The Birth of Intermediacy?

by John Holbo on February 1, 2018

I’m taking a break from reading stuff about political theory and liberalism and reading, instead, Other Minds: The Octopus, the Sea, and the Deep Origins of Consciousness [amazon]. It turns out Peter Godfrey-Smith on the octopus brain is more like Jacob Levy on Montesquieu and intermediacy than I was expecting. (The cover of Levy’s book is a bit tentacular. Maybe they should have played that up?)


The cephalopod body, and especially the octopus body, is a unique object with respect to these demands. When part of the molluscan “foot” differentiated into a mass of tentacles, with no joints or shell, the result was a very unwieldy organ to control. The result was also an enormously useful thing, if it could be controlled. The octopus’s loss of almost all hard parts compounded both the challenge and the opportunities. A vast range of movements became possible, but they had to be organized, had to be made coherent. Octopuses have not dealt with this challenge by imposing centralized governance on the body; rather, they have fashioned a mixture of local and central control. One might say the octopus has turned each arm into an intermediate-scale actor. But it also imposes order, top-down, on the huge and complex system that is the octopus body.

This is something a lot of people know about the politics of being an octopus: your various members enjoy semi-autonomy. Tentacles are federated, after a fashion. They continue to act in a purposive manner even if they are cut off from the center. Weird! (See also: Montesquieu on monarchy.) But what does he mean by ‘these demands’?

[Earlier] I contrasted sensory-motor views and action-shaping views of the evolution of nervous systems. The action-shaping approach is less familiar, and it took some effort, historically, to develop it. The central idea is that rather than mediating between sensory input and behavioral output, the first nervous systems came to exist as solutions to a problem of pure coordination within the organism — the problem of how to coordinate the micro-acts of parts of the body into the macro-acts of the whole.

Sensory-motor vs. action shaping? Let’s go back to a rather basic question – and, again, there are political theory overtones:

Why is it worth having such a brain [a smart one!], or any nervous system? What are they for? As I see it, two pictures guide people’s thinking about the matter. These pictures are visible in scientific work and they permeate philosophy, too; their roots run deep. According to the first view, the original and fundamental function of the nervous system is to link perception with action. Brains are for the guidance of action, and the only way to “guide” action in a useful way is to link what is done to what is seen (and touched, and tasted). The senses track what’s going on in the environment, and nervous systems use this information to work out what to do.

He calls this the sensory-motor view. Now, the second:

Modifying your actions in response to events going on outside you has to be done, yes, but something else has to happen, too, and in some circumstances it is more basic and more difficult to achieve. This is creating actions themselves. How is it that we are able to act in the first place? Just above, I said: you sense what’s going on and do something in response. But doing something, if you are made of many cells, is not a trivial matter, not something that can simply be assumed. It takes a great deal of coordination between your parts. This is not a big deal if you are a bacterium, but if you’re a larger organism, things are different. Then you face the task of generating a coherent whole-organism action from the many tiny outputs— the tiny contractions, contortions, and twitches— of your parts. A multitude of micro-actions must be shaped into a macro-action.

Basically, he is saying there has been a tendency to treat organisms as having only a foreign policy. But any complex, active body has, as it were, a domestic policy and politics as well. Godfrey-Smith really could make what he is saying clearer if he said: there is a fallacy similar to that made by IR realist theorists who neglect domestic factors in their explanations of interstate relations and conflicts.

But what explains the growth of big government in octopus brains if, in fact, here is a working system of federalism, with intermediate bodies – various members – enjoying semi-autonomy? Obviously this is guess-work:

A large nervous system evolves to deal with coordination of the body, but the result is so much neural complexity that eventually other capacities arise as byproducts, or relatively easy additions to what the demands of action-shaping have built. I said “or” just above—byproducts or additions—but this is definitely an “and/or.” Some capacities — such as recognition of individual people — might be by-products, while others — such as problem solving —are the results of the evolutionary modification of the brain in response to the octopus’s opportunistic lifestyle. In this picture, neurons first multiply because of the demands of the body, and then sometime later, an octopus wakes up with a brain that can do more. Certainly it seems that some of its impressive behavior is fortuitous, from an evolutionary point of view. Remember again those surprising behaviors in captivity, the mischief and craft, the engagement with humans. There is, it seems, a kind of mental surplus in the octopus.

