This book is really interesting. Unfortunately, it is also quite a mess. There’s a thesis buried in it somewhere, but Graeber never makes it clear. Here’s a stab at it:
-
The origins of money can be traced to violence: war and slavery. In both cases the “context” of a thing is removed, which exactly is the thing that makes it fungible. There are some societies where women where seemingly treated like currency (what Graeber calls human economies), but the important difference between these and a “real” economy that trades in money is that it is only in extraordinary circumstances (e.g. in conflicts with unknown, faraway tribes) that items are seen to be commensurate with human “property” and thus exchangeable for it. Human economies usually only traded women for women.
-
Historically, societies oscillated between virtual money (credit) and actual money depending on circumstances. During the Axial Age, there was a shift towards actual money because of the prevalence of war and the need of states to pay soldiers. Coin money is useful as payment because it is “contextless”.
-
The standard economic picture of development from barter to money to credit is exactly backwards; most human societies have only credit, and then in certain circumstances (outlined above) transitioned to use money. Barter, where people performed “spot trades” with no expectation of trading again in the future, is a recent development that only makes sense from early modernity (?) on. What people usually call “barter” they really mean credit---you give something to your neighbor today, with the expectation that he will be able to return the favor some time in the future. What he will give to you will be roughly commensurate, but never exact as in monetary compensation. Thus everyone is in a little bit of debt to each other all the time---exactly opposite of the economics picture.