Why doesn't hard content externalism lead to behaviorism?
A little background in what's meant by "content externalism:"
"Externalism—whether of content or vehicle—is often viewed as a peculiar eruption of late twentieth-century analytic philosophy. This, however, is historically myopic. Cognate views with fairly clear externalist credentials have been advanced and defended much earlier than this, and in some of the defining works of twentieth-century philosophy (see Rowlands 2003: Chapters 3&4). Wittgenstein (1953) argued that meaning something by a sign does not consist in an inner state or process. Rather, the possibility of such meaning depends essentially on the existence of a custom or practice of using a sign in given way, and the mastery of a technique—i.e., the ability to adjust one’s use of the sign to bring in into accordance with this custom. According to the dominant—community—interpretation of the notion of a custom, this entails that the possibility of meaning something by a sign is dependent on the existence of a practice external to the individual meaner, and that what this individual means by a sign on any given occasion depends, at least in part, on this external practice. As we shall see, this social version of content externalism was later developed, powerfully, by Burge (1979, 1986)." -- SEP on externalism.
The prevailing form of mental content externalism only says that some content is external. A more rare and somewhat confused version says that all mental content is external.
I think this brand of externalism leads to behaviorism and a pending collapse in meaning of any kind anywhere. How can this be avoided?
"Externalism—whether of content or vehicle—is often viewed as a peculiar eruption of late twentieth-century analytic philosophy. This, however, is historically myopic. Cognate views with fairly clear externalist credentials have been advanced and defended much earlier than this, and in some of the defining works of twentieth-century philosophy (see Rowlands 2003: Chapters 3&4). Wittgenstein (1953) argued that meaning something by a sign does not consist in an inner state or process. Rather, the possibility of such meaning depends essentially on the existence of a custom or practice of using a sign in given way, and the mastery of a technique—i.e., the ability to adjust one’s use of the sign to bring in into accordance with this custom. According to the dominant—community—interpretation of the notion of a custom, this entails that the possibility of meaning something by a sign is dependent on the existence of a practice external to the individual meaner, and that what this individual means by a sign on any given occasion depends, at least in part, on this external practice. As we shall see, this social version of content externalism was later developed, powerfully, by Burge (1979, 1986)." -- SEP on externalism.
The prevailing form of mental content externalism only says that some content is external. A more rare and somewhat confused version says that all mental content is external.
I think this brand of externalism leads to behaviorism and a pending collapse in meaning of any kind anywhere. How can this be avoided?
Comments (61)
"Behaviorism
Behaviorism was a movement in psychology and philosophy that emphasized the outward behavioral aspects of thought and dismissed the inward experiential, and sometimes the inner procedural, aspects as well; a movement harking back to the methodological proposals of John B. Watson, who coined the name. Watson’s 1913 manifesto proposed abandoning Introspectionist attempts to make consciousness a subject of experimental investigation to focus instead on behavioral manifestations of intelligence. B. F. Skinner later hardened behaviorist strictures to exclude inner physiological processes along with inward experiences as items of legitimate psychological concern. Consequently, the successful “cognitive revolution” of the nineteen sixties styled itself a revolt against behaviorism even though the computational processes cognitivism hypothesized would be public and objective — not the sort of private subjective processes Watson banned. Consequently (and ironically), would-be-scientific champions of consciousness now indict cognitivism for its “behavioristic” neglect of inward experience.
The enduring philosophical interest of behaviorism concerns this methodological challenge to the scientific bona fides of consciousness (on behalf of empiricism) and, connectedly (in accord with materialism), its challenge to the supposed metaphysical inwardness, or subjectivity, of thought. Although behaviorism as an avowed movement may have few remaining advocates, various practices and trends in psychology and philosophy may still usefully be styled “behavioristic”. As long as experimental rigor in psychology is held to require “operationalization” of variables, behaviorism’s methodological mark remains. Recent attempts to revive doctrines of “ontological subjectivity” (Searle 1992) in philosophy and bring “consciousness research” under the aegis of Cognitive Science (see Horgan 1994) point up the continuing relevance of behaviorism’s metaphysical and methodological challenges." --IEP entry on behaviorism.