Even if you have a robust and stable civic society of semi-autonomous members, the central government will grow, and grow crafty, and make all sorts of plans for messing with humans. It’s just like Hayek feared.



aw74 02.01.18 at 4:16 pm

This was a superb book – enjoyed it immensely. I hope there is no extrapolation to society from the fact that octopuses with semi-autonomous cognition can learn and adapt superbly, but are also fragile, vulnerable to predators, unable to pass on what they learn to descendants, and more or less fall apart spontaneously after their (short) lifespan.


Z 02.01.18 at 5:42 pm

I loved this book. Then again, I like cephalopods, evolutionary biology and cognitive science, so there was little surprise. In the “did you know” department, have you followed the most recent hypotheses on how cephalopods manage to imitate colors despite lacking enough color-sensitive cells in their eyes? and ever heard of pre-hatch learning among cuttlefish?

Even if you have a robust and stable civic society of semi-autonomous members, the central government will grow, and grow crafty, and make all sorts of plans for messing with humans. It’s just like Hayek feared.

Or complex systems of single computational units with retroactive loops have emergent cognitive properties. It’s just like Hayek hoped.


steven t johnson 02.01.18 at 7:14 pm

The correlation between encephalization and longevity suggests metabolic regulation and various biochemical and hormonal functions are also actions shaped by the nervous system.

Personally, my brain is not really competent to coordinate two arms and legs. I am awed by the ability of the octopus to handle eight, and coordinating the actions of chromatophores along the way too. Yet, although the number and size of these semi-autonomous arms is impressive, one can get somewhat distorted notion of how centralized the human organism is. There is fairly complex processing in the retina/optic nerve, and all those different neural reflex arcs whose pathways resolutely avoid the brain. The abbreviation CNS in the philosopher’s picture of the mind appears to mean “brain.” But it also includes the spine, which means the “center” has a length a significant fraction of the whole body length.


afeman 02.01.18 at 7:54 pm

It’s also a fascinating book about cephalopods!


Stephen 02.01.18 at 8:24 pm

Do have a look at the intensely venomous, actively hunting cubomedusoids, aka box jellyfish,
which have true eyes with corneas, lenses and retinas, and swim purposefully guided by what they see, but have no optic nerves transmitting the images perceived to their brain, because they have no brain. And yet they swim intelligently.

I’m not sure there is a political analogy for them.


BenK 02.01.18 at 10:29 pm

He (and most people) completely under-rate the complexity of bacteria.


John Holbo 02.02.18 at 12:13 am

“But it also includes the spine, which means the “center” has a length a significant fraction of the whole body length.”

Godfrey-Smith discusses the ’embodied cognition’ stuff briefly. One thing he points out – I don’t know if it’s true – is that a lot of discussions of embodied cognition presuppose bodies that are more like ours than like cephalopods. Which would make sense. It’s hard to discuss ‘the body’ without presupposing what kind. Their bodies are just radically more flexible. Flexibility means more decision making (or does it?)


John Holbo 02.02.18 at 12:13 am

I don’t mean conscious decision-making obviously, just more nervous architecture to manage all that flexibility.


jim of jim's blog 02.02.18 at 1:17 am

Obvious analogy to the box jellyfish: The market. No center, but diffuse coordination.

Also, ants. Seem to coordinate their attacks quite well without the queen getting directly involved.


alfredlordbleep 02.02.18 at 1:42 am

(Limiting myself to humanoid, if you please)

The phrase “embodied cognition” made me recall Hubert Dreyfus’ What Computers Can’t Do (1972 and later editions) and its section, “The Role of the Body in Intelligent Behavior”.

Which let me to “The World Is its Own Model or Why Hubert Dreyfus Is Still Right About AI” by Louis Savain (online 10/2017). This reading is so interesting I risked not surfacing for quite a while. . . Yet not so far as endangering my dilettante’s credentials.
On that I freely associate—

Deep as first love, and wild with all regret;
O Death in Life, the days that are no more!

