Cyber communist planning
- This topic has 11 replies, 4 voices, and was last updated 1 year, 10 months ago by
.
-
Posts
-
The article here seems to be a very good guide to the mathematics needed to plan a large scale economy, and the feasibility thereof. I’ve only skimmed it, as I think it’ll need a bit of close study to fully understand.
“At CibCom, we aspire to transcend this situation; we want to guide novices who venture into the (seemingly obscure) mathematical and computer science fundamentals of economic planning without ever having specialized in it or not having worked on it for some time. The general public may consult the document when they encounter difficulties in interpreting an algebraic idea or expression in any of these texts. It is especially extensive because we have not wanted to skip any of the explanations that are usually taken for granted in the treatises on this subject.”
“Finally, let us underline something that is not always made explicit: you do not have to be a cybernetic genius to discover the “magic” behind this research program. The vast majority of us will not have to invent any mathematical techniques or readjust the computational complexity of any five-year plan. Our initiative to disseminate it in this depth is intended, first and foremost, to make larger portions of our class aware of its existence, that we understand how hopeful our proposal is. This is our message
to all of us who feel the shackles of capital: Comrades, there are possibilities for our long-awaited democracy, that seed that germinated slowly for who knows what future harvests, and whose sprouts will not take long to burst the earth!”It might actually be worth organising some sort of study group around….
I have had a quick look and it seems to be an argument in favour of central planning as practiced in historical state capitalist economies (I saw references to the USSR), a demonstration that it could work under ideal conditions (which were absent in the USSR). I could be wrong but that’s what the articles in Spanish seem to be suggesting.
I thought that Pieter Lawrence and Robin Cox had convincingly argued that this this would not be necessary in socialism and that for most consumer goods and services production and distribution could be more or less self-regulating, operating like the market today except that demand would be real need not paying demand.
In this case, central planning would only need to be indicative or apply only to infrastructure projects.
I think Cockschott, whose work they seem to be influenced by, would argue that if it can be designed as a system at the firm level, it can be run by computer globally. I believe even Lawrence’s model requires ‘statistical clearing houses’ so some of the mathematics deployed here may come into play there. I certainly think Kantarovich’s polynomial spaces for optimal productive allocation may well play a part.
Where I disagree with them is that I think machine computers aren’t essential, what matters is the underlying methodology.
So far I have just looked at the introduction to this piece (which I agree is worth studying). I noted straightaway that it seems to repudiate the concept of a posteriori feedback as a means of spontaneous coordination and so, by implication, is advocating for an idealised system of society-wide planning. This is apparent in this comment:
“In capitalist society, characterized by the private ownership of the means of production, conscious coordination is non-existent and organization occurs at the atomic level (in companies). Capitalist planning, in spite of how much it has been technified in the last decades, occurs only within individual companies and, more importantly, it is fundamentally oriented towards profit expectations. Between different private companies, it is no longer that there is no harmonious planning, it is that there is no planning at all. Only a posteriori, and according to the logic of blind and impersonal market
automatism, can the different productive units be coordinated to supply the demands of the people. These demands (which are sometimes the most basic human needs) will be satisfied, or not, exclusively according to the level of income of each person and the availability of goods that each country has in the global supply chain.”Yes, as YMS rightly says, their work seems to be influenced by Cockshott and bears the same muddled thinking on the matter that Cockshott displays in his writings. Cockshott, on the one hand, seems to want a system of society-wide planning yet on the other, talks about his system being able to accommodate feedback or a posteriori decisionmaking
The logic of society-wide planning calls for a globalised input-output matrix in which the supply and demand for every conceivable good are matched up through a process of material balancing. It’s a sort of general equilibrium model of the production system. To talk of “clearing houses” in the context of such a model implies the application of feedback which is not really compatible with the idea of society-wide planning.
Input-output matrices are what is called a consistency model. In idealised terms, they eliminate waste by ensuring consistency in supply and demand in every case. This is different from linear programming which is an optimisation model in the sense that it plots a course of action that ensures maximised output or alternatively minimised costs (or both).
Linear programming can operate at different scales – from the small scale to the large scale involving thousands of variables (for example by identifying bottlenecks in a complex transportation system like the London Underground to increase the overall efficiency of the system). But that is far removed from society-wide planning
It always puzzles me when people like Cockshott and others seem to put forward arguments that appear to advocate for society-wide planning while repudiating spontaneous coordination or feedback (which they quite mistakenly equate with the market – the market is just one example of this but you can equally have non-market examples).
Perhaps, the greatest failing of these people is that they approach the whole question of a communist or socialist society from a thoroughly technocratic perspective. They don’t take as their starting point, the kind of social relationships that will pertain to such a society
To me, it is as clear as daylight that what they are advocating is fundamentally incompatible with the nature and entire ethos of such a society. They talk vaguely about the democratisation of planning but how is that remotely possible given the immensity of what needs to be planned under a system of society-wide planning?
