Goffman, Erving. Strategic Interaction.
This volume thus deals with the calculative, gamelike aspects of mutual dealings - what will be called strategic
interaction. By examining strategic
interaction in its own terms, we can become clear about what it is; being
clear, we will be better equipped to set it in its special place when looking
at face-to-face interaction. By seeing
that communication is of limited analytical significance in strategic
interaction, we can prepare ourselves to find its limited place in the
naturalistic study of face-to-face conduct.
On both counts we will be further on the way to segregating those
concerns which draw illustrations from the realm of immediate conduct (and in
turn illuminate it) but draw their basic analytical concepts from other
sources. (x)
Expression
Games: An analysis of doubts at play (3)
In pursuit of their interests, parties of all kinds
must deal with and through individuals, both individuals who appear to help and
individuals who appear to hinder. In
these dealings, parties - or rather persons who manage them - must orient to
the capacities which these individuals are seen to have and to the conditions
which bear upon their exercise, such as innate human propensities,
culture-bound beliefs, social norms, the market value of labor, and so
forth. To orient to these capacities is
to come to conclusions, well founded or not,
concerning them; and to come to these conclusions is to have assumptions about
the fundamental nature of the sorts of persons dealt with. (3)
In this paper I want to explore one general human
capacity in terms of the conceptions we have of its physical and social limits:
the individual’s capacity to acquire, reveal, and conceal information. The perspective here is that of an
organizationally committed observer who needs information from another
person. I will draw upon the popular
literature on intelligence and espionage for illustration, for no party seems
more concerned than an intelligence organization about the capacity we will
consider, and more likely to bring assumptions to the surface for us. Special attention will be given to occasions
when the informing individual is in the immediate presence of the party
collecting the information. (4)
I.
Assumptions (4)
1. Individuals, like, other objects in this
world, affect the surrounding environment in a manner congruent with their own
actions and properties. Their mere
presence produces signs and marks.
Individuals, in brief, exude expressions. (4-5)
The generating of expression, and hence making
its information available, is not an official end of the action, but (at least
ostensibly) only a side effect. Here,
then, is expressed information. (5)
2. There
are means, then, some quite standard, through which the individual expresses
information. (5)
Typically the signs employed are of the
generative kind, able to support an enormous number of different statements;
typically the message is somewhat free of context and its subject matter is not
at all restricted to the sender. I shall
speak here of communicated information (or transmitted information or a
message), and the process through which this information is conveyed, communication
or transmission. (6)
It need only be added that everyday language
tends to be ambiguous regarding the various relationships that have been reviewed,
and here common usage might best be avoided.
For example, when the term truth is employed, it is often
difficult to know whether the reference is to any self-believed statement, any
correct statement, or any self-believed correct statement. Similarly, the phrase “fully reply” fails to
distinguish between repleteness and candor. (7)
3. The behavioral and technical
process through which information is communicated, like all other human
activities, will naturally
exude expressions, and indeed that is why communication will have to be
considered in this paper. The least the
communicating can express is that the sender has the capacity and apparently
the willingness to communicate.
We can say, then, that as a source of information
the individual exudes expressions and transmits communications, but that in the
latter case the party seeking information will still have to attend to
expressions, lest he will not know how to take what he is told. Thus we have two protagonists, observer and
subject. When words are involved, we can
speak of an interrogator and an informant.
4. All
organisms after their fashion make use of information collected from the
immediate environment so as to respond effectively to what is going on around
them and to what is likely to occur. In the case of lower organisms, no appreciable
intelligence is involved and one can only say that in consequence of natural
selection information is gathered and used “in effect.” With higher organisms, especially man, instinct
is not sufficient and self-conscious intentional efforts are made to acquire
information from local events, with the purpose in mind of using this knowledge
to deal with these events. One may speak
here of a party assessing its situation, the assessment involving both the
collection of information and its use in helping to arrive at decisions. (10)
II.
