Walter Block posts a response to my earlier post. He still says that borrowing money from a depositor is like selling a "square circle": it is a "logical contradiction." It's not, as long as the contract right is defined correctly.
A and C enter the following contract. A is to give C $100 and agrees that C may do various things with it -- lend it out, invest it, whatever. The contract provides further that A may demand cash up to $100 (never aggregating more than $100) from C at any time. C must satisfy the demand in cash to A if C has cash on hand; if not, C must liquidate his assets up to A's demand and pay the proceeds to A. If, even after liquidating his assets, C does not have enough to satisfy A's demand, he must pay what he has, and A's ability to satisfy his claim is contingent on C obtaining additional assets.
Block says that C is "bankrupt" the moment that he accepts A's cash and turns around and loans out part or all of it to another party, B. This is not true as a matter of law or economics. C's loan to B is worth something and appears as an asset on C's balance sheet. It's perfectly possible that if A demands his $100 back on day 2, C will be able to sell the loan and use the proceeds to pay back A and even make a profit.
Still, if Block is right, I'd think libertarians would be troubled by the thought of a vibrant market, going back many, many years, in square circles, involving millions of sophisticated people on both side of the transaction, endorsed by thousands of common law judges who are responsible for all our other precious contract and property rights. It's not a result of fraud in the sense of deceit, Block says (even though that is what fraud means). It is a kind of "fraud" where both parties, with full knowledge of a transaction that is internally contradictory, nonetheless decide to go through with it. If I offer to sell you a square circle, would you buy? With delusional behavior on such a grand scale, fractional banking would be the least of our problems.