what was reductionism?

On the last day of a recent conference on the philosophy of biology, an eminent scholar pointed out that no one had yet uttered the R word. He wondered aloud whether the long struggle over reductionism had finally passed, and whether we could move on now please?

The reductionism/anti-reductionism debate is over. I mean, reductionism will always have its defenders – there is no position so hopeless that someone won’t defend it. But serious debate is over. It’s over because reductionism is useless – in the literal, practical sense, it’s useless. You can’t do anything with reductionism.

Ok, what do I mean by that? For the purposes of this post, let’s take reductionism to mean the following:

Everything in the world is nothing more than the sum of its smallest parts.

If you prefer, you could think of it instead about the way we understand everything in the world, but the result will be the same. Either way it is a useless position, because we simply cannot deal with the world in terms of its smallest parts, not all the time.

When confronted with this fairly obvious fact, philosophers will usually say something like “well yes, but in principle…” But principles are supposed to be our basis for activity. If a principle is utterly useless, even misleading for us, then it will be rejected. Reductionism is on the same footing as its opposite, the claim that everything in the world is fundamentally unified, that all is one. In some sense of ‘true’, that may be true. But you can’t use it as a basis for reason and action. Because life depends on chopping up the world into bits, we reject the idea that ‘all is one’ just by living. Similarly, our lives can’t function if we always and only treat the world as made up of quarks or whatever. So forget about reductionism. It’s over.

SMBC explained the situation clearly enough:


But I think there is still some work to be done in explaining, as the title of this post says, what it was. Why did people find the idea of the world-as-just-its-parts so compelling? Some even took it to be essential to scientific thinking, a mandatory prerequisite. Why?

birds flocking

I want to suggest that the actual prerequisite for scientific thinking is the analytic imperative. But being analytic isn’t the same as reducing something to its smallest parts. It just means not explaining the parts of a system in terms of the whole thing. So if I’m being analytic, I can’t explain the movement of a single bird by citing the movement of the flock it flies with, taken as a whole. I have to do it in terms of the local interactions between birds. For example, I can only cite things that my particular bird can see or hear. Or if I want to explain why a single rabbit was eaten by a particular fox, I can’t do it by citing the rising population of foxes. The whole population didn’t eat my particular rabbit, some specific fox did. That is holistic reasoning, explaining by embedding something in its context. If I want to be analytic, I have to explain the rabbit-level event at the level of rabbits, not some higher level like populations. But as long as I’m not explaining by citing the context, it doesn’t matter what level the explanation is pitched at. You can be perfectly analytic about big macroscopic things like organisms, carburetors and planets.

So what I’m suggesting is that people have confused the imperative to treat things categorically and in isolation from their context with the requirement that we always break things up into their smallest parts. Reductionism would then be a kind of mistaken reification of this methodological principle into a supposed metaphysical reality. You can see how that would be a tempting shift to make: the society I’m born in is my context, but to analytically (scientifically) explain my behaviour, you should look for the specific local influences that made me turn out the way I did. You have to find the things that actually impinged on my person, because the society I’m in is, from the perspective of me as individual, my context. But if we take some bit of my brain as the thing-to-be-explained, then my body as a whole becomes the context. Me as an individual is now the thing that cannot be dealt with analytically – you have to find the specific electro-chemical forces that are acting on that bit of my brain to explain it scientifically. If you follow this pattern, it looks like the only thing that won’t be a context for something is whatever the smallest bits are.

But as I said before, you can be analytic at any level. The imperative to be analytic which I think motivates science does not require that you should always only explain things at the lowest mereological level possible. But thinking that it does, I claim motivated reductionism.

Advertisements

16 thoughts on “what was reductionism?

  1. 1) “let’s take reductionism to mean the following: Everything in the world is nothing more than the sum of its smallest parts.”
    One might think that this isn’t the happiest way to characterise a substantive position, since it’s either a trivial truth, or a trivial falsehood, depending on how ‘the sum of’ is taken.