—Alfred Lord Tennyson


alfredlordbleep 02.02.18 at 1:55 am

P. S. I meant to add this from Savain on the usefulness of philosophers (!)

Dreyfus has been the foremost critic of artificial intelligence research (What Computers Still Can’t Do) since its early days. The AI community hates him for it. Here we are, many decades later, and Dreyfus is still right. Drawing from the work of famed German philosopher, Martin Heidegger and the French existentialist philosopher, Maurice Merleau-Ponty, Dreyfus’s argument has not changed after all those years. Using Heidegger as a starting point, he argues that the brain does not create internal representations of objects in the world. The brain simply learns how to see the world directly, something that Heidegger referred to as presence-at-hand and readiness-to-hand. Dreyfus gave a great example of this in his paper Why Heideggerian AI Failed and how fixing it would require making it more Heideggerian.


John Holbo 02.02.18 at 5:18 am

hey alfredlordbleep,

Funny, I was Bert’s TA at Berkeley for several years. And he was the second reader on my dissertation (despite not really being a Wittgenstein guy himself.) I am the only Bert student ever who just doesn’t get what the fuss is about with this Heidegger stuff. (And the Black Notebooks haven’t exactly caused me to rethink that.) But I should check out new writing arguing a Bert anti-AI line. I always thought that line of his was implausibly strong.


Hidari 02.02.18 at 9:25 am


There’s all sorts of political implications here that Dreyfus doesn’t draw out (and Heidegger wouldn’t, of course) but, at the risk of offending the, ahem, ‘philosophical community’, I still think there’s something in Popper’s critique of Plato (in The Open Society) and that this relates to Plato’s psychology. Most modern psychologists have brains on…well the brain, and are keen to make the (gigantic, although mainly unheralded) leap that because brains are necessary for cognition, they are therefore sufficient for cognition). Plato seems to have had a similar idea. But by definition if you are exalting the brain, you are (implicitly) demeaning the hand. And this leads to conclusion (more by inference and analogy than straight logic) that ‘book learning’ is, in an of itself, ‘superior’ to, in some objective sense, to ‘hand learning’: i.e. knowing that is better than knowing how.

And who knows ‘that’? Intellectuals, teachers, lecturers, journalists. And who knows ‘how’? Plumbers, machine workers, mechanics. There are unquestioned political implications here, in Plato, and, of course, in Socrates (cf which seems to have been mainly brushed under the carpet). Embodied psychology has political implications it seems to me and social psychology (i.e the idea that thought is ineluctably social) even more so.

If you don’t like Heidegger, which is understandable, almost all of the best bits in him can be found in other sources: the anti-representationalism in Reid and J.L. Austin, the dissolution of the subject-object distinction in Dewey, the linguistic aspects in the later Wittgenstein, and the emphasis on the body in Merleau-Ponty (and, some would argue, the later Husserl). With some minor exceptions, you don’t actually ‘need’ Heidegger to be Heideggerean, especially if you are uninterested in the ‘problem’ of ‘Being’ (of which, cf Lakoff and Johnson’s discussion in Philosophy in the Flesh).


hen sholar 02.02.18 at 2:38 pm


Sitting an old geezer seminar with Mark Johnson last Spring, when Bert Dreyfus died, I mourned Bert and dreyfussed as best i could in Johnson’s class on “mortal ethics,” to the point that he invited me to sit his Fall quarter class, reading through Dewey’s Experience and Nature. Scales fell from my heideggerian-tinted glasses from the second paragraph of the preface that Dewey writ in 1925 — alluding to the hermeneutical circle — that here in E&N was an agreeably comparable text to B&T, equally insurgent and along similar lines. So now I am cobbling together some sort of Dewey-Heidegger Reconciliation Project, and just found Godfrey-Smith essay/review of E&I, always looking for co-readers, as Dewey is such a master prose stylist, also as was our Heidegger. At any rate, my first read through E&N, and further ‘firsts’ in Dewey’s thought shows a great deal of congruence between these two old boys, and bonus: Dewey was a nice man… Why wasn’t I told about this 40 years ago?


Z 02.02.18 at 2:45 pm

Using Heidegger as a starting point, he argues that the brain does not create internal representations of objects in the world. The brain simply learns how to see the world directly, something that Heidegger referred to as presence-at-hand and readiness-to-hand.

That goes way beyond an anti-AI critique, that’s an anti-cognition critique (an anti-I critique, perhaps)! The whole piece is more nuanced than this sentence might indicate (and quite interesting, thank you for the link!), though. Still, it refers positively to a couple of claims that I have a hard time reconciling with things I think we by now really do know about human (or octopoid) cognition, if I understand them correctly. “[T] he meaningful objects … among which we live are not a model of the world stored in our mind or brain; they are the world itself” for instance. I like to start small, myself, so let me ask: where are questions in the world itself? (Obviously, our cognition can produce questions; mine just did.)