While there is a certainly a role to play for planning tools like input-output matrices or linear programming we should be very wary of fetishizing these procedures
Well, we will need society-wide planning, if we want to build, say, a railway across the Bering Straight, or to move everyone into Texas while we re-wild the earth, and we’ll need some sort of co-ordination against external shocks. Lawrence’s model includes a role for the FAO as some sort of worldwide co-ordinating body, and similar other such bodies, so there is a role for this sort of planning.
Yes but that is not what is strictly meant by society-wide planning, YMS. Building a railway across the Bering Strait is certainly a very big project involving large-scale planning. Undoubtedly an element of this will be needed in a socialist society although I am bit wary of the big-is-beautiful school of thought. Big dam projects, for example, have often turned out to be ruinous for displaced people and ecologically disastrous.
But that is something different. Society-wide planning means the suppression or elimination of all independent planning below the level of global society and the replacement of a polycentric planning system with a unicentric planning system and a single society-wide plan. Pieter Lawrence’s model certainly did not envisage this but a spoke instead of a tiered (i.e. polycentric) system operating at global, regional, and local levels.
I understand the point you are making but we should be careful about not using terminology that might convey a completely false and misleading idea of what we are talking about. “Society-wide planning” is one of those terms….
The paper also references Stafford Beer, so they must have the viable system model in mind. so subordinate units would still have relative autonomy within the needs of the wider system.
Beer’s concept was structures should be recursive, so global planning is possible, but in a steady state sub-systems would run pretty much without interference. In the chart above, system 5 would be the population as a whole, through worldwide bodies.
Speaking of Stafford Beer, he turns up in this short film I watched recently “Cybernetics and Revolution”
https://endnotes.org.uk/posts/groupe-brume-cybernetics-and-revolution
From watching this film and reading books like ‘Seeing Like a State’, I’m pretty sure that this kind of thing is a non-starter for us. Human political society, and the world, is more complex than can be modelled by any of the cyberneticians mathematics.
I’m no expert on this topic though, so persuade me otherwise.
Well, these sorts of systems are necessary to some extent in any widescale organisation, computerisation doesn’t add magic, but it does add interconnexion and co-ordination: if we’re going to have trainlines running or nuclear power plants, we will have to have vertically organised command control systems to do so, and the same could be applied to the whole logistics chain. After all, we will have Amazon’s systems to use for ourselves once we change over, absent the compulsion. That plural and myriad organisations can organise themselves means they can link up and effectively function together.
Here’s what the Zeitgeist Movement has to say on this. They seem to be siding with the “cyber communists”.
“Two people are born every second on this planet, and each one of those humans needs a lifetime of food, energy, water and the like. Given this fundamental need to understand what we have, the rates of depletion and, invariably, the need to streamline industry in the most efficient, productive way, a Global System of Resource Management must be put in place. It is just common sense. This is an extensive subject when one considers the technical, quantitative variables needed for implementation. However, for the sake of overview, it can be stated that the first step is a Full Global Survey of all earthly resources. Then, based on a quantitative analysis of the properties of each material, a strategically defined process of production is constructed from the bottom up, using such variables as negative retroactions, renewability, etc. (More on this can be found in the section called Project Earth in the ZM lecture called “Where Are We Going?”). Then consumption statistics are accessed, rates of depletion become monitored, distribution is logically formulated, etc.. In other words, it is a full Systems Approach to earthly resource management, production, and distribution, with the goal of absolute efficiency, conservation, and sustainability. Given the mathematically defined attributes, as based on all available information at the time, along with the state of technology at the time, the parameters for social operation within the industrial complex become self-evident, with decisions “arrived at” by way of computation, not human opinion. This is where computer intelligence becomes an important tool for social governance, for only the computation ability/programming of computers can access and strategically regulate such processes efficiently, and in real time. This technological application is not novel. It is simply ‘scaled out’ from current methods already known.”
“In other words, it is a full Systems Approach to earthly resource management, production, and distribution, with the goal of absolute efficiency, conservation, and sustainability. Given the mathematically defined attributes, as based on all available information at the time, along with the state of technology at the time, the parameters for social operation within the industrial complex become self-evident, with decisions “arrived at” by way of computation, not human opinion.”
Peter Josephs’s thought seems to have developed since then. I wonder what he makes of statements such as this now? Anyone here know?
He is certainly still opposed to electing people to take technical decisions. But in episode 12 of his podcast series Revolution Now! (2 December 2020) he does speak of “economic democracy” but it seems to be limited to taking part in designing the system:
“In a future economy that’s truly efficient, it will be truly integrative and you will have an economic democracy where people participate through CAD systems and design, as I touched upon the prior podcast. They will engage a network like Amazon, and such robust infrastructure that Amazon currently possesses would be utilized for efficient and sustainable design production and distribution and recycling.”
This would seem to be a change of language from what he was saying ten years when he didn’t even mention the word “democracy”.
- You must be logged in to reply to this topic.