The Discovery of Money

Money  shares with  things like time and space the sort of obvious-mysterious quality that can utterly puzzle us.  Do we need a philosophy of money? I think we do.  Today’s financial crisis reminds me of the case of Bill #240 introduced in the Indiana legislature in 1897, which attempted to define Π (pi) as having the value 3.2, a kind of deep silliness that arises from understanding mathematics technically without understanding it philosophically. Imagine if we’d lacked an intuitive visual understanding of the idea of a circle, and the wheel had evolved like money in a universe where the Indiana episode was not a historical joke:wheelmoney

When you attempt to impose the arbitrary and contingent on things which have a hidden fundamental character, you run into trouble. In our hypothetical hexagonal-wheel world (one with no natural circles to inspire us, with Π set to 3), I imagine sophistry would have reigned. People might have been tried as criminals for traffic accidents caused by a too-hasty adoption of heptagonal or octagonal wheels (complaints in legislatures about “obscure derivative wheels that nobody understands”).  Ayn-Rand-like wheel philosophers would have argued for a return to square wheels. This is not to condone crime, greed and stupidity but to point out that in hexagonal-wheel world, it would be easier to be criminal, greedy or stupid. Better truths mitigate all  human failings. Einstein noted that the significant problems of life are not solved at the same level that we create them. Do we need to elevate the notion of money from the level of cultural construct, where we created our problems, to the level of universal fundamental, where we might be able to solve them? I think we do, and the first step is to start thinking of money as something yet to be discovered, like the concept of zero, or the idea of Π (or equivalently, circularity). It is not something to be invented.

Towards a Real Philosophy of Money

When philosophers consider money at all, they seem to think of it merely as an example of the construction of social reality. John Searle, in a book of that name, talks money-metaphysics, but really, he talks about language and communication, and money merely serves as an example of how we create and sometimes physically manifest shared meaning. There is one book apparently considered the classic, George Simmel’s The Philosophy of Money, which is available on Google Books. I’ve started reading it, but I find its approach dissatisfying; there is not as much ontology or metaphysics as I think is necessary. We hint at money being different when we say things like “time is money” or “money is power” or “money is the root of all evil,” but generally we don’t poke to see if there is more to these epigrams. Money is unique among human creations in attracting such metaphysical remarks. We do not remark that the wheel is the root of all evil.

We don’t seriously suspect that money might have fundamental characteristics because it seems like an invention. Designing a currency seems like a practical-aesthetic endeavor, like designing a car, that is only distantly constrained by physics, logic and mathematics. With cars, you have to dig quite a bit before you get to the underlying physics constraints (from wheels with bearings, axles and tires to the geometry of the circle; from the internal combustion engine through the Carnot cycle to the laws of thermodynamics). With money, we hit a wall pretty quickly as we dig deeper. Much of the complexity of money is in the direction of emergence: stocks, debt, derivatives, banking systems and so forth, phenomena analogous to the impact of automobiles on the sociology of suburbia and the distribution of gas stations. You have to assume a basic notion of currency as a given.

Money is closer to something like the wheel. There is much less meaningful/functional diversity possible in wheels than in cars. While money shares this low-diversity characteristic, it is not merely closer to the fundamentals. It is a fundamental. This is clear from the evolutionary direction towards simplicity. Originally, wheels were presumably just sections of round trees. Things got more complex with axles, bearings, rim-grooves, spokes and tires. With money, things went the other way, groping towards a simpler, more fundamental essence. Where possible money has shed the arbitrary and gotten simpler. In the move from barter to commodity, it shed most of its diversity. From commodity to token (backed by commodities like gold, but otherwise of nominal value), it traded intrinsic value (a variable number) for the simpler number ‘zero’ (at least as a design target). As communication improved, geographically-enforced diversity in otherwise economically-integrable regions gave way to unified currencies. From token to fiat, money shed its commodity backing and association with an arbitrary contingent reality (the scarcity of gold on earth). Each of these simplifications has had its discontents, but the evolution seems to me to arise from natural dynamics rather than some sort of political “ever simpler” camp winning.

Inventions differentiate, specialize, cross-breed and proliferate by natural selection. Fundamental discoveries converge (to cite Einstein once more) to “as simple as possible but not too simple” ideal. When your fundamentals are wrong (as with the five elements of alchemy that preceded the periodic table), your efforts to engineer an ecosystem of invention around them end up deeply flawed. Something like that seems to be happening today. We are engaged in fruitless quests for miraculous means to turn lead into gold and find elixirs of immortality.

If we reframe our exploration of money as a process of fundamental discovery rather than creative invention, we will, I believe, have achieved the right elevation of context to solve our problems. An example of the problems inherent in thinking “invention” is the following. Local currency advocate Thomas Greco said in an interview (as best as I can recollect the quote), “Money has gone from being our slave to being our master, and we have to make it our slave once more.” This sort of moralizing is just barely defensible for true inventions (such as automobiles), but it is entirely indefensible for discoveries. This sort of attitude turns metaphysical problems into moral ones, and engineering decisions (the optimal degree of localization of  currency systems) into religious ones. Money can be neither our master, nor our slave, any more than time can.

So yes, we need a philosophy of money, just as we have a philosophies of other conceptual primitives like time, space, consciousness and information. To finish the analogy to the wheel, money already has the equivalent of physics (economics), engineering (banking) and technical-vocational knowledge (accounting).  To start with, we must ask, as we do of any discovery quest, what is the most natural concept of money. The question we’ve been asking so far (and have never gotten beyond) what is the best design for money? is the wrong first question.

p.s. my opinion of Simmel’s book is tentative. I might change my mind as I read more

Only 1169 words!

