A third concern about game theory as the foundation of the analysis
of economic institutions and behavior is its narrow scope. Society is not
well-modeled as a single game, or one with an unchanging structure. An
approach to games that would be adequate to understanding society
would have to take account of the following characteristics. Games are
overlapping: people regularly participate in many distinct types of social
interaction ranging from firms, to markets, to families, to citizen-state
relationships, neighborhood associations, sports teams, and so on.
Credit markets are often linked to labor and land markets, for example,
and loan agreements that would be infeasible in a credit market taken
in isolation may be possible when the borrower is also the employee of
the lender, or the renter of his land, and in both cases subject to eviction
should default occur. The overlapping character of games is also important
because the structure of one game teaches the players lessons and
imparts direction to cultural evolution, affecting not only how they play
the game in subsequent periods but how they play the other games they
are involved in. Citizens endowed with well-defined individual liberties
and democratic rights in their relationship to their government may, for
example, seek to invoke these in the workplace. Games, in other words,
are constitutive of the players’ preferences. Furthermore, not only the
players evolve; the rules do as well. The games are thus recursive in the
sense that among the outcomes of some games are changes in the rules
of this or other games. In the pages that follow, I will introduce overlapping
and asymmetric games in the analysis of firms, credit markets, employment
relationships, and class structure. Constitutive and recursive
games will be used to analyze the coevolution of preferences and
institutions.

1) having power to enact, appoint, or establish

т.е.:
Другими словами, игры конститутивны, то есть они претворяют в жизнь предпочтения игроков.

???