Not that we should assume there would automatically be some awful problem if it did. But,
Quoting bongo fury
It's a game of pretend. There won't be any fact of the matter of exactly what anyone was pretending. How could there?
That's why I say hard behaviorism collapses all meaning, including the theory itself. Is this wrong?
So you're saying externalism does lead to behaviourism? Contrary to your thread title?
I'm looking for a reason to say it doesn't.
It (meaning, mental content, what have you) is a game of pretending that words and pictures refer to things (externally, not in the head). We're just second guessing each other's fantasies about the reference. We can't use observed behaviour to justify any particular fantasy.
But the kind of behaviorism I'm thinking of doesn't allow that there is any kind of referencing involved in communication.
Speech itself is merely behavior. Some behaviors are repeated because they bring about something positive.
Wouldn't a sort of "hard" externalism seem to mesh with that kind of behaviorism?
Fair enough. By justify I meant choose. If there were some fact of the matter of which symbols mapped to which things then behaviourism wouldn't be wrong to reduce or replace that mapping (the reference that 'happened') with some description of behaviour. But according to Quine there is always a choice of mappings. We never know for sure what 'happened'. We have to play the game of pretend, and allow for, second guess, multiple hypotheses of what was pretended.
Which is how a 'hard' externalism about the mappings, i.e their not being in the head, can fail to mesh with behaviourism.
So for Quine, it's quite possible that you understand what I mean in the sense that we're consciously referencing the same stuff, we just don't for sure that this is happening?
Why would you be expecting Quine to be getting into our heads?
Quoting frank
Why would you be expecting Quine to regard this question as a matter of fact?
If we're planning to do some scientific research on what's in our heads, I think he would say we shouldn't do that due to the unavailability of facts. That doesn't mean he ruled out subjectivity, though, right? In our off hours away from the lab, we could say that we might be thinking about the same thing?
This sounds wrong. Not sure what you meant.
Quoting frank
I don't know what "that" is, but he was usually happy to rule out such notions. Are you surprised?
Quoting frank
Again, why would you be expecting Quine to be internalist (getting into our heads) and determinatist (matter of fact) about reference?
I've actually always assumed he did rule out subjectivity, but I'm rethinking it, actually based on what you said.
There's a difference between saying that we can't know whether we're thinking of the same thing, and ruling out subjectivity. Why should I take Quine as saying the latter?
Then ignore what I said.
Quoting frank
Because the word is carrying too much baggage.
Quoting frank
Can't because it's hard to determine :down: , or can't because it's indeterminate :up: ?
And why 'think' when this was about externalism?
The OP makes it pretty clear that we aren't talking specifically about Quine. You brought him up and now ditch the effort. That's weird.
Quoting bongo fury
All sorts of things are indeterminate for all practical purposes for me with my poor 21st technology. I don't therefore rule out those things.
I think we're done?
Brains might sync as people interact.
Yes, to offer one straight answer to your question.
Quoting frank
How so?
Quoting frank
So, contra Quine. As long as you see that.
Quoting frank
Okely.
Quoting frank
Of course. Cool stuff.
Thanks. I appreciate it. I didn't understand it, but I'll return to it shortly.
That's a hard solution to avoid as "strong" content externalism just seems like a version of behaviorism.
I suppose that one can argue that the content of our perceptions are veridical only when the external content actually coincides with what we imagine. That is if the cow we think we see is actually a cow. So on this view, thinking that you see a cow that turns out to be sheep would be equivalent to saying that the content in your head is false. So even if you think you saw a cow, your thinking that you say it is not content after all.
But this just makes no sense. It's an arbitrary stipulation on what content is, which leaves out almost everything.
If, however, you use content externalism in a much less radical manner, then perhaps it can be articulated a little.
What I don't understand is how to make sense of a strong version of behaviorism. Doesn't it rule out sense in communication?
Quoting Manuel
But there really isn't anything newsworthy about weak content externalism, is there?