Fun fact (?), after reading Other Minds, I wrote a science-fiction short story in which octopoid alien scientists (with the expected distributed intelligence) puzzle around human cognition and especially how humans are able to systematically produce recursive cognitive structures despite apparently lacking the sensorimotor apparatus required to detect recursion (just like human scientists puzzle on how cephalopods can imitate so perfectly outside colors despite apparently lacking the sensorimotor apparatus required to detect colors).


MisterMr 02.02.18 at 3:40 pm

@Hidari 13

It seems to me that, while there is an obvious “class preference” in the opinion of value of the ancient philosophers about the various kinds of knowledge, there is also an important distinction in the way the relation between “will” and “intelligence” is conceived.

If we take a modern evolutionary view, we first and foremost are individuals who need to eat and want to reproduce, and in order to eat more and reproduce more we evolved the ability of creating “model worlds” in our brain, in order to predict what happens in the real world.
In this view, “intelligence” is a byproduct of “will”. I think that this is the point of view of german idealists, though they spun it in a weird mysthical way, and is also the point of view of C.S. Peirce (“pragmaticism”).

But before this, “intelligence” was often assumed to be a distinct thing from “will”, and often seen as superior, in some way, even to “noble” forms of action.


Whirrlaway 02.02.18 at 5:09 pm

Rorty explored this, did he not? But he sadly died young and left the field to Searle.

Cognition, esp social cognition, flows as much from what is ignored as from what is seen (… the famous gorilla …). So the brain is a repressor as much as a sensor. That would match the point about the Octopus as the oppressor of the Arms.


Antonin 02.02.18 at 7:43 pm

Why do you link to [amazon]?


Stephen 02.02.18 at 8:27 pm

jim@9: no, I don’t think so. The market doesn’t make a conscious, enforced decision to move in one way or another for its own benefit. A cubomedusoid, though without a central brain, does.


M Caswell 02.02.18 at 10:13 pm

The only mention of the brain in Plato I can think of is Timaeus (73b ff). Not what I would call “exalting”…


Hidari 02.03.18 at 12:29 pm

Just to be precise, I specifically didn’t say that Plato ‘exalted the brain’, I said he had a ‘similar idea’. According to Wikipedia, it was Alcmaeon of Croton who first posited the brain as the seat of the intellect. This was (like Plato’s speculations) in no way a scientific hypothesis: Alcmaeon believed that ‘the human soul was immortal and partook of the divine nature, because like the heavenly bodies it contained in itself a principle of motion’ (he seems to have been influenced here by speculative metaphysicians like Thales et al) (

Much of the core work in the Western intellectual tradition (especially as it relates to the social or human sciences), especially in the last 200 years, is an attempt to secularise or naturalise what are essentially metaphysical (or even religious) guesses, and then pass them off as ‘science’ by assuming what you are meant to be proving, and if these speculations have convenient political implications, so much the better.

@17 Rorty is a great philosopher (perhaps one of the last in the Western tradition) and his anti-representationalism is crucial. Unfortunately discussions of his name tend to get sidetracked into discussions of ‘relativism’ (most contemporary discussions of ‘relativism’ tend to be the modern equivalent of ‘how many angels can dance on a pinhead’: parlour games for aspiring intellectuals), but it’s his anti-representationalism which is the real gauntlet thrown down to (e.g.) Fodor et al. Most philosophers and cognitive scientists in the representationalist tradition have dealt with Rorty’s arguments by ignoring them IMHO.


Hidari 02.03.18 at 4:07 pm

This got quoted by John in another thread but it seems more germane to this one:

This new book also seems to be relevant:

‘From Plato onwards, western philosophy has favoured mind over “mere” body, so that by the time we get to Descartes, the human has become hardly more than a brain stuck atop a stick, like a child’s hobbyhorse. This is the conception of humanness that Damasio wishes to dismantle. For him, as for Nietzsche, what the body feels is every bit as significant as what the mind thinks, and further, both functions are inextricably intertwined. Indeed, from the very start, among the earliest primitive life forms, affect – “the world of emotions and feelings” – was the force that drove unstoppably towards the flowering of human consciousness and the creation of cultures, Damasio insists.’

Most Western psychology presupposes that there is some kind of Medieval Chain of Being in which Man (sic) stands at the ‘top’ and so Human Consciousness(or ‘cognition’) is, so to speak, the ‘best’ type to which animals (and plants!) can only approximate. But as Stephen Jay Gould tirelessly pointed out, there is no possibly scientific way by which that could be true. Animals (like Homo Sapiens Sapiens) can only be ‘judged’ in terms of their ability to survive and reproduce. One might assume therefore that it is bacteria that are the ‘highest’ form of life in Earth as they are the most prevalent. Therefore we might ask the question:

in what ways are human intelligence inferior to bacterial intelligence, given the relative reproductive success of the latter?

Or as Douglas Adams put it: “For instance, on the planet Earth, man had always assumed that he was more intelligent than dolphins because he had achieved so much—the wheel, New York, wars and so on—whilst all the dolphins had ever done was muck about in the water having a good time. But conversely, the dolphins had always believed that they were far more intelligent than man—for precisely the same reasons.”

We humans think that we are terribly terribly clever because it is us who is doing the assessment of exactly what we mean by the word ‘clever’. If octopuses were to do the same thing, they might be able to list quite a lot of reasons as to why ‘their’ intelligence is superior to ‘ours’. *

*To the obvious rejoinder that ‘ah but octopuses don’t do they?’ the response is, of course, ‘and you know this, how?’


Adam 02.04.18 at 7:00 am

“Octopuses have not dealt with this challenge by imposing centralized governance on the body; rather, they have fashioned a mixture of local and central control. One might say the octopus has turned each arm into an intermediate-scale actor. But it also imposes order, top-down, on the huge and complex system that is the octopus body.”

So what? That’s how our nervous system works as well. Single synapse reflexes, local pattern generators in the spinal cord, supra spinal reflexes, descending control from the motor regions, fine motor control from the cerebellum, all guided by higher order processes…


Stephen 02.04.18 at 7:52 pm

Hidary@22: show me a city or a machine built by octopuses, show me a work of philosophy written by octopuses, of poetry written or even sung by octopuses, and I might begin to take you seriously.


F. Foundling 02.05.18 at 3:41 pm

@ Hidari 02.03.18 at 4:07 pm

>Animals (like Homo Sapiens Sapiens) can only be ‘judged’ in terms of their ability to survive and reproduce.

Everything can be judged in terms of anything: the judge decides, and the only judge in sight is Homo Sapiens Sapiens. Evolution is not a judge or a deity, just a fact. I humbly submit that in my experience, conversations with animals tend to be far less rich in information, meaning and complexity than conversations with humans, especially once one sets aside the initial element of exotic novelty, charm and refreshing primitiveness. Octopuses are free to assess this differently, but I reserve for myself the right to act in accordance with my own assessment and not with that of the octopuses. They are, of course, also entitled to act in accordance with theirs, and I wouldn’t expect anything else.

>To the obvious rejoinder that ‘ah but octopuses don’t do they?’ the response is, of course, ‘and you know this, how?’

And the response to that question is always the same, regardless of what one is discussing: ‘it sort of looks like that’. Mostly negative evidence, of course (the conspicuous absence of [anything that looks like] octopus blogs, octopus philosophical treatises and octopus phonology, morphology and syntax, for instance), but you might as well be a theist – or, more consistently, a mineral – if you aren’t OK with basing assessments and decisions on negative evidence.


Hidari 02.06.18 at 9:12 am

And of course it’s the questioning of human wonderfulness that brings out the bile. But ( I’m quoting from memory) as Richard Dawkins once remarked (in the Blind Watchmaker), after quoting from a vaguely racist comment of Darwin’s:

‘Don’t be put off by Darwin’s assumption of white superiority. White superiority was as taken for granted in his day as human superiority is taken for granted in ours.’

This is true, and if Pinker et al of the Evolutionary Psychology crowd weren’t so keen to distort the science to make tedious right wing comments they might care to reflect that taking Darwinism on board really tends one towards vegetarianism (that’s the most obvious ‘political’ point to infer from it).

Which leads to the conclusion that in 200/300 years people might back on the fact that many of us eat meat in much the same way as we look back on the fact that (e.g.) Washington etc. owned slaves.

Indeed someone has made a movie based on this premise:

Comments on this entry are closed.