保罗·科克肖特（Paul Cockshott）是苏格兰格拉斯哥大学的准教授、计算机专家。他们的主要研究领域包括数组编译器（array compilers）、经济物理学（econophysics ）和可计算性（computability）的物理基础等。他撰写了一系列书籍，其中包括《走向新社会主义》（Towards a New Socialism ）和最近的《经典经济物理学》（Classical Econophysics）一书。

本访谈最初发表于“矛盾的精神”（Spirit of Contradiction）网站。

采访者的问题为粗体。

问：你最近在研究经济学力学。你可以简单介绍一些什么是经济物理学，以及它如何与社会主义者相关吗？

答：我明白，它可能看起来有点晦涩。但是你得知道，马克思说过他打算发现资本主义运行的规律——这是一个深受物理学影响的观点。什么是经济物理学？物理学发展了一套研究高自由度系统的概念装置。经济物理学主要讨论的就是用这一套概念装置来理解经济现象的尝试。因此，经济物理学大量借用统计力学（statistical mechanics）中发展起来的理念。

经济物理学有两个主要的来源。数学家Farjoun 和Machover （他们在1983年出版了《混沌规律：政治经济学的概率论方法》<Laws of Chaos, A Probabilistic Approach to Political Economy>一书，将统计力学的方法引入到经济学研究中。该书得出了价格和劳动量的关系、利润率分布和劳动生产率提高的趋势都服从概率规律的结论。反对一般利润率的假定，认为《资本论》第一卷的劳动价值论更符合经济现实，从而<用作者的话说>“消解”了马克思的转型问题。这本书被认为是经济物理学的开端。——译者注）把资本主义经济作为一个混沌系统进行研究，基于统计力学，将价格的出现归结于劳动价值。最近，有许多物理学毕业生进入金融部门工作。他们把自己的概念背景应用于金融业的经济问题。（物理学家到华尔街工作早已不是新闻，中文媒体也有报道。

由于后一种工作主要由雇主们付酬，所以倾向于聚焦金融市场，但是它创造了一个机会。以数学和物理学为主要学科背景的人，已经开始不带预先形成的意识形态来审视经济。如果他们获得了经济学学位，就可能会有这种先入为主的意识形态。相比经济学家而言，这就生产了更多以经验为依据，更少基于意识形态的研究。

经济物理学与社会主义者的相关之处在于，你可以使用这些方法来分析资本主义，理解为什么劳动价值论站得住脚，以及为什么收入分布变得如此扭曲。

问：你可以简单描述下你认为在这个领域中已经发现的重要结果吗？

答：我觉得，重要的研究结果是：

1，劳动价值论基本上是准确的。

2，任何市场制度都有一个非常不均匀的收入分布——甚至工人持股企业的制度下，也是如此。

3，然而，相对于如上考察所预期的结果，现存的制度收入分布更加不均匀（美国马里兰大学物理学教授Victor Yakovenko在一次访谈中解释了为什么现实的资本主义制度收入分布更不均匀：

由于后一种工作主要由雇主们付酬，所以倾向于聚焦金融市场，但是它创造了一个机会。以数学和物理学为主要学科背景的人，已经开始不带预先形成的意识形态来审视经济。如果他们获得了经济学学位，就可能会有这种先入为主的意识形态。相比经济学家而言，这就生产了更多以经验为依据，更少基于意识形态的研究。

经济物理学与社会主义者的相关之处在于，你可以使用这些方法来分析资本主义，理解为什么劳动价值论站得住脚，以及为什么收入分布变得如此扭曲。

问：你可以简单描述下你认为在这个领域中已经发现的重要结果吗？

答：我觉得，重要的研究结果是：

1，劳动价值论基本上是准确的。

2，任何市场制度都有一个非常不均匀的收入分布——甚至工人持股企业的制度下，也是如此。

3，然而，相对于如上考察所预期的结果，现存的制度收入分布更加不均匀（美国马里兰大学物理学教授Victor Yakovenko在一次访谈中解释了为什么现实的资本主义制度收入分布更不均匀：

问：“利润率下降趋势”是一个追溯到亚当斯密的想法。你可以快速介绍一下吗？

答：是的。斯密最早注意到利润率具有下降的趋势。他将其归结为资本家们挤进某个行业，从而导致价格降低。马克思则认为，这不是影响某个行业，而是影响整个经济的趋势。他谈到他称为“绝对过度积累（absolute overaccumulation）”的现象。当资本存量的增长比劳动力增长快的时候，就会出现这种现象。