The Basic Moves. (11)
1. Inanimate objects can certainly be said to be
indifferent to whether or not they are under observation. Or presumably whatever state the object is
in, it will persist in this state whether observed or not - the exception that
is famous at the subatomic level being of no concern to us. It is also the case that there will be
occasions when animate subjects including the highest will be, for all
practical purposes, unoriented to and unconcerned
about being under observation - at least observation by particular observers
and with particular interests - whether this is due to actual ignorance that
such observation is occurring or dim appreciation associated with quite genuine
indifference. In any case, the observer
can feel that he does not have to correct his observations for the possible
masking that the subject might engage is because at face value, an ingenuous
uncalculating expression or candid communication; and in taking this view the
observer, at times, can be quite justified. (11)
A subject’s observable behavior that is oriented
to the assessment an observer might be making of it can be called an unwitting
move. Such activity is at once part
of the expression game and not part of the game and requires a paradoxical
title. (11)
2. The term naive move can be used to
refer to the assessment an observer makes of a subject when the observer
believes that the subject can be taken as he appears, that is, that he is
involved in an unwitting move. (11)
3. The term control move will be used to
refer to the intentional effort of an informant to produce expressions that he
thinks will improve his situation if they are gleaned by the observer. (12)
A control move, we should see, is made with
reference to what is already part of the game, namely, the second move; it is
made relative to a world that has already been generated by the game. (13)
A common covering move is found in the act of
open secrecy and privacy, whereby the subject keeps observers from perceiving
something but makes no effort to prevent their perceiving that they are being
kept in the dark (14)
If the subject has not decided on a course of
action yet, he can feign that he has, or he can feign that he
hasn’t when he has. Also, he can feint a
course of action when has not started his actual course or has started a
different one. Note that here feinting
refers to faked courses of action, as when the military employ a “cover target”
for purposes of misdirecting the countereffforts of
the enemy, whereas feigning refers to beliefs, attitudes, and
preferences misrepresented strategically. (17)
Further, an expression game may involve
contestants who have a shared purpose, or a partially shared one, for such
players can still be involved in the necessity of reading another’s expression
and ensuring that their own is correctly read.
In the usual case the contest is between observer and subject; in “games
of coordination,” we can think if the observer and subject teamed together in a
contest against nature or against the score achieved by other such pairs, as in
match-point duplicate bridge. (17)
4. Three basic moves have been considered so far:
the unwitting move, whereby the subject acts mindlessly relative to impression
management; the naive move, whereby the observer draws information from what he
takes to be an unwitting move; the covering move, through which the subject
attempts to influence the conclusions that the observer comes to. A fourth basic move may be considered. The observer, suspecting that what he might
have treated as an unwitting move is actually or possibly an obfuscation or
misrepresentation, suspecting what appears to be ingenuous fact could be shot
through and through with a gamesman’s manipulation and design, suspecting this,
he can attempt to crack, pierce, penetrate, and otherwise get behind the
apparent facts in order to uncover the real ones. The observer performs an uncovering move. (17-18)
5. Just as subjects can be aware that they must
mask their actions and words, so they can appreciate that the controls they
employ may be suspected, the covers they use penetrated, and that it may be
necessary to attempt to meet this attack by countering actions, namely counter-uncovering
moves. (19)
A concluding note about
counter-uncovering moves in particular and the other kinds of moves in general.
Expression games are subject to constant developments and change as new
cues are discovered, new instruments of observation perfected, and once-secret
techniques become familiar and thereby less effective. (27)
Two particular developments in expression games
are worth noting. First, as
sophistication increases concerning what is given away
through bodily expression, guardedness during face-to-face interaction may also
increase; certainly it will shift in focus.
Second, with the recent rapid advances in the technology of
surveillance, there has been a marked tension in the kind of social setting
that can be realistically suspected as insecure, that is, subject to
monitoring; and with this we can expect an increase in care regarding certain
expressions and an increase in willingness to be exposed regarding other
expressions. (28)
III.
The constraints on play (28)
He who would analyze expression games must
consider the ungamey conditions that gamesters must
face whenever they engage in an actual game of expression. For the moves open to the subject are
established by the restrictive conditions of play the observer faces, just as
the ones open to the observer are established by the constraints affecting the
subject. What one party must face as a
limit on play the other party can exploit as a basis of advantage. Further, somewhat similar kinds of
limitations face both subject and observer. (28)
1. The
first constraint and condition to consider is derived from the interplay of
three physical factors: what is to be hidden; what is to be used as cover; and
the means of perception available to those from whom something is to be concealed. (28)
2. A
general limitation on play is the state of the participants’ technical
knowledge and competence. (29)
3. A third set of constraints on the assessment
game derives from what we can view commonsensically
as the constitutive features of human players - their all-too-human nature,
especially as subjects...One such limitation is that of emotional
self-control. (31)
By acutely awakening the subject’s ordinarily
latent interests he, in effect, becomes split in two, with one of these
interest-serving persons forming a coalition with the enemy. There are standard methods of mobilizing
these interests…There is seduction in its various forms. The observer’s object here is to maneuver a
definition of the situation such that the subject is led to believe that the
observer is to be treated as something of a teammate, to who strategic
information (among other things) can be voluntarily entrusted. (37)
Seduction as a means by which the observer can
attempt to break the subject’s cover is to be distinguished, I feel, from coercive
exchange, that is, the kind of exchange which the subject participates in
under pressure and in spite of his finer feelings in the matter. (38)
4. I have
mentioned three types of limitation on the play of expression games: physical
factors, knowledge, and “human nature.”