It rules out almost everything. A human being is reduced to a stimulus-reacting nothing. You wouldn't be able to tell a human from an amoeba under strong behaviorism. So it's not even good science, unless it's radically altered.
Quoting frank
Nope. Not that I can see. Yet I see more papers talking about externalism of all varieties than internalism, which is coherent and looks to me to be scientific. Everybody's internalist when it comes to other animals, we all assume they have an intrinsic nature such that a dog will behave dog-like and not chimp -like, etc.
But when it comes to human beings, internalism becomes this mysterious magic stuff to some. This is strange to me.
A human being will react to poem, an amoeba won't. What's unscientific about that?
Under strong behaviorism, how would you know that?
Any movement made by the amoeba can be taken as sign that it is reacting to the poem.
But of course amoeba's don't react to poems and humans do, as you point out. So you need to postulate an innate mechanism that allows human beings to react to poems that amoeba's lack.
If it's based on behavior alone, without assuming an innate structure to each creature, I don't see how one could make that distinction, unless you believe there is no difference between creatures.
How do you think we found anything out about amoebae then? Theories all rely on repeatability. If the amoeba repeatedly responds to a poem in some predictable way (but not to some control, like a random collection of words), then we've got everything we thought about amoebae pretty wrong haven't we?
Quoting Manuel
I think you've misunderstood what behaviourism is. None of its proponents suggest that there is no mechanism, that the brain's just a non-functioning blob.
Which is why I used the term "strong" behaviorism, one which would do away with any innate mechanism. I agree that I doubt any proponent today would hold such a view.
Less radical versions are fine and necessary.
That's what I thought. Thanks!
Quoting Manuel
So content internalism is satisfactory to the average scientist, but philosophers insist on externalism? Why? Because they think they'll be conceding some religious view otherwise?
Absolutely no one, ever, would hold such a view. There's little point in constructing arguments against views that no one holds is there?
It's difficult to generalize in this respect, there are philosophers who accept internalism without much problems such as Strawson, Chomsky, Haack, McGinn, etc.
I suspect some philosophers think that by sticking to externalism, they're putting aside spooky stuff like experience (looking at the blue of the sky, or explicitly thinking about a beach, etc.) and then stick to things that are publicly observable and hence be "scientific". It's hard to say.
I don't agree. Putnam, for example, suggested the thought experiment of a brain-in-vats as an exercise. Nobody literally believes we are brain in vats, though some believe in the simulation hypothesis. It's still useful to look at what extreme ideas would look like.
I thought the BIV was useless too, so that's not a persuasive argument. Perhaps highlight some of the conclusion you find useful?
I think that in principle, if we knew enough about the brain we wouldn't need an external world. You would just need to stimulate the appropriate are of the brain to recreate an experience which would be indistinguishable from one in ordinary life.
Granted, this scenario is science fiction, but I think the principle is correct. The main point I take from it is that the world is not necessary to explain our experience.
But as you hinted at, what's useful depends on the person involved. I don't find the "do machines think?" question to be useful, but clearly, other people do.
I understand this was intended as an example of the sort of conclusion you find useful, so my question is a bit off-topic, but, very briefly, if we could do without the external world, from where is the stimulation to the appropriate area of the brain coming? Whence the electricity to power it, the mechanism to convert it, the materials from which this 'stimulator' is made?
From a super scientist or an alien species, or God (for those who believe in Him/Her/It). Yes, you can always ask that where is that super scientist or alien species located? It must be in a world in which all these things happen. Perhaps or maybe we lack the imagination to think of how such a situation in which a brain in a vat could be carried out. All I'm saying is that there need be nothing in the world to which our representations are about.
One last question, the physics community does a pretty good job of at least trying to communicate with the nonacademic world. Do you think the philosophy world could do more of that?
I think so. Look at Bryan Magee's work, he did an excellent job in explaining philosophy.
But as far as the modern technical stuff goes, I'm not so sure. The philosophy world could surely use better communicators for the lay person. I mean Dennett's lectures are fun and Chalmers has appeared in popular science shows. But a lot of the interesting details need a bit more clarity, it's just too much jargon at this point.
