what no one seems to be realizing in this forum is that fiat currency gets its "initial value" in the same way as any currency, and that regulation and taxation are neither necessary nor sufficient to explain its value. (Of course they aren't, if you put any stake in the idea that bitcoin has value!) There has been a lot of research on this subject for decades, and there are many different theories on what gives fiat it's value, including the state theory to the commodity theory. The commodity theory says that currency such as fiat which has no intrinsic value (this is actually the central concept behind the definition of a fiat currency), is valuable only as an intermediary between other goods, i.e. because it is expected to be tradeable for another asset in the future. Consider an economy of three people: A, B, and C. A has something B wants, B has something C wants, and C has something A wants, so that if they were to trade in a circle, the total utility of the economy would increase. However, assuming that all transactions in the economy must be pairwise (an exchange between two parties), there is no way to actually make the optimal trades in this economy, since there is no coincidence of wants. Now suppose we introduce a fiat currency into the market by giving it to one of the people. This extra degree of freedom allows for the 3 parties to complete their trades in pairwise exchanges, and increase the total utility of the economy. This increase in total utility is exactly the value of currency.

That is really the simplest answer to the OP's question. Of course, that oversimplified explanation has its problems. Most importantly, there is the built in assumption that the other people in the economy will accept the currency. It isn't rational to assume this, so it seems that the commodity theory of currency value, while sufficient to explain the VALUE of currency, is not sufficient to explain its existence, which I think is an important distinction to make. This is where the state theory comes into play.. the regulation and enforcement of this currency provides the requisite trust system that allows for the currency to take hold. However, this idea is also considered a vast oversimplification. There have been other ideas proposed that would allow us to get around this trust problem without the use of regulation and law enforcement.  You see, the main problem in the example with only three people is that after the first two trades have taken place, the third person will not be able to trade the worthless currency with anyone, everyone else has already received what they want, and no longer have interest in the new currency. This disadvantage for being the last guy "holding the bag" is precisely what destroys the viability of the currency, since there is no way of knowing if you will be last. A theory proposed by (Kovenock and De Vries, 2002) gets around this problem by making the number of players in the economy sufficiently large, though still finite.  This theory says that if the economy is large enough, there is a very low probability of being the last person holding the currency, and since you know that all of the other parties know that as well, your expected utility from using the currency outweighs the uncertainty about being the last person holding it. It has been proposed that this type of model is exactly what gives bitcoin currency value despite its lack of regulation or instrinsic value. Or said differently, the thing that gives a currency its "initial value" is simply the fundamental utility argument above, provided a sufficiently large core number of people trading said currency.

To sum up, I think it is kind of ridiculous that in a forum about bitcoins, where people assume that a currency can have aggregate value even if it is instrinsically worthless and has no regulation, almost no one has stopped to think that largely the same forces are at work in fiat currency. Yes, of course government can use force to make you pay your taxes, and that this kind of aggression can alter the value of the currency. But that is not what the OP asked. The OP didn't ask what forces us to use one currency over another (regulation/taxation), or what causes the value of currency to be stable (regulation), or what causes currencies to be homogeneous within a certain nation-state (taxation). As some posters have pointed out, there are many examples, such as the USD, of currencies that started out without strong legal enforcement, and there are countries without taxes, like somalia, which still use currencies. Does it really seem plausible that fiat currency and digital currency like bitcoin really derive their value in completely, fundamentally different ways? Of course not. Yes, bitcoin is radically different, but it is still a currency and the way it derives its value is fundamentally the same as fiat. All of the posters here who claim that the only reason fiat has value is because someone tells you at gunpoint that it does are really just fueling an "us and them" mentality while ignoring the hypocrisy of turning around and saying that bitcoin is valuable without regulation or taxation.