A final constraint to consider is that of social norms. (43)
Take the subject first. Although varying from one social circle to another, and from one set of circumstances to another, there
is nonetheless a special morality about impression management. There are rules against communicating
self-disbelieved statements. (43-44)
A final comment about the norms
that constrain subjects and observers. These various rules of play can
be followed, and gamesmanship correspondingly constrained, for a variety of
reasons, good and bad: incorporation of
the norms, causing an offender to feel guilt when he deviates even secretly;
genuine concern for the good opinion of witnesses, causing feelings of shame
when this is threatened; fear of legal penalty; perceived long-range
expedience, and so forth. (46)
IV. The Strategic
Properties of Play (47)
Just
as certain properties of players have a special significance in expression
games, so do certain properties of play itself. (47)
1. Note should be taken, first, of the difference
between real moves and virtual or tacit ones. As G.H. Mead argued, when an individual
considers taking a course of action, he is likely to hold off until he has
imagined in his mind the consequence of his action for others involved, their
likely response to this consequence, and the bearing of this response on his
own designs. He then modifies his action
so that it now incorporates that which he calculates will usefully modify the
other’s generated response. In effect,
he adapts to the other’s response before it has been called forth, and adapts
in such a way that it never does have to be made. He has thus incorporated tacit moves into his
line of behavior. Now when the courses
of action, actual and imagined, consist of assessment and response to
assessment we find ourselves dealing with virtual or tacit moves in an
expression game. The observer imagines
the likely consequence of the subject discovering that he is being assessed,
and attempts to offset the likely control of impression before it has had a
chance to occur. Thus the observer’s
first and second moves, constituting moves of the naive and uncovering types,
are collapsed into one. (47)
It should be apparent that not only can a player’s
move be a response to a tacit move on the other’s
part, but may have to be if it is to be effective. Precautions not taken before the opponent has
had a chance to make his move may not be worth taking at all, and the player
may clearly appreciate that this is the case.
We must then expect some insuring, some diffusion of wariness, some adaptation
to moves which one knows the opponent may not even have considered making. (50)
2. An
important feature of expression games has to do with the significance for the
play of the player’s knowledge of the other’s knowledge of what is occurring in
the game. For the play itself has its own strategic implications. (50)
The informing implications of gaming with
expression - the strategic implications of the game itself - are nicely
illustrated in double agentry, accounting for
whatever longevity is found among those who are so employed. (56)
The
Degeneration of Expression (58)
The more the observer suspects the subject of
control, or the more he wants to guard against this possibility, the less
weight he will give to the face value of the subject’s behavior and the more he
will seek out expressions that seem immune to fabrication and
dissimulation. For the observer this can
mean a heightened dependence on a special and therefore small part of the
subject’s expressions. However, the more
the observer concentrates his interpretive effort, and the more he gambles on
it, the more it will pay the subject to discover what this discovery about
himself is, and control the control by extra-ordinary efforts of expression
engineering. If an observer can learn
about the significance of a cue, then the subject can too, and there seems no
evidence that, once learned about, cannot be doctored. Uncovering moves must eventually be countered
by counter-uncovering moves. (58)
Given the corruptibility of minor cues, major
cues, and meshing of cues, this follows: the more the observer relies on
seeking out foolproof cues, the more vulnerable he should appreciate he has
become to the exploitation of his efforts.
For, after all, the most reliance-inspiring conduct on the subject’s
part is exactly the conduct that it would be most advantageous for him to fake
if he wanted to hoodwink the observer.
