So does that mean that gold is a "fiat" currency if we're using the "unbacked currency" definition? Bitcoin has more in common with gold than it does with U.S. dollars. Bitcoin is a scarce digital commodity. It's true that Bitcoin isn't "backed," but it seems to me that the concept of "backing" only makes sense when the thing you're backing isn't reliably scarce, e.g. printed pieces of paper. The purpose of backing is to keep the issuer honest by acting as a constraint on its ability to inflate the currency supply. That scenario isn't possible with Bitcoin.
The economics definition is that gold is "commodity money" because gold has intrinsic value (it directly enters the utility function). Money which does not directly enter the utility function is called "fiat money."
State backing, etc. is irrelevant to the classification.
Of course, in bitcoin-land, fiat means something different which is not in accord with the definition in economics.
No, the economics definition will not change. These are well-defined technical terms. Money is sometimes modeled as "fiat money" and other times as "commodity money". The terms are referring to how money operates in mathematical models, not to how money operates in the real world. If you randomly revise the language, it would make communication difficult.
When I google "define fiat currency" here are the first five results.
From wikipedia:
Fiat money is money that derives its value from government regulation or law.
From investopedia:
Currency that a government has declared to be legal tender, despite the fact that it has no intrinsic value and is not backed by reserves. Historically, most currencies were based on physical commodities such as gold or silver, but fiat money is based solely on faith.
From businessdictionary.com:
Common type of currency issued by official order, and whose value is based on the issuing authority's guarantee to pay the stated (face) amount on demand, and not on any intrinsic worth or extrinsic backing. All national currencies in circulation, issued and managed by the respective central banks, are fiat currencies.
From thefreedictionary.com:
Money that is not backed by anything other than a government trust. Fiat money has no intrinsic value; it only has value at all because all participants in an economy agree to trust the government issuing the currency. All modern money is fiat money. While deflation is possible for fiat money, it is much more susceptible to inflation.
From dictionary.com:
paper currency made legal tender by a fiat of the government, but not based on or convertible into coin.
Each of those definitions includes the idea of government, but they don't all include the idea of a lack of "intrinsic value." I suppose you
could define "fiat currency" as currency that has no "intrinsic value," i.e. currency that has no non-monetary usefulness, without reference to whether or not that currency has any kind of official government status. And maybe that's how many people (including some self-described "economists") DO use the term. But clearly there are a large number of people who use and understand the term "fiat" to refer to money that derives its value solely from government decree. Which definition is "right"? Both. Neither. Language belongs to its users. (But I'd note that based on the above admittedly-extremely-preliminary research, my definition does appear to be the more widely used one.) What's more interesting to me is the fact that this distinction has probably never been that important. My guess is that there are very few examples of a market adopting an intrinsically-worthless currency in the absence of government coercion.