他认为，这一现象一定会倾向于推动利润下降。理由两个。首先，它造成了对工人的争夺，使得工资上涨；其次，由于劳动是利润的来源，如果工资对劳动的比例提高，那么每投资一英镑资本所能得到的利润就一定会减少。

这对于发达国家很重要，因为它们的人口自然增长率很低。这就意味着，随着时间的推移，利润率倾向于降到很低的水平。就像我们在日本看到的那样。

问：你觉得利润率下降趋势与欧洲当前的危机有何种关系？

答：与其说它是危机的直接原因，不如说是背景原因。说它是背景原因，是因为相比当前的危机而言，七十年代和八十年代的危机更直接地是由于利润率下降趋势。积累率的降低使得九十年代出现的部分复苏成为可能。这就意味着，资本存量的实际价值增长得更慢了，实际上相对于劳动人口来说下降了。然而，副作用是利润不在作为实际资本进行积累，而是投资到金融系统，转变为对国家和消费者的债务。2008年后这些借款人的信用价值的耗尽，意味着利润既没有被借出去，也没有被投资到生产性领域，这造成了衰退。

问：在未来的十年里，这对社会主义者来说意味着什么？

答：嗯，我认为，这里重要的考虑因素是中国的利润率下降。向中国的工作输出一直是顶在这里工会头上一把枪。中国利润率的下降会减弱这种威胁。更普遍而言，这意味着八十年代出现的积累的矛盾的解决办法，正在耗尽自身。而且，基于老的新自由主义原则来重启欧洲经济，也是非常困难的了。

这样的重组危机为进步政治创造了潜在的空间，如果劳工运动有一个替代性经济战略的话。

问：我们知道，安德鲁·克莱曼（ Andrew Kliman，美国佩斯大学戴森艺术与科学学院教授,是左翼经济学家、著名的马克思主义理论家及分期单一系统解释<Temporal single-system interpretation,TSSI>学派的代表人物之一。2012年,克莱曼出版了研究2008年全球金融危机及之后的大衰退的《大失败:资本主义生产大衰退的根本原因》一书,全面阐释了当前的危机是资本主义发展的结果。——译者注）对利润率下降趋势以及对这一过程的抵消因素提出了一些观点。具体来说，资本的毁灭（通过物理的破坏或者陈旧过时而被淘汰）或者生产资本中用于生产资本自身的量的减少造成的资本价值的降低，可以保持住资本的有机构成。

答：科技进步往往会通过加速过时资本存量的折旧降低资本劳动比。我倾向于同意这一点。不过，要把收益率恢复到更高的比如说1960年代的水平，需要劳动生产率前所未有的增长，而这是不太可能的。

问：自动化不是一个新现象。不过，考虑到越来越多以前是人干的活现在都被机器干了，自动化不断增长的应用范围似乎使得实现充分就业变得不太可能。你认为这只是一个暂时现象，就跟18世纪晚期引入动力织机时一样吗？你认为当前的自动化会造成新的重要问题，或者能够提供阶级斗争中新的机会？

答：我认为把失业归结于自动化是一个错误。自动化已经进行两百年了。它导致劳动从一个领域切换到另一个领域，而不是减少雇佣的总体水平。就业的下降总是总需求不足的直接结果，而这又反过来被经济中的收入分配以及利润率，而不是被自动化水平所决定。

问：你如何看待最低基本收入提案，考虑到生产足够多的使用价值所需要的劳动量在减少，以及出于环境方面的原因不管怎样都得降低经济增长率。

答：我倾向于反对这些提议。在资本主义制度下，这意味着对低工资被雇佣者提供补贴，而这个补贴是从其他工人通过税收支付的。相反，劳工运动应该要求实现他们的劳动所创造的全部价值。

问：苏联有相当大比例的人口参与到计划工作，但是他们应付的是一个不太先进的经济，产品比我们现在少得多。你认为我们可以对与现存的商品组合同样多样化的经济进行计划吗？

答：我认为真实情况不是这样。我知道的是，苏联国家计划委员会（GOSPLAN）只有区区几千人。而且，我们现在有了好得多的信息技术，所以我不认为计划体制所需要的劳动会是一个问题。

问：苏联在质量控制方面存在严重问题。由于投入品的质量低劣，有时候工业部门会自己再生产投入品，或者为了完成计划任务，找一些能办事的人非法地寻求供货来源。资本主义有时候会通过生产商之间的竞争来解决这一问题。生产商可以从许多可选的来源获取零件。虽然有时候资本主义在这方面也会有严重的失败，但是一般说来，不会是如此系统性的问题。