The very fact that the observer finds himself looking to a particular
bit of evidence as an incorruptible check on what is or might be corrupted is
the very reason why he should be suspicious of this evidence; for the best
evidence for him is also the best evidence for the subject to tamper with. However many moves the observer thinks he is
ahead of the subject in an expression game, he ought to feel that it is just
this sense of being ahead that the subject will find of maximum use in finally
trapping the observer. The harder a spy
must work to obtain startling secret information, the more confidence his
masters may put in his findings; but the very fact that masters do behave in
this way provides the best reason why the enemy should be careful to leak false
information only to those who have worked hardest to get the true facts, or
insist that “turned” agents put up a show of having to work hard for the
information planted with them. (63)
In addition, we can expect that certain
expression games will have their own natural limits. It is true that an agent will have to go
through many turnings before his value is entirely used up. But once the information state is reached
wherein both sides know that the other side knows that he has been turned,
then, I think, the game is pretty well over - and, of course, for him, pretty
well used up. (66)
Expression, then, becomes transformed into
communication and retains this role until reasons develop - as they are likely
to - for mutual suspicion to arise, this time, however, in connnection
with the normative commitments of the parties to communication. And so the event again drops out of active
status in the expression game between the observer and his subject, but this
time through a different hole. (70)
The Observer-Subject Model (70)
But even these qualifications leave us with too
simple a picture. For purposes of
analysis it is convenient to restrict expression games to a contest over one
item of information. In real life,
however, when the observer is engaged in uncovering one fact regarding a
subject, this subject is likely to be engaged in uncovering rather unrelated
facts regarding the erstwhile observer.
Pairs of players involved in one expression game against each other are
likely to be involved in additional games too, but this time
are likely to be in reversed roles.
Each seeker is therefore doubly a concealer,
and each concealer is doubly a seeker. Two individuals can, then, play against each
other, even while playing past each other.
(p71-72)
These various qualifications notwithstanding, I
do not think that the division into two roles, observer and subject, is
entirely arbitrary or that for any particular aspect of interaction it doesn’t
matter which of the participants is accorded which of
the slots. (72)
In general, however, one can expect that triads,
like dyads, will have some instability.
Because of this, there is sometimes to be found the practice of “clearing
the channels” - that is, a special effort is made by one participant - say, Tom
- to make sure that his relation to neither Dick nor Mary can be undercut by a
divulgence from the other. To do this,
Tom tells each what he tells the other, and also tells each that he has told
the other. Of course, Tom would then
still find himself with entrusted secrets of Dick’s (which Mary would like to
know about), and entrusted secrets of Mary’s (which Dick would like to know
about). A complete clearing of the network
is found when each of the three participants clears his own channels. It might be added that the longer the life of
the triad, and the more matters of concern the members share, the more likely
is the system to be unstable, or stable for unpleasant reasons, and perhaps,
the more the members yearn for a time when matters were less sticky. (77)
VII
Conclusions (77)
It is certainly the case that nations at war (and
is likely the case that individual organizations at peace) have been
wonderfully served by effective spies; Sorge and
Cicero are only two examples. But just
as certainly, the overall value of intelligence organizations of the national
kind can be questioned. (77)
And note, just as we are like them in significant
ways, so they are like us. In the little
service contacts we have in stores and offices, occasions are always arising when
we must ask for advice and then determine how to read the advice by trying to
analyze the sincerity of the server’s manner.
