concentration. Note that this whole thought or judgement has a content or meaning that might be deployed in other ways, in other attitudes. I may fear that teenagers lack concentration. You may deny it. But there is something in common to these cases namely what the different attitudes are ‘towards’. This is the content of the thought that teenagers lack concentration. This content is sometimes called a ‘proposition’ because it can be expressed in a whole sentence. It is the meaning of a whole sentence. Everyday or folk psychology, which is used to make sense of people, is thus sometimes called ‘propositional attitude psychology’.
Different sentences in different languages can express the same thought or proposition because they can say the same thing or have the same meaning. What they have in common is nothing to do with the symbols or the ‘shape’ of the sentence but rather something abstract. Not the ‘vehicle’ of the content, such as a particular sentence, but the content itself: what the sentence says. This is abstract.
Once we have a whole thought, we can also break it down into smaller parts. Our judgement – the one I suggested just now - is made true, if it is – and it is! – by the nature of teenagers. Of those teenagers, what it says is that they lack concentration. So it refers to the teenagers and it ‘predicates’ of them a lack of concentration. It may then be possible to break that down further into the notion of concentration and also its lack.
Such articulations into concepts of whole thoughts also suggests relations between different thoughts, such as the thought we have been discussing, and the thought that teenagers have a tough life, and also the thought that parakeets lack concentration. So another constraint on concepts that seems plausible is Gareth Evans’ Generality Constraint. Evans says:
It seems to me that there must be a sense in which thoughts are structured. The thought that John is happy has something in common with the thought that Harry is happy, and ... something in common with the thought that John is sad.... Thus, someone who thinks that John is happy and that Harry is happy exercises on two occasions the conceptual ability which we call ‘possessing the concept of happiness’. (Evans 1982: 100)
We thus see the thought that a is F as lying at the intersection of two series of thoughts: on the one hand, the series of thoughts that a is F, that b is F, that c is F... and, on the other hand, the series of thoughts that a is F, that a is G, that a is H. (ibid 104 fn21)
if a subject can be credited with the thought that a is F, then he must have the conceptual resources for entertaining the thought that a is G, for every property of being G of which he has a conception. This is the condition that I call ‘The Generality Constraint’. (Evans 1982, p. 104)”
“But is that obvious?” Lottie objects, finally catching her breath. “Might one be able to think that Derek the dodo is happy - he often seems so - but be incapable of thinking that Masongill the mole is. He never seems so! He always seems gloomy ‘of character’ as you might say! Perhaps the idea that happiness might apply to Masongill simply never occurs to me? And you wrote in my birthday card last year, stupidly, that ‘green ideas sleep furiously’ which made no sense! But your scarf can be green. It is! So not every concept applies to everything. Evans sounds silly!”
“The late Gareth Evans, who died sadly young, was a very gifted philosopher. But you are right in that a plausible version of the Generality Constraint should not require that every combination of name and predicate makes sense. It requires some generality, not every generality. So perhaps Evans is wrong in saying ‘for every property of being G’ in the quotation above. (In fairness to him, he says in a footnote ‘With a proviso about the categorical appropriateness of the predicates to the subjects’ (p101).) In your examples - aside from green ideas - you can surely entertain, if prompted, the idea of Masongill being happy if only to dismiss it as obviously false.
But imagine that someone were to insist that the idea of happiness could apply only to the King and no one else. Would not that make you doubt that it was happiness that they were really talking of? Perhaps they have a special concept of the-King-being-happy which is limited only to him. That would be a different concept which, for example, you could not express with a laugh, or self-ascribe, unless there is a regicide and you get lucky. Perhaps it is a vital concept in a kingdom with a cruel tyrant as king, not unlike a weather forecast. Sun is coming! The King is happy! But it is not happiness. And yet, on the other hand, as you say, the fact that a subject can understand my scarf being green cannot require that they must understand an idea being green. Not every predicate can be used of every ‘name’. Well done! [This topic returns later, Ed.]
Let us go back to the idea that this came from. Whole thoughts can be articulated into subsidiary aspects. These articulations of the whole thought into smaller parts are abstract or ‘abstracta’. They stand to the whole thought as pitch and tone stand to a musical note rather than as individual bricks stand to a whole wall. They cannot have independent existence. Rather, they are aspects of the structure of a whole thought that can be shared with other whole thoughts. For example, our judgement (above) has things in common with the very silly thought or judgement that teenagers possess concentration.
One reason to start like this is the claim made by the C19 philosopher Gottlieb Frege that a whole sentence or a whole thought is the smallest unit with which we can do anything. We need the whole of it before, for example, we can see whether it is true of the world – or teenagers.
Frege’s Context Principle: only in the context of a whole sentence does a word have meaning. Similarly, concepts exist only as abstractions from whole thoughts.”
“But wait a minute” interrupts Lottie. Why do we have to start with whole thoughts and think of concepts as abstractions from them. All that just sounds like a story but not real. Couldn’t we think of them as like bricks building up a whole wall?” Lottie kicks her football as though to imply that something like that is her standard of real existence.
“Well done, Little Lottie” – Lottie squirms with embarrassment – “that is a good question”
Concepts as illata
“So let’s see. I’ve said two things. That we start with whole thoughts. And that concepts are abstractions from them. Could concepts be abstractions from anything else? It is not obvious unless we follow the American philosopher Donald Davidson and say something like: it is only in the context of a whole language that a word has meaning. (I will talk about him tomorrow.) If we do that, then we will think of thoughts as abstractions from whole systems of communication or perhaps conversations. And we will think of concepts as yet further abstractions. This would not address your worry.