About Venkatesh Rao

Venkat is the founder and editor-in-chief of ribbonfarm. Follow him on Twitter


  1. Maybe I’m totally missing the point here, but isn’t money in a sense just an abstracted version of the sort of inter-personal accounting that is associated with and required by reciprocity (especially indirect reciprocity) as it has evolved among humans? A bit of googling turns up these papers on the evolution of indirect reciprocity and money as an externalization of confidence (not in English, unfortunately). If nothing else they are at least suggestive of a potentially fruitful approach to discovering the true nature of money.

  2. Sorry, I messed up the link for the second paper.

  3. Frank:

    The first line of thought worries me because it builds in notions of society, agency and subjective value as fundamental. In fact the biological examples of direct/indirect reciprocity (such as the famous example of bats sharing food) show that those phenomena don’t actually need money to happen, just enough memory to play the underlying information games such as prisoner’s dilemma. I think the money may well lubricate and accelerate such phenomena (because it reduces the transaction cost known as the “dual coincidence of wants“), but it would be a emergent effect of money, not a part of the definition.

    The second one is more fundamental, since “confidence” is a type of information that can be encoded and therefore (as you say) externalized. I have been groping along the same lines myself.

    I think what will turn out to underlie both phenomena will be a definition in terms of information theory. The tentative definition I am playing with is “money is known-unknown information about which there are shared expectations.” I haven’t yet worked out the details, but I’ll let that statement stand baldly there while I do :)

    Thanks for the papers, they are going into my gradually accumulating bookmarks folder on the subject.

  4. Quick comments from my iPhone: “externalized confidence” is a nice phrase; hope you’re able to get an English translation of the paper in question. Your information theoretic definition is interesting, hope you’ll post more on this in future.

  5. Venkata Pingali says:

    There are several dimensions to this question.

    If you go with the broad consensus in economics that price carries information, money becomes a way to represent this information. At different points in time, different forms worked because this representation has some prerequisites such as durability and trust. The information view also means that since there are infinite variety of things that can be price, there is no one notion of money that will work. We routinely trade (concrete) favors. The favors are currency too.

    One issue with money is the control of the representation (bills, etc.). Whoever gets to define it/own it, gets political power. In all cases that power is abused. It goes on until the price of corruption of the system is much lower than the transaction value it enables. Countries/economies collapse when that is not true.

    I like the idea of “emergent” currency. But the state/powerful elite will not allow that to happen because they want the ability to corrupt (e.g., to wage a war)

  6. This is an interesting project. Two ground-clearing questions.

    1. Would a theory of money have to be historical? Would it aim to analyze what absolutely all money is, to account for currencies like wampum, cattle, &c.; or would it aim for a pure definition of Money that might well exclude many categories of trade goods and tokens which happen to have been its historic predecessors or alternatives?

    2. How would a theory of money relate to the theory of value? One the one hand there is work to build upon in the theory of value; on the other hand severing the theories of money and value has the appeal of an elegant scientific gambit, like the way information theory severs the concept of information from the engineering of the system that carries it. But such value-neutrality would contradict the idea that value is identical with price (unless you generalize price as something only sometimes or partly monetary). Information theory doesn’t care whether information is true or false. Likewise a general, value-neutral theory of money would have to be applicable to exchange under conditions of force or fraud.

  7. How about “pragmatic karma”?

    Small scale prisoner’s dilemma games can be played with memory. When you have much larger sets and you want to discriminate, you really have to use an external data structure viz. money.

    Quote from Dawkins’ “The Ancestor’s Tale”:
    The idea (I attribute this inspired way of expressing it to Steven Pinker) is that before the invention of the freezer the best larder for meat was a companion’s belly. How so? The meat itself was no longer available, of course, but the goodwill it buys is safe in long-term storage in a companion’s brain. Your companion will remember the favour and repay it when fortunes are reversed. Chimpanzees are known to share meat for favours. In historic times, this kind of IOU became tokenized as money.

  8. Venkata: the corruption issue is one I hope to address in a future piece. But check out the views of Thomas Greco on localization… (google him). He thinks local currencies are the best way to limit the power of central currency issuers.

    Paul: you are touching on exactly the right issue. I think price and value need to be separated, OR value needs to be relativized (i.e. not ‘good’ or ‘bad’ but ‘is mine the same as yours?’). I like the idea that pricing should be agnostic to value. Force or fraud SHOULD be included. There is presumably a set of pricing dynamics even in protection rackets, where street apple sellers get taken for more than store merchants, and each level of defaulting has its own punishment (“if you miss a protection payment this week, you get a slap in the face. Miss it for 2 weeks, you get beaten up at night. 3 weeks, house burned down. 4 weeks, you get killed”).

    Tubelite: the animal world examples are great. I hadn’t thought of the friend’s-stomach=refrigerator analogy at all. But it makes sense. Iterated prisoners’ dilemmas get to evolutionary stable states (ESS) of tit-for-tat if discounting of future isn’t “too high.” Your Dawkins/Pinker example formalizes that notion. If discounting of future value of trust (in PD sense) is less than degradation rate of meat, it makes sense to put it in stomach-refrigerator. There may be a general model here that combines NPV-DCF analysis with capital asset depreciation dynamics.

  9. IMHO any sustainable and insightful philosophy of money will stand on the shoulders of a giant: the late 19th century economist Carl Menger, specifically his essay On the Origin of Money. I’d be curious to hear your take on his approach.


  1. […] most recent one is on the discovery of money, where he asks Do we need to elevate the notion of money from the level of cultural construct, […]