When we come into contact with the person who employs us, a similar task
arises; he has reason to almost cover his actual assessment with an equable,
supporting air, and we have reason to try to read his for what he “really”
thinks. The same is true in the warmer circles of social life. Surely every adult who has had a friend or
spouse has had occasion to doubt expression of relationship and then to doubt
even while giving the other reasons to suspect that something is being
doubted. These, then, are the occasions
for our expression games, but a nation’s gamesmen play here too. He who manages the affairs of state has to
make fateful decisions on the basis of the appearances of good faith of those
with whom he negotiates; similarly, an intelligence officer is dependent on
being able to appraise correctly the show of loyalty displayed by his
agents. (81)
In every social situation we can find a sense in
which one participant will be an observer with something to gain from assessing
expressions, and another will be a subject with something to gain from
manipulating this process. A single
structure of contingencies can be found in this regard which renders agents a
little like us all and all of us a little like agents. (81)
Strategic
Interaction (85)
Whenever students of the human science have
considered the dealings individuals have with one another, the issue of
calculation has arisen: When a respectable motive is given for action, are we
to suspect another ulterior one? When an
individual supports a promise or threat with a convincing display of emotional
expression, are we to believe him? When
an individual seems carried away by feeling, is he intentionally acting this
way in order to create an effect? When
someone responds to us in a particular way, are we to see this as a spontaneous
reaction to the situation or a result of his having canvassed all other
possible responses before deciding this one was the most advantageous? And whether or not we have such concerns,
ought we to be worried about the individual believing that we have them? (85)
In recent years this traditional concern about
calculation has been taken up and refined by students of game theory. This paper attempts to isolate the analytical
framework implied in the game perspective, and show its relationship to other
perspectives in analyzing interpersonal dealings. (85)
I
Individuals typically make observations of their
situation in order to assess what is relevantly happening around them and what
is likely to occur. Once this is done,
they often go on to exercise another capacity of human intelligence, that of
making a choice from among a set of possible lines of response. Here some sort of maximization of gain will
often be involved, often under conditions of uncertainty or risk. This provides one sense in which an actor is
said to be “rational,” and also an ethically neutral perspective from which to
make judgments concerning the desirability and advisability of various courses
of action. Rational decision-making
is involved. (86)
The kind of entity this paper will be concerned
with can be called a party, to be defined as something with a unitary
interest to promote. By the term coalition
is meant a joining of two or more, ordinarily opposed, parties, and their
functioning, temporarily in regard to specific aims, to promote a single
interest. (86)
The interests of a party are promoted by action
taken on the party’s behalf by individuals who are authorized to act for it and
are capable of doing so. An individual
agent is called a player (or sometimes, an “actor”). The player exercises human intelligence,
assessing “his” party’s situation, selecting from the available courses of
action and committing his party to this selection. Note that an individual may simultaneously
play for a party and - in another of his relevant capacities - be part of the
party itself, as every executive with a stock option knows. Moreover, the individual often acts for a
party of which he himself is the acknowledged sole member; in poker (but not in
bridge) we usually expect that the individual will “play for himself.” The analytical differences between party and
player here become easy to neglect, as I shall sometimes. (86-87)
The game-theoretical approach is clear enough on
the need to divide the individual into party and player, but less clear about
distinctive ways in which the individual can function in the gaming
situation. He can serve as a pawn. I mean here that conditions can be such as to
place in jeopardy the social or bodily welfare of an individual, and how this
welfare can be the interest that is at stake in the game. Now it should be seen that although this kind
of welfare is obviously an attribute of the individual in jeopardy, a concern
to preserve this state may well be the interest of a friend, a family, or even
a tribe. It is so when an infant is
given or taken hostage, or when pursuit is dampened because the owners of a car
have been forced to ride along with the gunmen who are stealing it. (87)
An individual may also function as a token,
that is, a means of expressing and marking openly a position that has been
taken. The early history of Western
diplomacy recognized this possibility with a term, nuncio, to refer to
someone who had the capacity to substitute for the presence of his principal,
that is, his party, although not to negotiate for it. (A procurator, on the other hand,
could negotiate but not ceremonially represent, being a player solely.) (87)
Given the several ways in which the individual
can perform, we must expect to find these functions often overlayed
upon each other. An individual who
functions as a pawn need not, of course, participate in any other capacity,
being “innocently involved,”
as we say; but just as clearly, he who functions as a pawn can
also be the player, as the story of Judith tells us. Our modern term “ambassador,” to take another
example, combines the functions of the earlier roles of nuncio and procurator. Similarly, he who functions as both pawn and
player can also function as the party - the complication routinely found in
sexual seductions and duels. Note
finally that although an individual can function as a pawn or as a token, other
objects of concern and value can also, and, more typically than individuals,
do. The player function, on the other
hand, is largely restricted to individuals, the doomsday machine being one of
the exceptions. (88-89)
I shall use the word turn to refer to
Harry’s moment and opportunity for choice, and move to refer to the
action he takes consequent on deciding to play his turn now. (90)
Although a game against nature is a rudimentary
game indeed, some concepts important to game analysis can be introduced by
it. One of these is the move. A move, analytically speaking, is not a
thought or decision or expression, or anything else that goes on in the mind of
a player; it is a course of action which involves real physical consequences in
the external world and gives rise to objective and quite concrete alterations
in Harry’s life situation. (91)
A second matter that Harry will want to know
about is what has been called operational code, namely, the orientation
to gaming that will diffusely influence how the opponent plays. Here several factors must be considered. There is the opponent’s preference pattern or
utility function, namely, his ordering, weak or strong, of aims and goals. There are his normative contraints,
the self-imposed conditions on furthering his objectives he is likely to
observe for reasons of moral sentiment or conscience. And a residual category: style of play.