So let’s change the other claim. Perhaps concepts are not abstract, or abstracta, but entities that can exist in isolation and have, let’s say, causal powers. Like an electron. We cannot see an electron but we think it is a real and non-abstract entity. Perhaps concepts are like that. Such posited entities are called ‘illata’ following the introduction of the distinction of illata and abstracta by Hans Reichenbach (and used more recently by Daniel Dennett [Hipparchus returns briefly to Dennett here]). Are concepts illata?
So let’s think what concepts would be if they were illata. What shall we call them? Well concepts are not very like bricks or footballs. But they connect in some way we are trying to get clear to meaning. Let’s call them, for now, ‘magical runes’ or MRs for short. (I do not think the connection between concepts and thoughts or meanings is at all mysterious but I think some philosophers in their effort to avoid mystery make mystery. As we will see.)
There are philosophers who think this way. Jerry Fodor thinks that MRs are neural entities with causal properties. But just as computers are designed so that programs can written which can run on machines because the internal states of the machines have causal properties that mirror the logical structure of programs, so MRs have causal properties that mirror the rational structure of meaning or thought. According to Fodor MRs are ‘mental representations’ which are also neurological states.
The computer offers him this analogy. But if you think about it, it is not clear that computer outputs could be about anything without creatures like us to interpret them that way. We programme them to give outputs apparently in English. But it is up to us to link the words on a computer screen to, for example, real teenagers not concentrating. How could bricks, footballs or MRs be about anything? How can there be aboutness in the world if we start this way?”
“Well” Lottie replies “even you told me that smoke means fire! Isn’t meaning just there in the world?”
“That’s a good point, Lottie. You do seem to be thinking today. And concentrating.” Lottie sighs, looks bored, kicks her chair leg. But perhaps she is a little pleased too.
“There are indeed regular causal connections in the world that are sometimes called ‘factive meaning’. But that really means that creatures like us, on seeing smoke, can, and should, think fire. We are the ones injecting the meaning. A bird falling dead to the ground does not in itself mean anything except to those who might worry about what killed it. Fodor takes a related line. You will have to think whether it sounds plausible. He thinks that, if MRs are caused by specific sorts of things, then they mean their cause. So let’s think about that. Suppose that you are lying in your bunk in the rafters of the Potting Shed where no one else sleeps. Quite spooky when you think about it! (Have you heard of the Bogley Strangler?) Anyway, suppose that the wood of this old shed creaks and the light from my oil light 3 flights of ladders below makes a sudden shadow up where you are. What will you think?”
“That there’s a ghost, obvs - or a strangler now, thanks!”
“But given my story, what caused you to think that? A ghost?”
“No obviously. You’ve just said. A creak and a flicker.”
“So take your time and think about what we have just been talking about.”
“Oh, so my magic rune was caused by one thing but I thought another. If Mr Foe-Door was right, I would have thought creaky flicker.”
“Well done. It seems hard to know how, according to the theory, anyone could think false thoughts. A plump horse seen at night might prompt the mistaken thought of a cow, but not on a simple causal theory of meaning because the thought, the MR, will mean whatever causes it. So if Fodor explains the meaning or content carried by some neurological MR through what causes it, how can he avoid it meaning a disjunction of everything that can cause it? One might want some specific MR to mean cow - to encode the content cow - but if it is also caused by plump horses when glimpsed in the dark - corresponding to seeing a horse and thinking it a cow - why does that MR not mean cow OR plump horse at night? How can the correct application of a mental representation be explained just using causal mechanisms?
In fact, Professor Fodor was well aware of this problem, which he called the ‘Disjunction Problem’ and offered a solution. He proposed that some causal connections might be more fundamental than others. Others only exist because of the first sort. So the MR we have mentioned is only caused by plump horses at night because it is also caused by cows in the day, but not vice versa. In a possible world with no cows but with plump horses, that MR would not ever be ‘tokened’, or bought about, but it would be in a world with cows but without plump horses. This is his Asymmetric Dependence theory. If the causal connection between a plump horse and an MR would not have existed without a similar link from cows to that MR, but not vice versa, then the MR means cow not plump horse. ”
“Can I say I’m getting a bit bored now?”
“Yes. We will stop soon for today. Let me just give you my own opinion. Fodor offers a very limited solution to a problem which is more serious than he thinks. Despite the addition of the idea of causal connections that depend asymmetrically on more fundamental connections, he has to construct a complicated causal mechanism which will track meanings. But he has to ensure that no fundamental causal connection ever corresponds to falsity. And machines simply are disposed to break down. Causal dispositions can never set the standard for correct thinking. In saying which causal connections depend on which others - in speculating what would happen in other non-actual hypothetical possible worlds - Fodor must not rely on us already having decided what an MR encodes (eg cow rather than plump horse). It would be a cheat in this context to judge which causal mechanisms are fundamental on the basis of a prior understanding of meaning. (In this respect, computers are quite different. We programme them so as to follow meanings, thus presupposing meaning.) But Fodor has given up the right to that check. He is on his own, in a sea of mere causes.”
“Football! Now!”
“Very well. But the moral of today’s lesson is that reductionism of concepts to bricks, footballs, MRs or illata faces a grave problem concerning the correctness or normativity of thought and hence concepts.
For your homework look up Ruth Garret Milligan’s ideas that concepts are encoded in some biological functions and see what you think. Biological functions look to have one foot in the realm of purposes but as Darwin taught us another foot in brute causal history. So ‘teleosemanics’, as this approach is called, looks to have a better chance of explaining the correct application, or the normativity, of concepts while still being grounded in nature.”
Without waiting, Lottie runs off to get some fresh air.