Literature Review 3

- 3

C onstructivism

: A U

ser's M anual

N icltolas O

ituf

C onstructivism

is a w ay of studying social relations-any

kin~ of so~a~ relations. W

hile it draw s from

a variety of other w ays of studying sue

d broad and com

plex subject, it stands on its ow n as a system

of conctts an

propositions. C onstructivism

is not a theory as _su~h. It doe\not \erg:~~ • 1ex lanations for w

hat people do, w hy soc1et1es differ,

ow t e w

o ~;,aan e~. Instead, constructivism

m akes it feasible to theoriz~ _about matters

that ;eem to be unrelated because the concepts and propos1ttons norm

ally used to talk about such m

atters are also unrelated. .

. .

A s presented here, constructivism

applies to all fields of ~oc1al inqu\~~ In recent years dissident scholars in m

any fields have selecttvely used d

language of so~ial construction to criticize existing social arrangem en;s an!

scholarly practices. A great deal of discord has ens_ued. (~!so se\ tt

o'. Introduction.) W

hen constructivism is used system

atically,_ ti has t~re ~~ne site effect. It finds value in diverse m

aterials and forges hnks w h

seem ed possible.

, ti Id

h' I Full of discordant voices, International R

_e!ations 1s the ie I'

this particular system of concepts and propostt1ons_

w as ti~st~pP;\. b'ts of

this m anual is intended for the use of anyone w

ith m et 10. tea

t m

ind its users are m ost likely to have an interest in the subiect of,inte~~:

tio11al relations. T hey m

ay have also had som e exposure to tl:e field s sc to

arly controversies. If this is indeed the case, they w ill soon d1scobver

thlatdthte .

. .

I x than they have

een e 0

subject is less d1st1nct1ve, but m ore com

p e ' believe.

C O

N ST

R U

C T

IV ISM

: A U

SE R

'S M A

N U

A L

59

- O

verview

Fundam ental to constructivism

is the proposition that hum an beings are social

beings, and w e w

ould not be hum an but for our social relations. In other w

ords, sociaf relations m

ake or co1is1111ct people-ow

:re/ves-into the kind of beings

that w e are. C

onversely, w e m

ake the w orld w

hat it is, from the raw

m aterials

that nature provides, by doing w hat w

e do w ith each other and saying w

hat w e

say to each other. Indeed, saying is doing: talking is undoubtedly the m ost

im portant way that w

e go about m aking the w

orld w hat it is.

C ountries such as France, the U

nited States, and Z im

babw e are am

ong the social constructions, or societies, that people m

ake through w hat w

e do. C

ountries arc self-contained w orlds because people talk about them

that w

ay and try to keep them that w

ay. Y et they are only relatively self-contained.

R elations am

ong countries--intem ational relations-constitutes

a w orld in

its ow n right. T

his is a self-contained w orld for the sim

ple reason that it covers the earth, but it is still nothing m

ore than a w orld of our m

aking-a society of relatively self-contained societies.

C onstructivism

holds that people m ake society, and society m

akes peo- ple. T

his is a continuous, tw o-w

ay process. In order to study it, w e m

ust start in the m

iddle, so to speak, because people and society, alw ays having

m ade each other, are already there and just about to change. T

o m ake a

virtue of necessity, w e w

ill start in the m iddle, betw

een people and society, by introducing a third elem

ent, rules, that alw ays links the other tw

o ele- m

ents together. Social rules (the term rules includes, but is not restricted to,

legal rules) m ake the process by w

hich people and society constitute each other continuous and reciprocal.

A rule is a statem

ent that tells people w hat w

e should do. T he "w

hat" in question is a standard for people's conduct in situations that w

e can identify as being alike, and can expect to encounter. T

he "should" tells us to m atch

our conduct to that standard. If w e fail to do w

hat the rule tells us to, then w

e can expect consequences that sonie other rule w ill bring into effect w

hen other people follow

the rule calling for such consequences. A ll the w

ays in w

hich people deal w ith rules--w

hether w e follow

the rules or break them ,

w hether w

e m ake the rules, change them

, or get rid of them -m

ay be called

practices.E ven w

hen w e do not know

w hat a rule says, w

e can often guess w

hat it is about by looking at people's practices. A

m ong m

uch else, rules tell us w ho the 'active participants in a society

are. C onstructivists call these participants agents. People are agents, but

only to the extent that society, through its rules, m akes it possible for us to

participate in the m any situations for w

hich there are rules. N o one is an

agent for all such situations.

•60 NIC IIO

L A

S O N

U F

O rdinarily, w

e think of agents as people w ho act on behalf of oth~r

people. C onsidering the m

atter m ore abstractly, w

e see that rules m ake tt

possible for us to act on behalf of social constructions, w hich m

_ay be ourselves, other hum

an beings, or even collections of people, along w ith the

rules, the practices, and the actual things that w e m

ake and use. C onversely,

agents need not be individual hum an beings to be able to_ act on behalf of

others (here l refer to agents in the third person to em phasize that_ the terrm

people and agents arc not com

pletely interchangeable). Agency ts a social condition. T

hus the governm ent of a country is a collection of people and a

social construction. A ccording to the relevant rules, these people act, to-

gether and in various com binations, on behalf of that country as a m

uch larger collection of people.

. R

ules give agents choices. A s w

e have already seen, the m ost baste

choice is to follow the rule--to

do w hat. the rule says the agent should

do-or not. O

nly hum an beings can actually m

ake choices, because w e

alone (and not all of us) have the m ental equipm

ent to consider the probable consequences of m

aking the choices that are available to us. N evertheless,

w e alw

ays m ake such choices on behalf of, and in the nam

e of, social constructions, w

hether ourselves, other people or collections of other peo- ple, or practices and artifacts.

, A

gents act in society to achieve goals. T hese goals reflect p7 ople s needs

and w ishes in light of their m

aterial circum stances. Every society has rules

telling agents w hich goals are the appropriate ones for t~em

to pursue. O f

course there are situations in w hich people are perfectly aim

less. For exam -

ple, w l1cn w

e freeze up in fear or fall asleep from exhaustion, w

e are no longer agents or, for that m

atter, social beings. .

. .

W hen w

e, as hum an beings, act as agents, w

e have goals m m

tnd, eve? tf w

e are not fully aw are of them

w hen w

e act. If som eone asks us to _thtnk

about the m atter, w

e can usually form ulate these goals m

ore or less tn ~he order of their im

portance to w hom

ever w e are acting as agents for, s~artm

g w

ith ourselves. M ost of the tim

e, agents have lim ited, inaccurate, or incon-

sistent inform ation about the m

aterial and social conditions that affect the likelihood of reaching given goals. N

evertheless, agents do the b_est they can 10 achieve their goals w

ith the m eans that nature and society (to-

gether-alw ays

together) m ake available to them

. A :~_chi~~e_g~als_is

rational conduct and agents faced w ith choices w

ill act rationally. V iew

ed fro;1 ~~utsiJc, th~se choices m

ay appear to be less than rational, but this is due to the com

plexities of agency and hum an fallibility.

A gents m

ake choices in a variety of situations. R ules help ~o d?fine

every such situation from any agent's po_int of view

. In m an~ s1tuat1ons,

rules arc directly responsible for presenting agents w ith choices. A

gents

C O

N ST

R U

C T

IV ISM

: A U

SE R

'S M A

N U

A L

61

- have m

ade or acknow ledged these rules in the belief that follow

ing rules generally helps them

reach their intended goals. In these situations, rules are related to agents' practices, a11d to each

other, through the consequences that agents intend their acts to have. W

hether by accident or by design, rules and related practices frequently f~111!a__S(~bJe (but never fixed) pattern suiting agents~ intenttons-:-'ffiese

l / ~att~i:t)Sa.rei'.1stit111ions.

A s recogniiab!e-paffem

s of ~Jes ~nd r~lated prac- tices, institutions m

ake people into agents and constitute an environm ent

":'ithin w hich agents conduct them

selves rationally. W hile it is alw

ays pos- sible, and often useful, to think of agents-all

agents-as institutions in

~hei: o~n right, w e m

ore com m

only think of agents as operating in an m

stttutlonal context that gives them at least som

e opportunities for choice. ~xercising choices, agents act on, and not just in, the context w

ithin w

hich they operate, collectively changing its institutional features and

the~selves, !n the process. N evertheless, from

any agent's point of ~iew ,

socte~ consists of diverse institutions that seem , for the m

ost part, to be hel~ m

_ pl~ce b~ rules _linking them to other institutions. A

ny stable pattern of m

st1tut1ons (t~clu~m ~ agents of all sorts) is also an institution. A

gents are aw

are of the m st1tu_t1ons

populating their environm ents, and not sim

ply because the rules form

tng these institutions directly bear on their conduct. T

o the extent that som e agents m

ake choices, and other agents are affected by these choices, institutions produce consequences for other agents that they cannot help but be aw

are of and respond to. In a com

plex w orld, agents often m

ake choices that have consequences, for them

selves and others, that they had not anticipated or do not care very m

uch about. U nintended consequences frequently form

stable patterns w ith

respect to their effect on agents. A perfect m

arket provides a com pelling

illustration of this phenom enon. O

ne by one, a large num ber of sellers and

buyers _are inca~able. of affect_ing the supply of, and dem and for, a good.

~ollect1:ely, their rational choices have the unintended consequence of set- ting a pnce for that ~ood w

hich they m ust individually accept as fixed.

A nyone m

ay notice such stable patterns of unintended consequences. In th~ case of a ~arket, no one could fail to notice it in the form

of a good's p~ce, over w

hich no agent seem s to have any control. Som

etim es agents

:v1t1 choose to prev_ent changes in such patterns by adopting rules that are m

ten~~d to_ have this effect. A rule fixing the, price of a good under certain

conditions ts only the m ost obvious exam

ple . . A

ny s~able pattern of rules, institutions, and unintended consequences , _,,

gives society ~.E !:!!E

!_1re,recognizable as such 'ioaiiy obsi,rver.-A ge.nts ·are-

· alw

ays observers. I_nsofar as they observe consequences that they had not intended, and accept them

, such consequences are no longer unintended in

63 62

N IC

I IO LA

S

- O

N U

F

the usual sense of the w or<l. If agents <lccide that these consequences are

hat! for them , they w

ill act to change them , perhaps w

ith other unforeseen consequences rcsu lting.

O utside observers (agents from

a different society) m ay recognize a

m ore com

plex structure than agents do as observers. O utsiders can stand

back, so to speak, and see patterns that insiders cannot see because they are too close to them

. A s agents on the inside becom

e aw are of w

hat observers have to say, observers becom

e agents, w hatever their intentions. W

hen agents in general take this new

infonnation into account in m aking their

choices, an even greater com plexity of structure results.

Scholars w ho think of them

selves as constructivists have given a good deal of attention to the "agent-structure problem

." (Sec H arry G

ould's con· tribution to this volum

e in C hapter 4 for a thorough review

of these discus· sions.) T

he tenn structure is the source of m uch confusion (an ontological

confusion), because scholars cannot agree on w hether structures exist in

reality or only in their m inds. T

he im portant point to rem

em ber is that

structure is w hat observers see, w

hile institutions are w hat agents act

w ithin. N

evertheless, structure can affect agents. W e are often affected by

phenom ena, natural and social, that w

e do not or cannot see, but w e then

respond as agents by putting w hat has happened to us in an institutional

context. W hen agents do this, they institutionalize

structure by bringing rules to bear on their situations.

G enerally speaking, scholars today tend to think that the structure of

international relations is not institutionalized to any great degree. T his is so

even for som e scholars w

ho think of them selves as constructivists. T

hey believe that countries are highly institutionalized as states, but that states, through their agents, conduct their relations in an anarchic w

orld. T he term

anarchy points to a condition of rule am

ong states in w hich no one state or

group of states rules over the rest. It also :m p lies that there is no institution

above states ruling them . W

hen w e say that states are sovereign, w

e are saying the very sam

e thing. B

y calling international relations anarchic, scholars are not saying that there is an absence of rule. T

his w ould be chaos, not anarchy. Instead, they

seem to be saying that structure-and

especially a stable pattern of unin· tended consequences-rules

the day. In the sam e sense, w

e m ight say that

the m arket rules the behavior of sellers and buyers.

Starting w ith rules, as constructivists often do, leads quickly enough to

patterns of relations that w e can only describe as a condition of rule. U

su- ally this condition is sufficiently institutionalized that w

e can recognize specific agents as rulers. Som

etim es there is very little evidence of institu·

tion.alization, as in m ob rule, but there is also little reason to think that this

C O

N S

TR U

C TIV

IS M

: A U

S E

R '-N

U A

L

condition w ill persist as a stable pattern w

ithout institutions em erging. In

other w ords, w

here there are rules (and thus institutions), there is rulc--n condition in w

hich som e agents use rules to exercise control and obtain

advantages over other agents. R ule is a stable pattern of relations, but not a

sym m

etrical one. A

narchy is a condition of rule in w hich rules are not directly responsible

for the w ay agents conduct their relations. T

o be sure, there are rules in the background. T

hey m ake sure that the unintended consequences of agents'

m any choices, and not rulers, do the job of ruling. If unintended conse-

quences seem to rule, it is because som

e agents intend for them to do so.

Som e agents w

ant to be ruled in this indirect sort of w ay because it suits

their goals m ore than any other arrangem

ent w ould. O

ther agents have little or no choice in the m

atter. Perhaps patterns just happen, but agents m ake

arrangem ents. A

rranging for anarchy is just one possibility. C

onstructivists should seriously consider dropping the w ord structure

from their vocabularies. Social arrangem

ent is a better choice. A ppearances

aside, international anarchy is a social arrangem ent-an

institution-on a

grand scale. W ithin its scope, m

any other institutions are recognizably con· nected. In every society, rules create conditions of rule. T

he society that states constitute through their relations is no exception.

W hether w

e, as constructivists, start w ith agents or w

ith social arrange· m

ents, w e com

e quickly enough to particular institutions and thus to rules. If w

e start w ith rules, w

e can m ove in either direction-tow

ard agents and

the. choices that rules give them an opportunity to m

ake, or tow ard the

social arrangem ents that em

erge from the choices that agents are m

aking all the tim

e. W hichever w

ay w e go, w

e ought to keep in m ind that rules yield

rule as a condition that agents (as institutions) can never escape. T

he practical problem is that, as constructivists w

e w ant to m

ove in both directions at the sam

e tim e. Y

et ifw e try to dos~, w

e com e up against the

staggering com plexity of the social reality that w

e w ant to know

about. It is im

possible to do everything. T he practical solution is to start w

ith rules and show

how rules m

ake agents and institutions w hat they are in relation to

ea~h other, T hen w

e can show how

rules m ake rule, and being ruled, a

um versal social experience.

T he rem

ainder of this user's m anual is dedicated to these tw

o tasks. T o

m ake points as clear and understandable as possible, it repeats m

ost of w hat

t~e reader has now had a taste of. In the 'process, it introduces m

any addi- ttonal concepts and propositions, expressed in the sim

plest term s that its

author can think of. U sed consistently and system

atically related, these concepts and propositions constitute a com

prehensive fram ew

ork for under- standing the w

orld in constructivist tenns.

65 -64 NIC

H O

L A

S O N

U F

R ules M

ake A gents, A

gents M ake R

ules

R ules m

ake agents out of individual hum an beings by giving them

opportu- nities to act upon the w

orld. T hese acts have m

aterial and social conse- quences, som

e of them intended and som

e not. T hrough these acts, agents

m ake the m

aterial w orld a social reality for them

selves as hum an beings.

B ecause agents are hum

an beings, acting singly or together on behalf of them

selves or others, they act as they do for hum an purposes-they

have goals reflecting hum

an needs and w ishes. The tangled connections betw

een agency (w

ho is acting on w hose behalf?), goals (w

hose goals are affected by w

hat acts?), and circum stances (w

hich features of the w orld actu'ally

m atter?) m

ake it difficult for agents to explain fully and convincingly w hy

they act as they do. E ven if they seem

confused, observers can often figure the reasons for their conduct from

the evidence at hand. A

gents use w hatever m

eans are available to them to achieve their goals.

T hese m

eans include m aterial features of the w

orld. B ecause the w

orld is a social place, at least for hum

an beings, rules m ake the w

orld's m aterial

features into reso11rcesavailable for agents' use. Som e resources are not

directly m aterial---rules also constitute agents and institutions as resources.

W hether agents are able to spell out their reasons for using the resources

available to them , or observers figure them

out from the evidence, recogniz-

able patterns in the results constitute agents' interests. A

gents need not know w

hat their interests are to act on them . O

nce they learn m

ore from other agents (as observers) about their ow

n interests, they m

ay act differently. Indeed, hum an beings do not need to think about them

- selves as agents to be agents. W

hile being an agent does not require the degree of self-consciousness that w

e associate w ith having an identity,

agents are usually aw are enough of their identities, singular and collective,

to have an interest in fostering those identities. A

s agents, people can m ake other people into agents by giving the latter

the opportunity to act on the farm er's behalf for particular purposes. T

he form

er m ay do so individually or collectively, and the latter m

ay be one or m

ore individuals acting on the form er's behalf. A

gents acting collectively becom

e a singular agent. B y using resources, they acquire a m

aterial exis- tence, and, as the previous paragraph suggests, they becom

e objects of identification.