如果生产者不能独立地寻求产品供货来源，我们如何确保一个可以应付质量控制的问题的更合理的计划体制。

答：我认为，采取更加实时的计划体制能够部分地解决这个问题。这样的话，生产单元就能够更频繁地基于新的投入品来注册新的生产计划，效果与切换供货商差不多。另一方面，经济以接近完全产能运转会造成一些影响。即使采用新的计划体制，也很难减弱这种问题。

问：替代性经济方案已经有好几个了。其中包括“参与型经济”（Parecon：Hahnel 和 Albert出版的著作《Parecon: Life After Capitalism》一书中提出的经济方案——译者注），荷兰国际共产主义团体（GIK， Group of International Communists of Holland。该组织在1930出版的《共产主义生产和分配的基本原则》（Fundamental Principles of Communist Production and Distribution）一书是第一本根据马克思《哥达纲领批判》来勾画社会主义经济蓝图的著作——译者注），你在《走向新社会主义》（Towards A New Socialism，少年中国评论网站有该书中文版连载——译者注）一书中提出的方案，还有包容性民主（Inclusive Democracy，希腊经济学家Takis Fotopoulos提出的经济方案——译者注），以及其他许多方案。无论这些方案如何可取，都存在难以付诸实施的问题。你怎样想象我们实际实施这些制度？

答：困难有两种：1.政治的，2.组织的。第一个困难更重要。有了政治意愿，组织问题都可以解决。人们可以在尝试组织经济的过程中根据经验修改推荐的经济模型，直到他们找到一个在特定的情况下能够实现的版本。

问：你认为某种形式的互助主义（mutualism），把合作社合在一起，建立某种与资本主义共存的计划形式，可以作为过渡形式吗？

答：可以。但是只有当立法支持把大部分经济都转换到合作的基础上，而不顾当前股东意愿的情况下才行。

问：目前，迈克尔·海因里希(Michael Heinrich，他是德国《社会科学批评杂志》的主编——译者注）和其他价值理论家在社会主义者圈子里很受欢迎。你如何看待海因里希提出的价值理论（value theory）？

答：比起海因里希，我对英国和美国的价值形式理论家更熟悉。但是我感觉到，价值形式理论对当代经济学让步太多，而且在价值创造的问题上，赋予市场过多的力量，这是不符合实际的。我对他们的方法感到不快的地方在于，他们给社会必要劳动赋予了太多意义，以至于如果使用他们的说法，就会使得劳动价值论变成非科学的东西。

通过与他们的讨论，在我看来，他们认为商品在某个给定价格的销售建立起了物化在其中的劳动的社会必要性。但是如果真是如此，那么，就没有检验善品价格是否决定于其中的劳动量的独立的办法。他们最终得到了这样一个理论，是价格决定了他们当做是劳动内容的东西。与其说这是劳动价值论，不如说是劳动价格论。科学的因果论要起作用，如果我们说A导致B，那么A和B必须能够独立地测度。如果你只能通过测度B来测度A，那么甚至A存在的推断都变成不必要的，这当然是萨缪尔森反对劳动价值论的理由。

经济物理学的方法是，A（劳动量）和B（产出销售的货币流）都可以采用实证方法测度，而且我们能够说明，A的变动导致了B的变动。

问：麻省理工学院的Jonathan Gruber教授最近在一个在线微观经济学课程里说，“经济学基本上是一个右派科学…我们的模型永远假定市场最了解情况，而且会一直被灌输这种基本立场。”你认为经济学的确像Gruber说的那样有偏见吗？

答：是的。

问：你认为，就经济学作为一门科学而言，社会主义者应该关注哪些重要的领域？

答：我认为，要取得进展，你必须永远面对针对正在实际上发生的实际现象提出因果模型。理论发展必须来自于与实际现象的遭遇。你必须审视当前所发生的事情，尽力用你当前的概念装置对其建模。经常会发现你不能。这时候，你就需要发展理论，但只有在面对经验事实的情况下，你才能发展。为了发展理论，你必须从其他领域借用概念，但是这种借用必须在之前已经存在的实际问题的引导之下进行。

问：如果你年轻的社会主义者对经济学感兴趣，你会建议他们怎样参与到社会主义经济研究计划的学习和推进中来？

答：我强烈建议他们对生活于其中的经济做局势分析（conjunctural analysis）开始。

附件：访谈英文原文

Interviewer questions are in bold

You’ve done work recently in Econophysics, can you give a brief introduction to what that is and how it’s relevant to socialists?