And the books are really expensive. I'll check out Magee, thanks.
Yes, stupidly expensive. :groan:
Quoting frank
:up:
Isn't that the conclusion you drew from the thought experiment? You're now using it as a premise. That doesn't sound like the thought experiment has done anything.
I think that's accurate. The things we see in manifest reality: trees, rivers, apples and so on, need not be aspects of the world. They happen to be so recognized by virtue of the cognitive capacities we have.
I find this idea useful. If you don't or you think I'm wrong, that's fine. We initially were speaking about the usefulness of extreme thought experiments. It seems that you think you don't find them useful, or maybe I misunderstood.
Either way usefulness is subject to a person's preferences. As you said, this topic is now removed from the OP.
Well that's why I asked where the stimulation comes from. Because if it comes from 'the world', then the thought experiment doesn't suggest we don't need the world, does it?
Quoting Manuel
Maybe. It seems related to me - externalism, stimuli (as in stimuli-response)...
I think the view you're advocating is internalism: that the world stimulates the brain to form representations.
Fair enough.
Quoting Isaac
The stimulation could come from the world or it could come from a brain in a vat. We assume, very plausibly, that these come from the world. But they could also come from a brain in a vat, given a genius scientist.
If one is interested in philosophy of mind, I think that it makes sense to see what happens in experience. After all, very similar experiences could have different causes. For example a dog starts barking when they look in a mirror, assuming they see another dog.
A moth will fly to lamp and kill itself, confusing it for the moon.
What I think happens in these cases is that the stimulus gets interpreted as belonging to something in the world (another dog, the moon, etc.). I don't think it's unreasonable to suppose that if we knew enough about moths or dogs, we could induce these experiences in a lab.
Likewise, for us, when we confuse one object for another. We interpret a stimulus in a certain manner, regardless of the source.
My aim was to inquire rather than advocate. I don't mean to disrupt your thread if it's off topic...
Quoting Manuel
You mean the brain stimulates itself? It certainly happens. Real experiments show that. You were talking about external stimulation earlier, that's all.
Quoting Manuel
But it does belong to something in the world, a mirror and a lamp respectively. Mirrors can cause dogs to bark, lamps can cause moths to fly to them. What does it matter that they're not other dogs or moons? Maybe the dog 'thinks' it's another dog, but the moth doesn't 'think' it's the moon, it doesn't 'think' anything, it hasn't got the substrate in which 'thinking' takes place. Yet, if its behaviour, including errors can be modelled in the same way as the dog's, then on what grounds do we say the dog 'thinks' there's another dog? These errors of modelling (the dog's, the moth's) don't make the sources of the data internal, they're about generating appropriate responses. If the model generates an appropriate response, then in what way is it an error of interpretation?
I don't believe I said it was an error of interpretation. We would say that the moth made a mistake, on the assumption that living creatures generally speaking, don't commit suicide.
Yes, "the model generates an appropriate response...". I agree here.
Isn't the model internal?
That's kind of rare on this forum. :grin:
Quoting Isaac
Not at all. I'm interested in your conversation.
All I ever do. Admittedly sometimes I inquire quite obdurately...
Quoting Manuel
A mistake in behaviour though, no? It ought not have flown into the lamp (to its death), the result of any modal of lamp/world should have had it remain alive at the very least. Soft behaviourism?
How could we understand such an error, in a functionalist sense, without an external world being one way such that some model of it can be another?
Quoting Manuel
So stimuli-response then...?
Quoting Manuel
Yes, I believe it has to be by definition. In my early days I wrote from a behaviourist perspective, it was the advances in computational cognitive science that changed my mind. The degree to which we can accurately assume a model-dependant cognition. To declare anything to be a model it has to represent something else (otherwise it's the thing itself, not a model), so we already have this division simply by the definition of 'model'. When talking about minds and worlds, 'internal' is just a marker for 'mind' - one half of the division we set up definitionally.