(95)
Next to consider is resolve, that is, the
opponent’s determination to proceed with the game at whatever price to himself. Resolve, of
course, is to be considered relative to a particular game and is not solely a
function of the opponent’s resoluteness, that is, his general tendency to preserve
or not in whatever game he is playing. (95)
Also, it will be useful for Harry to know about
the other’s information state. I
refer here to the knowledge the opponent may possess about the important
features of his own situation and of Harry’s.
(95)
A further matter for Harry to consider is the
opponent’s resources or capacities - the stuffs that he as a party
can draw upon in his adaptations to the situation. (Harry’s own resources will be important to
him too, of course.) (96)
Now a central resource of the other as a party is
his having someone to play for him. And
given that the other has a player, in this case the other himself,
it is important for Harry to know something about the attributes of this
player. (Thus, when a mental hospital
psychiatrist deals with a patient who is new to him, the patient’s dossier will
be a useful thing to examine beforehand, as is true also when diplomats,
friendly or hostile, deal with each other, except here all the
participants are likely to have had access to dossiers on the others.) (96)
Perhaps the most important attribute of players
is their gameworthiness. I include here: the intellectual proclivity
to assess all possible courses of action and their consequences, and to do this
from the point of view of all the contesting parties; the practice of setting
aside all personal feelings and all impulsive inclinations in assembling the
situation and in following a course of action; the ability to think and act
under pressure without becoming either flustered or transparent; the capacity
to refrain from indulging in current displays of wit and character at the
expense of long-term interests; and, of course, the ability and willingness to
dissemble about anything, even one’s own capacities as a gamesman. (96-97)
Another important attribute of players is their integrity,
that is, the strength of their propensity to remain loyal to a party once they
have agreed to play for it, and not to instigate courses of action of some
other party’s interests, notably their own.
One is concerned here with the “conflict of interests” problem, with
normative constraints, but constraints that pertain not to parties in
interaction but to intraparty elements. (Coalition formation, strictly speaking, is
not at question here; the issue is not that of two parties coming into
unanticipated temporary secret joining forces, but rather that of a player
coming into a corrupted relation to his erstwhile party. (97)
Finally, it should be stressed that if an issue
cannot arise as to the integrity of an individual, then he cannot really
function as a player at all, although he may form an important part of the mere
or instrumental resources of a party. (98)
By the term position I shall refer to the
place Harry has arrived at in his game against the other. Harry’s position is created by the past opportunities
he did and did not avail himself of, and consists in the framework of possible
moves (with their probability of success) that are now open to him. (100)
Given an extensive game, one can characterize
intermediate moves (and the positions they give rise to) in various ways, an
important instance being the distinction between viable and nonviolable ones, where the concern is whether or
not a particular move, however much the player sacrifices by making it, still
leaves him able to continue on with the game should his move prove
unsuccessful. Similarly, one can speak
of viable and nonviable positions. (100)
Now it is possible to review the defining
conditions for strategic interaction.