I can see that it may seem a bit obscure, but you have to remember that Marx said he was out to discover the laws of motion of capitalism–a sentiment very influenced by physics. What is econophysics? Well in the main it covers any attempt to understand economic phenomena in terms of the conceptual apparatus that physics has developed for the study of systems with a high degree of freedom. As such it borrows heavily from ideas developed in statistical mechanics.

It has originated from two main sources. The mathematicians Farjoun and Machover who studied capitalism as a chaotic system and deduced the emergence of prices which would be proportional to labour values on statistical mechanics grounds. More recently there has been an influx of physics graduates into jobs in the financial sector where they have applied their own conceptual background to economic problems.

Since this latter work is paid for by their employers in the main, it tends to be rather focused on financial markets, but it has created an opening whereby people with a primarily mathematical or physics background have started looking at the economy without the prior ideological formation that they would have gotten from an economics degree. This creates studies that are much more empirical and less ideologically based than what economists tend to turn out.

Its relevance to socialists is that you can use these methods to analyse capitalism and understand why the labour theory of value holds, and why income distribution becomes so skewed.

Can you briefly describe what you think are important results that we’ve found in this area?

To my mind the important results are:

That the labour theory of value is basically accurate.

That any market system has a pretty uneven distribution of income–this would apply even to a system of worker owned firms.

That the existing system, however, contains an even more uneven distribution of income than would be expected just from the considerations above.

The “Tendency of the Rate of Profit to Fall” (TRPF), is an idea that I believe goes back to Adam Smith. Can you give a quick description of the idea?

Yes–Smith originally noticed that the rate of profit tended to fall. He put this down to capitalists piling into a line of business and driving the price down. Marx argued that this is not something that affects just individual lines of business but the whole economy. He spoke of what he termed ‘absolute overaccumulation’ which occurs when the capital stock builds up faster than the growth of the workforce.

He argued that this phenomenon must tend to drive down profits for two reasons. Firstly it generates competition for workers and allows wages to rise; secondly since labour is the source of profit, if there is a rise in the capital to labour ratio, then the amount of profit for each £1 of capital invested must go down.

This is important for developed economies since they have a low rate of natural population growth and that means that over time the rate of profit tends to fall very low as we see in Japan.

How do you think the TRPF is relevant to the current crisis in Europe?

It is a background cause rather than an immediate cause. Background, because the crisis of the 70s/80s was more immediately due to the falling rate of profit than the current one is. The partial recovery that occurred in profitability in the 90s was made possible by a lowering of the rate of accumulation, which meant that the real value of capital stock grew more slowly, or due to depreciation, actually fell in proportion to the working population. The side effect though, was that the profits which were no longer accumulated as real capital, were invested in the financial system and converted into loans to the state or to consumers. The exhaustion of the credit worthiness of these borrowers after 2008 meant that profits were neither being lent out nor reinvested productively, which generated the recession.

What does this mean for socialists in the coming decades?

Well, in my view an important consideration here is the declining rate of profit in China. This will make the threat to jobs in Europe from the export of the jobs to China will be less of a gun to the head of the trades unions here. But more generally it means that the mode of resolution of the contradictions of accumulation that came into place in the 80s is exhausting itself, and it is very difficult for the European economy to re-start on the basis of the old neo-liberal formulas.

Such restructuring crises create potential openings for progressive politics if the labour movement has an alternative economic strategy.

I know that Andrew Kliman has made some statements about the TRPF and countervailing tendencies to this process. Specifically, the destruction of capital (either through physical destructor or obsolescence) or its decline in value due to the lower amounts of labour in producing capital used in production itself can keep the organic composition of capital.

I would tend to agree that the improvement in technology does tend to lower the capital labour ratio by accelerating the depreciation of obsolete capital stock. However, to restore profitability to–say–the levels of the 60s would require an unprecedented rate of growth in labour productivity which is unlikely.

Automation is not a new phenomenon. However, its increasing scope seems to make it unlikely we will attain full employment, considering that a lot of the tasks previously done by people are now done by machines. Do you think this is a merely temporary phenomenon, as it was with the introduction of the power loom in the late 18th century? Do you think our current stage of automation could cause significant new problems or offer new opportunities in the class struggle?