So, relating back to the thread... externalism would be simply mistaken from a grammatical point of view. That which the model is 'internal' to it is internal to by definition - the model, not the modelled. We might say that models span more than one mind, but we'd be mistaken to say that makes them external in the same way that the world they're modelling is, otherwise they are 'that which is modelled', and they're not just by definition.
In a sense, @frank's right. This kind of externalism does lead to the sort of extreme behaviourism that you theoretically posit (though no-one actually believes it). It removes models entirely and says that all there is is that which is modelled. If that were the case, there'd be no errors, the moth would have meant to fly into the lamp. Since talk of 'errors' and 'intentions' seems so useful, I can't see the utility of a system which would exclude them.
But to be fair, I've never fully understood externalism, and hard behaviourism doesn't exist outside of one's imagination, so it's possible that I've missed the point entirely...
Loosely speaking, in a model in which all sensations (stimulus, sense data, etc.) of the type X are interpreted as the moon, things that resemble X close enough, would lead the moth to act as if the X is the moon.
Of course, the moon could not be out that night due to cloudy weather or it could cease to exist. The moth would still interpret anything that causes X as the moon. Something like that.
Quoting Isaac
That's fine.
Quoting Isaac
There are several versions of it, linguistic, perceptual, etc. I don't know what's supposed to be revealed in most versions of it that I can recall.
I think that for phil. of mind, what matters is how the relevant creatures acts (behaves, responds, interprets) given sensory data.
But to be fair, the OP quotes discussion of a "social version" in which
Quoting SEP
I.e., cutting out, at least in part, the middle man in this too-universally-accepted picture:
A step which (taken in full, and ironically further than Ockham) Wikipedia calls "childish", but is arguably the natural perspective of any child or other socially-embedded language-learning machine. E.g., "what does this symbol refer to, in this language game?"
All of that seems to require an external world. The stimulus or sense data of type X must come from outside of the model.
Quoting bongo fury
I see, thanks. That makes sense. It seems little more than a context dependant frame. In linguistics it's not useful to consider the thought node at all (since a single individual can't have a meaningful symbol/referent link PLA etc). But in a different context - say psychology or neuroscience, it's simply a cold hard fact that the sign is modelled by the brain so as to be attached to a referent. The mid-stage is there whether we like it or not.
I haven't been as clear as I would've liked.
When I say there doesn't need to be a world, I mean the world we take for granted so you look outside the window and you see all that you see: cars, sidewalk, people, trees, etc.
A moth, granted, wouldn't see these things. A moth would see whatever it is that moths see: ultraviolet light and everything else they interact with.
But it could be, in principle at least, stimulated in a lab such that the world in this sense (described above) isn't necessary for the moth to experience (or react to) its "world". It would all be a stimulation in the moths brain.
Yes, the stimulation is external. But in this case the stimulation is not caused by something in the world (the moon), it would be caused by whatever electrical signals lead the moth to behave as if a moon existed.
There's also knowledge and justification externalism. That might have more to do with reference than mental content externalism.
To be fair.
Yes. I get what you're saying, I think.
What I take issue with is the (what I see as) false distinction between the data coming from 'the moon' and the data coming from an electronic signalling device. Neither has any better claim the be the 'actual' moon. We would not have been deceived if we found out that, rather than a lump of rock, our model of the moon modelled electrode signals. More it would be the case that electrode signals are what lumps of rock are.
There is no moon, there are only hidden states which we model as being the moon, the model just is a connection between signal and response (a dynamic and interactive model, mind, not a passive, fixed one).
This is why I don't get what the BIV gives us of interest. It says that the hidden states we model as 'the moon' might be electrodes. Well, they might, but 'electrodes' are themselves just a model of some hidden states. No deception has taken place, we always assume that our models might be slightly (or even massively) off sometimes, to not assume so is to believe one is always right.
Ah.
You already believe something akin to BIV when you say "there is no moon... only hidden states which we model as being the moon...". We are liable to find more "hidden states" the more we discover about the brain.
So it makes sense why such a thought experiment would not be appealing to you. Others might have Sellars' distinction in mind in terms of thinking about the manifest image (the world as we experience it) and the scientific image (the the world as it is absent people).