    2) Secondly, it seems that when you say “When confronted with this fairly obvious fact, philosophers will usually say something like “well yes, but in principle…” But principles are supposed to be our basis for activity”, and then when you say “In some sense of ‘true’, [substance monism] may be true. But you can’t use it as a basis for reason and action” you’re implicitly ruling out the idea that we might believe something just because it’s true.

    Now maybe believing things just because they’re true, not because we want to do something ‘useful’ with them, is actually a stupid idea that we should reject. But it seems that the burden of proof is on the person claiming it, because it seems that precisely in ‘everyday life’, we sometimes display things like ‘curiosity’, and think that it might be nice to know how things really are, and what’s really true, and then believe that. So if you think that’s just misguided, you need to explain why.

    Or else, just say “I’m talking about reductionism as a practical paradigm for scientific investigation, not reductionism as a philosophical doctrine.”

    3) Incidentally, I sometimes wonder – if you think the quoted principle is substantive, and thus that it makes logical sense to say something is ‘more than the sum of its parts’, shouldn’t you also think that it makes logical sense to say something is ‘less than the sum of its parts’? And that there is therefore a substantive philosophical principle, which people might defend or attack, and might be true or false, that “Everything in the world is nothing less than the sum of its smallest parts”?
    Yet I don’t think I’ve ever heard anyone utter those words.)

    • Of course you’re right that there is much assumed in this post, and I’ve offered little in the way of a rigorous defense. It would be truly surprising if anyone could settle anything in philosophy in less than 1000 hastily thrown together words. It’s equally unlikely that I’ll be able to provide that rigorous defense in the span of a comment on a post – but I’ll do my best to gesture at where I think a good response could come from.

      I have in mind a broadly Wittgensteinian framework for thinking about what ‘true’ means. We learn how to use the word ‘true’ through the activities that make up our lives. Its meaning is originally tied to the affordances which make up our experienced world. Philosophers have tended to want to extend that meaning to something transpersonal – something like “a claim is true if it corresponds with reality”. But if by ‘reality’ you mean the noumenal ground of existence, then we can no longer claim any of our beliefs are true. We have no access to noumenal reality, only our phenomenal experience. So by trying to extend the meaning of ‘true’ beyond its normal use, we’ve actually made it meaningless to ourselves. Under the burden of pure objectivity, the notion of ‘truth’ collapses altogether.

      Now, you’re probably not going to be convinced by any of that unless you’re already sympathetic to the Wittgensteinian/Kantian framework. I tend to think there are very good reasons from the philosophy of language and mind to accept such a framework, but again, defending it is a bit too involved for a comment on a blog post.

      As for your last suggestion, I think that’s nifty. Some of the work on self-organization actually could be phrased that way. Guys like Haken and Kelso argue that, given the right conditions, the micro-realization of a system actually becomes irrelevant to explaining its behaviour. The huge diversity of degrees of freedom the system had outside of a self-organized regime get reduced to just a few degrees of freedom – in a sense, the system becomes simpler than before. It may be a bit loose and metaphorical to say that the system is now ‘less’, but it nicely captures something about the way they present self-organization.

      • Being as uncharitable as possible, because why not:

        “We learn how to use the word ‘true’ through the activities that make up our lives. Its meaning is originally tied to the affordances which make up our experienced world.”
        I feel like ‘tied to’ is weasel words for ‘I learnt a word in the context of X, therefore it means X’, which is a clearly fallacious move when spelled out.

        “if by ‘reality’ you mean the noumenal ground of existence, then we can no longer claim any of our beliefs are true. We have no access to noumenal reality, only our phenomenal experience…you’re probably not going to be convinced by any of that unless you’re already sympathetic to the Wittgensteinian/Kantian framework”
        I used to be sympathetic to the Kantian framework, but was persuaded that it simply must be false in its central claim, on pain of vicious scepticism, by a fairly simple argument that no-one has yet shown me a convincing response to.
        The argument is: we learn facts about other people’s minds by observing and interacting with them; these facts cannot be phenomenal facts in any normal sense; therefore we have some determinate empirically-based knowledge that goes beyond phenomena.

