[A mode of relating] in which two parties act, or are disposed to act, toward one another in equivalent ways. Graeber, Toward an Anthropological Theory of Value, p.225

[…] we are not really dealing with reciprocity here-or at best, only with reciprocity in the broadest sense. What is equal on both sides is the knowledge that the other person would do the same for you, not that they necessarily will. The Iroquois example brings home clearly what makes this possible: that such relations are based on a presumption of eternity. Society will always exist. Therefore, there will always be a north and a south side of the village. This is why no accounts need be taken. In a similar way, people tend to treat their mothers and best friends as if they will always exist, however well they know it isn't true. Graeber, Debt, p.100

This element of competition can work in completely different ways. In cases of barter or commercial exchange, when both parties to the transaction are only interested in the value of goods being trans­ acted, they may well-as economists insist they should - try to seek the maximum material advantage. On the other hand, as anthropologists have long pointed out, when the exchange is of gifts, that is, the objects passing back and forth are mainly considered interesting in how they reflect on and rearrange relations between the people carrying out the transaction, then insofar as competition enters in, it is likely to work precisely the other way around-to become a matter of contests of generosity, of people showing off who can give more away. Graeber, Debt, p.103