I think it is a mistake to attribute unemployment to automation. Automation has been going on for two hundred years now, and results in a shift in labour from one area to another rather than a reduction in the overall level of employment. A decline in employment is always the direct result of a slackening of aggregate demand, which in turn is influenced by the distribution of income within the economy and the rate of profit rather than the level of automation.

What do you think about minimum basic income proposals, given that sufficient use values can be produced with decreasing amounts of human labour, and that there may be environmental reasons to curtail economic growth in any event?

I tend to oppose these. Under a capitalist system they amount to a subsidy to low wage employers and a subsidy that other workers pay out of their taxes. Instead the labour movement should be demanding the full value that their labour creates.

The Soviet Union had a quite large percentage of the population involved in planning, and yet they were dealing with a much less advanced economy with far fewer goods than we have at present. Do you think we can plan with as diverse an array of commodities as currently exists?

I don’t think this is true. My understanding was that there were only a few thousand people employed in GOSPLAN. Moreover, we now have a much better information technology so I don’t think that the labour required by the planning system would be a problem.

The USSR had serious problems with quality control. It was sometimes the case that industries would reproduce the input products themselves because of shoddy quality, or source them illegally through fixers in order to make planning schedules. Capitalism sometimes deals with this problem by having competing producers which allow producers to source parts from any number of alternatives. Sometimes capitalism also has serious failures along this dimension but it is not generally such a systemic problem.

How can we ensure a more reasonable system of planning that will deal with the problems of quality control if products can not be sourced independently by producers?

I think that in part this can be dealt with by moving to a more real time planning system so that units of production can register new production plans based on new inputs more frequently, thus getting much of the effect of switch in suppliers. On the other hand there are some effects that come from running the economy at close to full capacity that may be hard to mitigate even in these circumstances.

There are quite a number of alternative economic proposals, including Parecon, the GIK proposal, your proposal from your book Towards a New Socialism, Inclusive Democracy and many others. Whatever one thinks about the desirability they all have the problem of being difficult to put into practice. How do you imagine we can practically implement these systems?

Well, there are two kinds of difficulty: 1. Political, 2. Organisational. The first is the more important. Given the political will the organisational problems would be soluble. People would modify the suggested economic model in the light of experience trying to organise it until they had a version they could attain in the given circumstances.

Do you think some form of mutualism, where cooperatives come together in order to institute some form of planning coexisting with capitalism might be a transitional form?

Yes, but only if there was the legislative support to convert a large portion of the economy to a mutual basis, irrespective of the wishes of current shareholders.

Currently Heinrich and other value theorists are enjoying a certain popularity in socialist circles. What do you think of value theory as put forward by Heinrich?

I am more familiar with the English and American value form theorists than Heinrich but I feel that value form theory concedes rather much to contemporary economics and attributes more power to the market in creating value than is realistic. My unhappiness with their approach is that they overload the meaning of socially necessary labour in such a way that, were their meaning to be used, they would make the labour theory of value unscientific.

From discussions with them, it seems to me that they hold that it is the sale of commodities at a given price that establishes the social necessity of the labour embodied in them. But if that is the case then there is no independent way of checking whether the price of commodities is determined by their labour content. They end up with a theory in which it is prices that determine what they count as labour content, and you end up with a price theory of labour rather than a labour theory of value. For a scientific theory of causality to be of any use, if we say A causes B, then A and B must be independently measurable. If you can only measure A by measuring B, then the inference that A even exists becomes unnecessary, which of course was Samuelson’s objection to the labour theory of value.

The econophysics approach is that both A (labour content) and B (money flow from sales of output) are in fact empirically measurable, and that we can show that variations in B are caused by variations in A.

Jonathan Gruber, an economics professor at MIT recently said in an online microeconomics course, “Economics is fundamentally a right wing science…our models always assume the market knows best, and you are going to be indoctrinated into this basic position.” Do you think that economics is as biased as Gruber makes out?

Yes.

What do you think are important areas for socialists to be looking at in terms of economics as a science?

I think that to make progress you always have to face up to getting a causal model for some real phenomenon that is going on in the world. Theoretical advances have to come from a confrontation with real phenomena. You have to look at what is happening and see if you can model it with your current conceptual apparatus. Often you will find that you cannot. At that point you need to develop the theory, but it is only in confrontation with the empirical that you can advance. To make the advance you may have to borrow concepts from other domains, but this borrowing has to be guided by pre-existing real problems.

If young socialists are interested in economics, how would you suggest they get involved in learning and advancing a programme of socialist economic research?

I would strongly suggest that they start off by doing a conjunctural analysis of the economy they live in.