        “It may be a bit loose and metaphorical to say that the system is now ‘less’”
        I think that’s right – claims about ‘less’ and ‘more’ are metaphorical ways of directing people’s attention to things we think are highly relevant, interesting, etc. But it follows from this way of interpreting them that they’re never literally true…

        “Luke Roelofs’s criticism that your collapsing true into empirically verifiable…seems off to me because…reductionism as sometimes defended can be logically and not just empirically identical to substance monoism as sometimes defended.”
        I see no suggestion in the post that reductionism and substance monism are *logically* equivalent. Why would that be? Corey just seems to say that they “on the same footing”.

        (He also calls them ‘opposites’, which strikes me as false – they might be combined in a several ways, particularly given that Corey’s definition talks only about things “in the world”, hence need not apply to ‘the world’ itself.)

        • “I feel like ‘tied to’ is weasel words for ‘I learnt a word in the context of X, therefore it means X’, which is a clearly fallacious move when spelled out.”

          Another way of interpreting ‘tied’ to in this context is ‘the meaning of the word x was given thus and so; therefore the standards for evaluating whether something is an x depend on those circumstances, because that’s how the word got its meaning’. That, at least, is the move Wittgenstein makes repeatedly in On Certainty.

          “we learn facts about other people’s minds by observing and interacting with them; these facts cannot be phenomenal facts in any normal sense; therefore we have some determinate empirically-based knowledge that goes beyond phenomena.”

          I don’t understand why purported ‘facts’ about other minds don’t count as phenomena? Even if you were to have access to my mind as direct as I do, I don’t experience the noumena, so why would you seeing my mind give you access to the noumenal?

          • You’ll notice I didn’t say noumena, I said beyond phenomena. Consider the fact that when you read my comment you experienced the sort of feeling associated with linguistic understanding. That’s not reducible to any purely physical description, behavioural or neurophysiological.

            It’s still an ‘appearance’, but it’s ‘how things appeared to you’. That’s irreducible to any way things appear to me. To put it another way, the fact that things appear to X is not an aspect of how X appears. So I can know at least one type of fact about things that doesn’t remain within how they appear to me, and hence isn’t ruled by how I ‘synthesise the manifold of appearance’.

            Once that’s granted, though, metaphysics comes rushing back – do things appear to stones? Do things appear to the universe? Did things appear to God before the beginning of evolution? Can one thing, to which things appear, be composed of two other things, to which things appear? Can it be thus divided infinitely? Have there always been appearances to one being or another? If I set fire to someone’s entire brain, do things continue to appear to them some way? Etc.

            • Not only that! But arguably, a large part of how I am able to infer what your mental life is like by my experience of causation: I know what sort of mental effects tend to be produced by what observable causes, and what sort of mental causes tend to produce what observable effects. But this suggests that I have some understanding of causation that can go beyond me using it as a concept to synthesise how things appear to me. And then you get all the metaphysical questions about causation – what has causal power? Can anything cause itself? Etc.

            • I think I still don’t get it. If it’s not phenomena and it’s not noumena, what is it?

              Is there perhaps someone I could read on this? I suspect I’ll need a more thorough treatment to get the argument properly.

        • I don’t think that Cory’s criticism is dependent on a narrow theory of meaning. It is easy to imagine a reductive theory that can not be refuted empirically but none the less has a meaning we can accept or reject (it is not vacuous). For example take the idea that we can only have true statements about the fundamental elements of things and their relations (sort of a strengthen of the Democritus’s claim that all that existed was atoms and void, taste etc. mere appearance), this can’t be empirically refuted because empirical evidence is not itself given or formulated in terms of those elements so if its true those empirics are false and therefore no inference from them can be trusted. We can however reject this thesis for being logically contradictory or because it leads to radical skepticism or any number of other non-empirical reasons. A more realistic example, for a contrary principle, is the near total rejection of Parmenides monism (change is impossible, there is not many but only one etc.). I would contend you can’t argue against Parmenides on empirical grounds without a vicious circle, but there do appear to be rational grounds to reject it. Is there any reason to think these were the kinds of reductionism argued for or against in biology?