I think the impact of the BIV depends on how you think about this distinction.
No doubt. It seems there's two levels on which one could interpret it.
One is saying that the form of the hidden state we currently share a model of might be something radically other than we think (electrodes, Cartesian demons, etc)... For me, this misunderstands what it is for an object to 'exist'.
The other, which I prefer, is to say that there is no 'something' at all that is not a model of the hidden states outside our Markov blanket. So it makes no sense to say that our world might be something other than it seems, some way it seems is the sum total of all things, there is nothing 'other'. To be a 'thing' is to be some way it seems.
Attached directly? Sure. So,
Quoting Isaac
Apparently not. Not in the model.
Even in contexts of individual judgement, a speaker (game-player) may infer (model) the individual occasion of reference as a relation from the token (utterance) of the symbol to whatever the inferred referent(s), without psychologizing per the picture.
I only meant that it cannot be eliminated materialistically. It may be of no consequence in terms of systems analysis (though I'd argue it is, but later perhaps). For now, all I'm saying is that in certain contexts it can't be removed. For example, someone with damage to Broca's region will show a noticeably different relation between sign and referent than they would prior to that damage. To take account of that, we have to have a mid-stage to our model.
I would much rather you had waited five hours anyway.
Why people have to reply within five minutes I never understand. The result is rarely worth it.
Suggestion.
Sounds good. :up:
What an odd criticism. How do you suppose my seminars would have proceeded if every question was only answered following a five hour pause? Discussions with colleagues likewise. I was in a meeting only last week where I was being quizzed on a matter significantly more important and complex than our current discussion topic, yet I was expected to answer each question after only a few moments pause to gather my thoughts.
If seminars, research discussions and consultancy meetings can all proceed at pace, it seems odd to think that the little parlour game we play here requires five hours of deliberation before making the next move.
I was about ready to back-peddle on that, reminded of this,
Quoting Quine, Pursuit of Truth
So it's gratifying to find Kripke agreeing that,
Quoting Kripke p57
I'm not sure whether Kripke thinks that Quine would be happy with that way of regarding. I do. I think.
Quine shows that the human (or linguistic animal) condition is to have to hypothesise about a reference relation that is inherently indeterminate when conceived externally (cutting out the middle man).
Kripkenstein shows that the same indeterminacy arises for the relation conceived internally (per the diagram):
Quoting Kripke p57
42
The answer to your question is that there is a desire for knowledge to take the place of acknowledging the other person. But the above explanation takes the same desire for looking inside someone to create a picture of how we express ourselves only moving "meaning" to something external.
"Behaviorism.. emphasized the outward behavioral aspects of thought and dismissed the inward experiential."
I could go down a rabbit-hole to try to correct everything wrong with these pictures of our relationship with language, only to say that Wittgenstein was trying to remove our fixation with "meaning" being a thing either inside or outside--our desire to know that; internally, to actually salvage our ability to be individual, personal, secret; and outside, to show how we are responsible for what we say and do.
Quoting frank
I think the sense of loss is the continuing desire for knowledge to contain our entire relationship to the other. The most powerful image I've come across is that we do not know the other person's pain, their pain makes a claim on us that we either accept or ignore. Our relationship to the world is more than (just) knowledge.
Quoting frank
So we continue to search for a way (as it were into, or past, the other) that does not involve our actually engaging the other, and, in doing so, turn their actions into movements, their words into sounds. We are entirely separate, but still capable of expression, response; though our impulse is to find something to take our place.
What is missing from this account of behaviorism is the context of childhood development. Skinner claimed he could produce any sort of individual through controlling the environment of the child, thus the term "Skinner Box." As a matter of reference in the linguistic sense, the biggest problem is that the claim eliminates any method to check if the claim is true or not.
The natural antagonist to Skinner is Vygotsky. Vygotsky takes an environmental approach in noting that psychology has to go beyond the sphere of what any individual reports about their experience. But that experience is what an individual must be. In developing his idea of "zones of proximal development", Vygotsky says: