Distributed Process vs. Division of Labour

19 posts / 0 new
Last post
Tom Henry
Offline
Joined: 26-09-16
Jan 8 2017 03:09
Distributed Process vs. Division of Labour

I found this article interesting and thought that others might too.

https://aeon.co/essays/how-ant-societies-point-to-radical-possibilities-for-humans

It appears that ant colonies can't be used to show how hierarchy or 'competency' are efficient ways to organise the functions of people within society - thereby justifying the work ethic and capitalism, etc - since ants are not, according to recent research, organised in terms of division of labour but, rather, through distributed processes.

Tom Henry
Offline
Joined: 26-09-16
Jul 11 2017 03:38

How come this topic is 'updated' when I haven't done anything to it? Is it some kind of automatic thing?

Pennoid's picture
Pennoid
Offline
Joined: 18-02-12
Jul 11 2017 13:54

I mean there is a division of labor within multicellular organisms; specialization of function of tissues let alone limbs. It does provide some insight to the function of dol in the historical development of human societies, but I'm not clear on what precisely.

Efficiency is always defined in relation to particular ends; or evolutionarily, at the behest of particular environmental stimuli. So ants providing a 'truth' about efficiency that holds for humans is not a very strong argument.

Joseph Kay's picture
Joseph Kay
Offline
Joined: 14-03-06
Jul 11 2017 20:01

Re the 'updated', looking at the revision history it looks like you fixed a typo, it went into the approval queue, then an editor approved it. The approval would have reset the 'updated' flag.

I'm not sure ant societies are an alternative to a division of labour. There's not a biological division of labour as such, insofar as ants switch from one task to another as needs dictate, but at any given moment there's still a division of tasks. Reminds me almost of Toyotism, where workers are meant to be multi-skilled and able to switch to high priority tasks, vs being specialists on a Fordist production line.

That said, ant interactions are a great example of efficient system-level organisation emerging from purely local interactions. At an abstract level, that - and complex systems science, which studies these things - probably does have implications for communism. E.g. a non-market order need not be centrally planned, if it's arranged in such a way that autonomously interacting local agents (individuals, collectives...) can communicate with one other. What such a system would look like is an open question, but some of the ideas in complex systems, together with those about successful commons regimes, would be an interesting thing to explore.

It's a bit mathsy in places (GCSE level I think), but the Santa Fe Institute has a free online course introducing complex systems here, iirc ants get a mention, and there's a downloadable computer model of ant foraging behaviour to play with.

Pennoid's picture
Pennoid
Offline
Joined: 18-02-12
Jul 12 2017 18:21

I've always been skeptical of complex systems science: are there any other good intros than that you provided? Any texts?

Khawaga's picture
Khawaga
Offline
Joined: 7-08-06
Jul 12 2017 18:32

Why skeptical? I mean, that type of theory is quite important in modelling the climate/weather, the most energy efficient route for a vehicle to take through a network of roads and so on.

Now, I get your skepticism in the sense that I've seen it be used to naturalize free market economics in the sense that since [insert natural phenomenon] is a complex/chaotic system and the market is, therefore the market is natural. But that's just shitty argumentation and doesn't really undermine the science of complexity/graph theory/chaos science.

Pennoid's picture
Pennoid
Offline
Joined: 18-02-12
Jul 12 2017 19:16

You identified the root of my skepticism; I've only ever seen the term or concept deployed in pretty superficial ways. I'm curious about the abstract nature of the science as well, and how that limits it's applicability to issues of social relations. I'm not made up though because obviously I haven't dug into it.

Pennoid's picture
Pennoid
Offline
Joined: 18-02-12
Jul 12 2017 19:34

I wonder also about the extent to which concepts like centralization make sense to apply at certain levels, and about analogous relationships between things like cells, ants and banking firms; they certainly appear to exhibit very DIFFERENT qualities at the individual level, not to mention in the process and nature of their interaction.

But perhaps I'm wrong, we will see!

Joseph Kay's picture
Joseph Kay
Offline
Joined: 14-03-06
Jul 13 2017 10:25

Yeah, you do get people who try and use it to naturalise markets ('spontaneous order, just like ants!'), or to develop better economic theories of how markets really operate (complexity economics, non-equilibrium economics, evolutionary economics). But I think Khawaga's right that it's bad reasoning to infer 'markets are good/natural' from 'complex systems are ubiquitious'.

In fact given the ubiquity of complex systems, you'd expect to find them everywhere (I'm pretty sure you could model the soviet planning bureaucracy as a complex system of partially autonomous parts, information flows, commands etc, and some of its dysfunctions as emergent properties from those interactions. Arguably Marx's discussion of value as a regulating law is a discussion of an emergent property, i.e. something that only happens under large scale wage labour and market exchange that can't be found among any of the parts...).

I think this association is probably due to Hayek, whose 'catallaxy' (spontanous order of the market) sounds a lot like the complex systems notion of emergence. But interestingly, a lot of Hayek's ideas about the importance of local knowledge echo those of Otto Neurath, a socialist on the other side of the 'calculation debate' (in its first phase). John O'Neill's written a lot of good stuff on this, I'll see if I can find PDFs for the library as his books are expensive.

I don't really have the maths/programming skills to pursue it at the moment, but things like the technique of agent-based modelling could provide a reasonable way to look at objections to 'free-access communism', i.e. is there a 'free-rider problem', can reasonable models of peer-feedback keep it within viable levels, is a free-access communism a stable or unstable equilibrirum, if it appears unstable, what institutions/interactions could stabilise it, etc. This kind of modelling is hardly definitive, but it does provide a way to say 'ok, we reject X neoclassical assumption, Y seems more realistic, how does that change the viability of Z.'

Bowles & Gintis' A Cooperative Species could be seen as Mutual Aid rewritten by complex systems theorists. One of their arguments is that strong reciprocity is common, not only narrow self-interest. This means people incur costs to enforce group norms (of distributive fairness, or whatever). This is the kind of thing communists have long-argued would make free-access communism viable, though Bowles & Gintis don't make that argument themselves. However it's also possibly why stigmatisation of welfare recipients as 'scroungers' can be effective, people can be induced to take costly action to punish apparent violations of egalitarian group norms (e.g. 'everyone should contribute to the best of their abilities').

One day I'd like to have the maths and programming skills to really get stuck into modelling these kind of things, but you know, wage labour gets in the way.

Joseph Kay's picture
Joseph Kay
Offline
Joined: 14-03-06
Jul 13 2017 10:38

In terms good introductions, I haven't read it yet, but I've seen this recommended by a few people: Network Science by Albert László-Barabási: http://barabasi.com/networksciencebook/

Organisation of the Organisationless explicitly uses some concepts from network/complexity science to theorise social movements like Occupy/movement of the squares: http://www.metamute.org/sites/www.metamute.org/files/pml/Organisation-of-the-Organisationless.pdf

Joseph Kay's picture
Joseph Kay
Offline
Joined: 14-03-06
Jul 13 2017 11:02

As an aside, one of the topics in graph theory, an area of maths dealing with network-like structures called graphs, is matching. This studies how you achieve optimal allocations when you only have ordinal preferences (an ordered list), not cardinal ones (the price someone's willing to pay). I think economists study this as 'matching theory', but I haven't read any of the economics literature.

It's used today for things like allocating dissertation supervisors in universities, when every student submits say, their top 10 choices and an algorithm attempts to give as many as possible as high a preference as possible. But I suspect this kind of algorithm could have applications for a broad class of non-market allocations. Especially in cases where allocations are repeated, you could perhaps modify the algorithm to bump 'losers' in one round to the front of the queue in the next, to maintain a sense of fairness and prevent serial disappointment.

Together, I think a lot of this stuff - strong reciprocity, agent-based models, matching - could relieve some of the communist reliance on 'post-scarcity'. Things can be scarce, and still allocated in an egalitarian, non-market way (see also work on commons, like Elinor Ostrom). I mean I do think that much scarcity (food, housing) is an artefact of the market, but even communism's going to want a hard (even negative) limit on greenhouse gas emissions, we still only have 24 hours in a day, we can't do certain things at once, and we don't want too much/any unpleasant labour, all of which are sources of 'scarcity' in the sense economists use it, but none of which need derail communist arguments imho.

Pennoid's picture
Pennoid
Offline
Joined: 18-02-12
Jul 13 2017 15:27

1. I'm put off by how the santa fe intro is using certain terms. E.g. 'decentralized' seems to mean something really specific but used in a vague way. As you say, even Soviet planning (perhaps the most extreme example of centralism) wasn't centralized by their metrics. Maybe it makes more sense to say multi-centered (or a network as I think you say.

2. This ties to a second sort of feeling; this strikes me as 21st century dialectics; that's not meant to write it off, but that it is people trying to understand emergent phenomenon and complexity, often through bringing distinct disciplines in dialogue, abstracting away from these or those particulars to identify what is common - their complexity.

3. I'm interested in studying more concrete complex questions, especially evolutionary biology and developmental biology (as opposed to behavioral biology so much) because I think some of the info there seems to be quite rigorous and touch on common abstract questions of history, teleology, complexity, systems etc.

Thanks for the sources as well. I wonder about forms of distribution based on algorithms; it calls to mind AI, and also the problem Kautsky poses in his magnum opus; that all new social relations in human history have unintended consequences, which casts doubt on Engels even narrow conception of leaping from the realm of necessity to freedom. Spooky stuff!

Khawaga's picture
Khawaga
Offline
Joined: 7-08-06
Jul 13 2017 15:40
Pennoid wrote:
the problem Kautsky poses in his magnum opus; that all new social relations in human history have unintended consequences, which casts doubt on Engels even narrow conception of leaping from the realm of necessity to freedom. Spooky stuff!

Could you expand on this?

Quote:
1. I'm put off by how the santa fe intro is using certain terms. E.g. 'decentralized' seems to mean something really specific but used in a vague way. As you say, even Soviet planning (perhaps the most extreme example of centralism) wasn't centralized by their metrics. Maybe it makes more sense to say multi-centered (or a network as I think you say.

I haven't read the santa fe intro, you're referring to, but in most cases people that work in complexity theory tend to use centralized, decentralized, and distributed as borrowed from the theory of networks (hence, the Paul Baran definitions). The way you use multi-centered is actually how network theory defines decentralized.

This little blog post explains centralized, decentralized, and distributed quite well.

Quote:
2. This ties to a second sort of feeling; this strikes me as 21st century dialectics; that's not meant to write it off, but that it is people trying to understand emergent phenomenon and complexity, often through bringing distinct disciplines in dialogue, abstracting away from these or those particulars to identify what is common - their complexity.

I've had the same thoughts as well. Way back when here on libcom there were some discussions on this same topic and I think complexity as 21st century dialectics was mentioned. Can't seem to find the discussion though.

Joseph Kay's picture
Joseph Kay
Offline
Joined: 14-03-06
Jul 13 2017 15:59

I haven't really thought about how it relates to philosophical ideas like dialectics. Afaik modern complexity science comes via applied mathematics (cybernetics), which itself grew out of WWII research into feedback mechanisms (such as those used to aim anti-aircraft guns).

Feedback has some similarity with dialectical thinking, though (thinking out loud) I wonder if dialectics implies symmetry ('opposites'), whereas complexity (feedback, graph theory) generalises the notion that things can be both cause and effect, without presuming any particular structure in advance.

Otoh, Marx makes certain assumptions about how agents act - Moneybags wants M-C-M', workers want LP-M, but band together to improve their conditions. I don't know how easy it would be, but that sounds like part of the specification of an agent-based model, which could be explored for emergent properties (crises?), (in)stability etc (would the model be stable without a repressive agent to break up worker collective action?). Agent-based Marxism anyone?

Khawaga's picture
Khawaga
Offline
Joined: 7-08-06
Jul 13 2017 16:15
Quote:
whereas complexity (feedback, graph theory) generalises the notion that things can be both cause and effect, without presuming any particular structure in advance

That's in dialectics as well: how a system reproduces its own preconditions as results/products of that system, results that in turn are the preconditions for the system to reproduce itself and grow. The circuit of capital is a perfect example. Commodities and money, capitalist social relations were all the "causes" for the emergence of capitalism, but they are constantly reproduced as the effects of the system of capital. Not sure if that makes sense.

Quote:
Agent-based Marxism anyone?

That would be interesting.

Joseph Kay's picture
Joseph Kay
Offline
Joined: 14-03-06
Jul 13 2017 16:40

Thinking about the matching algorithm above, it's actually more like (local) central planning than a distributed system. I.e everyone submits preferences to one place, they run the algorithm, then inform people of their allocation.

One of the things 'Organisation of the Organisationless' argues is that networks make neat horizontal/vertical distinctions fuzzy. Like, is a highly followed twitter account a hierarchy? Not really, but then they're not really horizontal with the rest of a movement either (think of someone like DeRay during the Ferguson protests). The matching thing might be similar in that it's kind of a bit of both.

Pennoid's picture
Pennoid
Offline
Joined: 18-02-12
Jul 13 2017 17:32

Real quick: The allocation thing reminds me of how undergrad econ is rooted around utilitarian distribution theory (marginal econ) and not actually functioning markets and human interactions. It's abstract utopian. The 'systems' they propose actually all imply a central planner (0 transaction costs, universal knowledge, and holographic firms has the result of everyone constantly being on the same page - how is this different from an omnipotent and omnipresent god other than it being viscerally denied that it is such ad nauseum?)

On Engels, Kautksy and Freedom: The conception of freedom in Hegel and Engels (and Marx) requires a *mastery* of the necessary - an understanding of what is required or necessary, in order to understand and control it, subordinate it, and act beyond it, to escape, as it were, it's constraints. To make it more concrete; The division of labor in society results from the realization that one person, using tools, can produce more than enough food for himself. Having taken care of the *necessity* "food" frees up the labor of others to engage in, say philosophy.

In Socialism: Utopian and Scientific, Engels says that the change from capitalism to communism is the final leap from necessity to freedom in the realm of society; that rather than confronting humanity as 'natural laws' (those regulating capitalism) humanity will come to *regulate* material reproduction, plan production and distribution.

Kautsky writes (sorry for long quote in advance):

Quote:
Engels once expressed the opinion. . . that man’s lack o f freedom in history holds good only up to the present time, that it must necessarily disappear in a socialist society. [Kautsky quotes the passage from A nti-Duhring, pp. 309—10, to which he had devoted chapter 1:3:10, above.] The condition in which man confronts “the laws o f his own social activity. . . as external, dominating laws o f Nature” is abolished for production through the social regulation o f the latter. That is doubtless correct. But I cannot completely agree with Engels when he then draws the conclusion that men would thereby achieve control over their living conditions for the first time in
history. As far as this control is domination of nature, it has already been accomplished to a high degree by industrial capitalism. On the other hand, to the extent that it is control over the use o f the means of production by man (as producer and as consumer), according to plan, without crises, there was such control in early times prior to the emergence o f commodity production. Only the latter brought about anarchy in production and the subjection of men to its laws, which are effective without men being conscious of them. . . .
But Engels goes still further. From the fact that in socialist society men will themselves organize production with full consciousness, he concludes that they will thenceforth also make their history completely according to plan, so that “ the social causes set in motion by men will have, predominantly and in constantly increasing measure, the effects willed by men.” . . . The further social science progresses, the more accurately will m en be able to calculate in advance the effects of their new institutions, insofar as they function as man’s organs. But sooner or later, every one of these institutions, detached from the men by whom they were created, must become men’s environment, which they encounter as fully formed. The effects they have as such will presumably never be wholly predictable.
. . . In previous history, almost every new technological or economic institution has, in addition to the effects intended by their originators, also brought about unintended effects, if not always immediately, then after the passage of some time, effects that created new social problems. That was precisely what has kept the history of humanity in motion from its beginning onward, and there is no reason why the supplanting of commodity production should change this in any way. Certainly, the means for the solution of social tasks are im proving more and more… But along with those means, the tasks, too, are growing more and more. . . . The new institutions [primitive] men created for themselves remained predominantly their organs; only to a small degree did they become an environment determining those men. In contrast, how immense is the present-day technological apparatus o f society, how extensive are the states, how varied are the areas o f their involvement, how closely interwoven, internationally, is all of man's activity, not only their economic and political, but also their cultural one. Not the circumstances of a tribe numbering from two to three hundred people, but those of all of humanity must be taken into consideration today, if one wants to introduce a far-reaching innovation in men's life.
It will be an enormous task, indeed, to regulate according to plan the economic activity o f mankind. But we must not by any means expect that it will be possible to foresee all the consequences for men's entire noneconomic life, their scientific and artistic activity, their sexual and social relationships, that will arise out o f the efforts to solve that task, or to confront even before their birth the new problems that those consequences bear in their womb. We may assume that the social institutions the victorious proletariat will create will probably have the effect that the proletariat wants to achieve with them, and that it will realize its goal: the abolition of all exploitation. But we have no reason whatsoever to assume that with the solution o f the present-day problems of society, all social problems will be solved fo rev er,. . . But if the social problems of our time are solved, where then will new problems come from that will cause new movements, if the newly created conditions do not bring them forth, if these conditions do not, in addition to those effects intended when they were introduced, also contain effects that will one day make themselves felt, o f which, however, we do not yet have, nor can we have, any inkling?
The final goal of the proletariat is not a final goal for the development of humanity. The law of its motion always remains the same, though: the creation o f new institutions for the solution o f newly appearing problems. The new institutions solve not only these problems, but also contain in themselves new problems, which in turn make it necessary, in the interest of humanity, to create new institutions. And so on. Thus, like the process o f acquiring knowledge, that of social development is also unending— that is to say, a process that lasts as long as humanity, with its abilities and its existing natural environment, persists. An enduringly perfect society is as little possible as an absolute truth. And both the one and the other would mean nothing other than social stagnation and death."

Artificial Intelligence implies a common sci-fi problem; does the regulation of our material reproduction according to a plan require something *more* intelligent than us? (This has all sorts of problems, e.g. what is intelligence and how is it defined and so much more). I hope that clarifies.

The agent-marxism question is interesting; how do we theorize 'classes' if society is an organism? Or system? A network? Etc. Like in a society there is a division of labor and so too in an organism. But classes exhibit very specific phenomenon. You have 'classes' of people; their condition in the world with respect to their relationship to quantities of money and the division of labor. (how do you even theorize how the division of labor is organized?)

Again, in his book, Kautsky makes the (by now) familiar argument that humanity is an animal who's social organization and tool use are effectively forms of adaptation to environmental problems *prior* to natural selection - Kautsky calls them artificial limbs. Maybe systems science could have some insight into the relations between individual and class; between classes in society, between the technical and social divisions of labor etc.

Anwar Shaikh's recent book also argues that the 'classical economists' (Smith, Ricardo, Marx) understood capitalism to be a complex system with emergent properties (falling profitability/law of value, regulating capital etc.). He argues for supporting their idea of the market as a war between firms, and he outlines a rejection of much of contemporary microecon and the supposed (but abstract and utopian) reduction of macro to micro foundations. Instead he argues for the use of empiricism and straightforward models, many of which were pioneered by the classicals (though he draws on all of economic history for use and critique in the text).

Khawaga's picture
Khawaga
Offline
Joined: 7-08-06
Jul 13 2017 17:42

Yes, that clarifies it. Thanks.

Is the Shaikh's book you're referring to Capitalism: Competition, Conflict, Crises?

Pennoid's picture
Pennoid
Offline
Joined: 18-02-12
Jul 13 2017 18:24

Yes.