Two or more parties must find themselves in a well-structured situation
of mutual impingement where each party must make a move and where every
possible move carries fateful implications for all parties. In this situation, each player must influence
his own decision by his knowing that the other players are likely to try to
dope out his decision in advance, and may even appreciate that he knows this is
likely. Curses of action or moves will
then be made in the light of one’s thoughts about the other’s thoughts about
oneself. An exchange of moves made on the
basis of this kind of orientation to self and other can be called strategic
interaction. One part of strategic
interaction consists of concrete courses of action taken in the real world that
constrains the parties; the other part, which has no more intrinsic relation to
communication than the first, consists of a special kind of decision-making -
decisions made by directly orienting oneself to the other parties and giving weight
to their situation as they would seem to see it, including their giving weight
to one’s own. The special possibilities
that result from this mutually assessed mutual assessment, as these effect the fate of the parties, provide reason and
grounds for employing the special perspective of strategic interaction. (100-101)
The gamey situations which have been considered
so far are tight and pure. Once nature,
self-interest, and an intelligent opponent are assumed, nothing else need be;
strategic interaction follows. And, in
fact, some of the developments in game theory require no more than these
minimal assumptions, the object being to find a desirable strategy for Harry in
the face of opponents as intelligent and amoral as himself. (102)
One of these developments in the analysis of pure
games is important for us here. It has
to do with the role of assessability, especially as
this pertains to communication. The
problem here is what has come to be called “credibility.” (102)
We can think of this second player, the other or
opponent, as contributing in two ways to this assessment. First, he can give off expressions
which, when gleaned by Harry, allow the latter to make some sense out of what
is happening and to predict somewhat what will happen. (In this the opponent presumably is no better
than lesser animals and even inanimate objects, all of which can serve as a
source of information). Second, the
opponent can transmit communications, that is, convey linguistic avowals
(or substitutes thereof). These Harry
can (and is openly meant to) receive, and is meant to be informed thereby. Some special attention should be given to
communications in the analysis of strategic interaction, for many games involve
this kind of activity. (102)
Certain statements made to Harry by his opponent
can have crucial relevance for whatever is gamey in the situation between
them. Harry’s opponent can make an unconditional
avowal regarding his intention to follow a stated course of action,
affirming that regardless of what Harry does, he is going to do so and so. More important still, the opponent can make a
conditional avowal, claiming that he will pursue a given course of
action if Harry does (or does not) engage in another given course of
action. Two basic possibilities are
involved here. There is the promise,
where the outcome conditionally proposed is something Harry can be assumed to
desire, and there is the threat, where the conditionally proposed
outcome is something Harry would presumably like to avoid. (102-103)
IV. (113)
Although the issue of commitment is of central
importance in the analysis of game strategies, the empirical study of strategic
interaction must proceed beyond this point.
The idea of all-out, “zero-sum,” opposition, and of a pure and tight
game, does not cover all that is to be considered. And while the notion of a game of
coordination expands matters a little, too much is still left out. For in real-life situations it is usually the
case that gamelike interactions occur in a context of
constraining and enabling social norms.
For game theory as such, these norms can be usefully treated as a
regrettable limitation on the game-worthiness of players - a matter to be
temporarily set aside so that analysis can proceed. Or norms can be treated as conditions of the
game, or something whose maintenance is defined as an interest of the parties. For the sociologist, however, these normative
limitations on pure gaming - limitations which ideal games themselves help to
point out - may be the matter of chief interest. Let us proceed, then, to bring our little
scenarios closer to life in various ways.
(113-114)
First, there is the constraint to play. Once in these gamey situations, Harry cannot
decide to disdain the play or postpone it; his doing, nothing itself becomes in
effect, a choice and a course of action.
Second, there are constraints regarding courses of action. Harry is not faced with a vast choice of
moves, each a little different from the other, or with the possibility of
creating variations and modifications; he is faced with a finite and often
quite limited set of possibilities, each of which is clearly different from the
others. Harry’s situation, in other
words, is structured. Third, once
he decides on a given move and initiates it, he cannot change his mind; he
becomes committed to it. Fourth,
a tight connection exists between the game and the payoff. In the illustrations given earlier, life and
death are involved, however much Harry might wish he were playing for less
serious takes. More important, the
courses of action taken and the administration of losses and gains in
consequence of play are part of the same seamless situation, much as is in duels
of honor, where the success of the swordsman’s lunge and the administration of
an injury are part of a single whole. I
shall speak here of an intrinsic payoff.
(114)
The four factors here described - constraint to
play, structuring of choices, commitment to moves made, intrinsic payoff -
taken together are sometimes referred to as an enforcement system. (114-115)
Review the development along which we have taken
Harry. On one side there is a party and
an authorized player who commits the party to a hopefully best course of action
in a situation made up of the possible courses of action open to it and to the
opponent. On the other side there is
nature (pure or socially impregnated), or a social agency, either of which has
the job of enforcing play, structuring choices, committing the party to the
player’s move, and enforcing a particular payoff. Interaction, then, from Harry’s point
of view, refers to the following sequence: assessment, decision-making,
initiating a course of action, and payoff.
Where a social agency is involved as enforcer, moves can be made by
means of a communication, but communication, at least in the narrow sense of
that term, is not analytically relevant or necessary. (120)
We can illustrate where we have gone, and go on
from there, by looking at “equipment games” such as checkers, bridge, craps, and
the like. (120)
In this paper I have attempted to formulate a
definition of strategic interaction and clarify the special perspective this
concept implies. (136)
It should be noted that strategic interaction is,
of course, close to Meadian social psychology and to
what has come to be called “symbolic interaction” - since nowhere more than in
game analysis does one see the actor as putting himself in the place of the
other and seeing things, temporarily at least, from his point of view. Yet it is quite doubtful that there are
significant historical connections between the two types of analysis. In any case, strategic interaction appears to
advance the symbolic interactionist approach, by
insisting on full interdependence of outcomes, on mutual awareness of
this fact, and on the capacity to make use of this knowledge, provides a
natural means for excluding from consideration merely any kind of
interdependence. This is important, for if
all interdependence is included in the study of interaction, hardly anything
distinctive can remain. Second,
following the crucial work of Schelling, strategic
interaction addresses itself directly to the dynamics of interdependence
involving mutual awareness; it seeks out basic moves and inquires into natural
stopping points in the potentially infinite cycle of two players taking into
consideration their consideration of each other’s consideration, and so forth.
(137)
Now the main analytical
argument. The framework of strategic interaction is
quite formal; no limit is placed on its implication, including the type of
payoff involved, as long as the participants are locked in what they perceive
as mutual fatefulness and are obliged to take some one of the available, highly
structured courses of action. Because of
this inclusion of any kind of payoff, the game approach has an easy application
to almost everything that is considered under the ill defined rubric “interaction.” Furthermore, howsoever interaction is
defined, the actors involved must be accorded some attributes and given some
internal structure and design, and here the propensities of a gamesman will
have a place. The strategic approach
will therefore always apply in some way; it is important to be clear, then,
about the limits of this application. (136-137)
Look now at one further social element of the communication
system in question: what might be called the issue of “frame.” Given a well-received, easily understandable
message, what light is the message to be seen in, what systematic, word-by-word
rereading is to be given it? Is the
sender engaged in what he appears to be doing namely, sending a serious,
reliable message? Or is he merely
practicing his sending, or engaging in a joke, or sending a false message
because he is now working for the other side, or sending a message at the point
of a gun, the message being designed by him to make this evident to the
recipients? (141)
It is possible, then, to specify a communication
system and to consider the strategic implications of its various
conditions. Taking the same focus - a
particular communication system - it is similarly possible to consider the
bearing of social relationships and face-to-face interaction upon that
system. In this way, various substantive
areas can be drawn upon in a clear and subordinated way for what they can tell
us about a communication system. But in practice, when the term communication is used, little
clarity and consistency is found as to just what it is that is being
investigated. (142-143)
In this paper, of course, communication was not a
central subject matter. Assessments were
one central issue, it is true, but these assessments were as often a result of
expression gleaned as they were a result of communications conveyed. (143)
At the beginning of the paper it was suggested
that the two main moves open to Harry the hunted - bolting for the tree or “freezing”
- raise the issue of visibility, and that, on this ground alone (although there
are others), the two moves were radically different. Although this is the case, it is also the
case that in the game between Harry and the lion a move is something in
addition to a resource for assessment - it is an objective
circumstances-altering action whose efficacy just happens to be influenced by
the issue of visibility. Now the
complication resides in the fact that if we are willing to forego considering
Harry’s full plight and are willing to restrict ourselves to issues about
visibility we can, in fact, construct a little game out of these contingencies,
a game wherein the whole value and character of a move has to do with
assessment and its management. We can,
in fact, abstract or excise from any occasion of strategic interaction an
expression game. And this I have tried
to do elsewhere. But while this
narrowing of focus is possible, we must here see that each of these expression
games can properly be considered also as one component, and a variable one, of
something more inclusive, a game concerning objective courses of action, an
occasion of strategic interaction. (144-145)
Let me repeat:
In the analysis of strategic interaction, moves are central, but these
constitute a class that is broader than the one derived from moves in
expression games. During occasions of
strategic interaction, a move consists of a structured course of action
available to a player which, when taken, objectively alters the situation of
the participants. Some of these moves
are concealed, some visible; when visible, the question will always arise as to
the reading that the opponent places on the event, namely the assessment he
makes in terms of it. But this reading
will be merely a contingency of the interaction, certainly not the whole
thing. What is effected
by strategic moves is not merely a state of information, but rather courses of
action taken. Thus we can expect to find
situations where Harry elects a course of action knowing that he thereby
provides the other side with information they can use against him, but in spite
of this cost finds that the other gains outweigh the price of information. (145)