          Cory’s presentation of reductionism as a dead horse suggested that we start in a context of philosophy of biology. So I think the kind of reductionism that he is talking about would be the kind that makes claims about some empirical facts or at least the efficacy of certain heuristic methods, Cory says investigation has failed to verify these claims. Accepting that the only way to save the reductionism in question from falsification is to lower the bar of empirical evidence, if we lower the bar to zero then what is left of the reductionist claim? It could have had extra-empirical conceits as part of it but I’m inclined to agree with Cory that the kinds of reductionism brought out in science debates tend to only have empirical truth makers and if you lower that bar then they lose meaning. The claim that remains is just the logically possibility to describe a physical situation in a reductive way, but a similarly weak monistic claim could be made in science and both could then be true and then a reductive description could be transformed by logical/truth preserving steps into a hollistic/monistic description making them logically equivalent (hence my extension of Cory’s argument). This is all independent of what the nature of truth or reality is but rather set by the context of the debate, which avoids those sorts of questions.

          • “Cory’s presentation of reductionism as a dead horse suggested that we start in a context of philosophy of biology. So I think the kind of reductionism that he is talking about would be the kind that makes claims about some empirical facts or at least the efficacy of certain heuristic methods”

            Which I would have happily accepted, had he not then added a paragraph roping in philosophical reductionism and criticised that too.

            “the logically possibility to describe a physical situation in a reductive way, but a similarly weak monistic claim could be made in science and both could then be true and then a reductive description could be transformed by logical/truth preserving steps into a hollistic/monistic description making them logically equivalent”
            Ah, I see what you mean. I’m not sure I accept that once empirical considerations are removed, it’s quite such a free-for-all. But even so, firstly, monism (about the universe) isn’t the same as holism (about particular systems within it), and usually opponents of reductionism don’t argue that holistic descriptions are possible, they argue that they’re necessary, because the reductive descriptions are incomplete. Saying that we can always fully describe things in either idiom sounds like just what reductionists should accept and be happy with.

  2. I would substantively agree with your position Cory, but I do see one other motivation. It is that their are two strains of anti-reductionism, and while one takes the tack you do, the other attempts to claim strong emergence: Ie that the parts of things in an aggregation obey a truly alternative set of rules as part of the aggregation than the parts obey outside the aggregation and not merely a special, if difficult or impossible to derive, subset of the usual rules. The thing is that this emergentism is the same sort of “in principle” idle wheel as reductionism, since it is rarely if ever used to try and find and prove any such violation of normal rules in special aggregations (and even those who would try and find such violations have yet to succeed).

    So I would say there is a non-idle, but weak, part of the reductive claim which merely consists in the denial of violation of principles like the conservation of energy even by systems that fail to be practically described by mechanical laws that would guarantee such conservation. This is non-idle since it is a bet against research programs that would focus on seeking out such violations (eg. build a perpetual motion machine running off the vital force). Now whether the claim of the absence of strong emergence is reductionism I guess is debatable but it is clearly a necessary component of reduction (to admit its falsity would be to deny reductionism as usually construed).

    Luke Roeful’s criticism that your collapsing true into empirically verifiable (verificationist theory of meaning?) seems off to me because your criticism of reductionism seems to me to suggest that reductionism as sometimes defended can be logically and not just empirically identical to substance monoism as sometimes defended.

    Taking something to be less than the sum of its parts would seem to mean taking the aggregate to be simpler than its constituents taken together, ie to assume some of those parts add nothing to the action of the whole. Such positions are hardly unknown fatalism would be an example…

  3. “If it’s not phenomena and it’s not noumena, what is it?”
    It’s other people’s phenomena, i.e. other people’s mental states (construed as appearances-of-them-to-themselves and appearances-of-other-things-to-themselves, and systematisations thereof, in line with how Kant construes them).

    “Is there perhaps someone I could read on this? I suspect I’ll need a more thorough treatment to get the argument properly.”
    Nothing not written by me, I’m afraid. Draw what conclusions you like about the soundness of the arguments that nobody except me finds convincing.

  4. It’s really a nice and helpful piece of information.
    I’m satisfied that you just shared this useful info with us.
    Please keep us up to date like this. Thanks for sharing.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s