A gency is alw

ays lim ited. A

gents are never free to act upon the w orld in

all the w ays that they m

ight w ish to. M

any lim its have a m

aterial com po-

nent. W e need air to breathe; w

e do not have w ings to fly. N

o rule can readily m

ake things otherw ise, even though rules allow

us, agents, to use resources to alter these lim

its, for exam ple, by fashioning scuba gear and

C O

N ST

R U

C T

IV ISM

: A U SE

R "'N

U A

L

airplanes. R ules that give any agent an opportunity to act create lim

its for other agents. R

ules in general lim it the range of acts that other agents arc

free to take. It follow

s from this proposition that no individual hum

an being, as an agent, has full autonom

y. B y the sam

e token, agents acting together never h?ve full independence. A

s noted, agents are alw ays lim

ited by rules thnt give other agents opportunities to act. A

gents acting together arc addition- ally lim

ited by the very rules that give them the opportunity to act collec-

tively. R ules allow

ing other agents, individual and collective, to act on their behalflim

it them even further.

W hen a very large num

ber of people collectively operate as an agent, w

hen they have agents acting for them , w

hen they have som e considerable

m easure of identity (including som

e place identified as theirs), and w hen

they are free to act w ithin very w

ide lim its, these people constitute a coun-

try. For several centuries, agents have had a consistent interest in talking about countries as if they are independent of each other and any other social ~onstruction. T

his is m ade clearest by defining sovereignty as absolute

m dependence and describing countries as sovereign states. A

s constructiv- ists, how

e_ver, we should alw ays bear in m

ind that full independence is a useful fictton, and sovereignty is a m

atter of degree. T

h~ freedom that agents do have depends on their ability to recognize the

m atenal and social lim

its that apply to them . T

hey m ust also be able to

evaluate the consequences of exceeding those lim its. T

o be an agent re- quires the m

ental equipm ent that individual hum

an beings norm ally develop

over ~he course of their social lives. A gents exercise their freedom

by choosing to act one w

ay or another, in an unending series of situations that m

a~e choosing unavoidable. It hardly needs saying that not choosing is a choice, presum

ably taken, as all choices are, to advance agents' goals. A

gents m ake choices in light of the skills that they possess and the re-

sourc~s that they have access to, for reasons that they are m ore or less able

to articulate. In short, they m ake choices in pursuit of their interests.

R ules o~er agents the sim

plest kind of choices. A gents m

ay choose to follow

a given rule, or to break it. C om

pared to m ost situations in w

hich ?gents m

ake choices, the choice of follow ing a rule or not follow

ing it m

volves consequences that are easy to calculate. W hile unintended conse-

qu~nccs are. alw ays ~ossible, rules give agents the opportunity to m

ake rational choices-<:hoices dictated by reference to goals-w

ith som

e assur- ance that they are m

aking the best choices available to them .

A ~le

m akes rational choice relatively easy by telling the agents to

w hom

it refers w hat they should do in som

e sort of situation that they m ight

find them selves in. T

hese agents m ay act on the contents of the rule w

ithout

67 6/,

N IC

I IO L

I\S -

O N

U F

realizing that the contents form a rule. In principle, how

ever, any agent (including any observer w

ith enough inform ation) can form

ulate contents of a rule in the form

of a rule. T here is nothing tricky about this. Saying w

hat a rule is--putting its contents in the right form

-is exactly the sam

e as speak- ing in a fonn that gets anyone w

ho is listening to respond to w hatever w

e arc saying. T

he point of speaking in this w ay is to have som

ething take placc--to accom

plish som ething w

ith the assistance of som eone else.

T he act of speaking in a form

that gets som eone else to act is com

m only

called a speech act. T he form

that a speech act m ust have w

ill be clear from

the follow ing exam

ples: (I) Y ou assert that duck season has begun (you

m ight actually say, "D

uck season has begun!"). (2) She dem ands that w

e all go duck hunting (she m

ight actually say, "L et's go duck hunting!"). (3) I

prom ise to roast duck for dinner (I m

ight actually say, "I'll cook!"). T he

generic form for a speech act is: I (you, etc.) hereby assert (dem

and, prom -

ise) to anyone hearing m e that som

e state of aIT airs exists or can be

achieved. T he three exam

ples suggest that speech acts fall into three catego- ries, here called assertive speech acts, directive speech acts, and com

m is-

sive speech acts. W

hether speech acts accom plish anything depends on w

hether others respond to w

hat they hear. T he response to your assertion about duck sea-

son w as obviously positive. I, at least, accepted her inclusive but im

perative dem

and to go hunting w hen I prom

ised to cook. W e m

ay surm ise that both

of you accepted m y offer, and w

e all three w ent duck hunting, perhaps after

w e checked the new

spaper to be sure that duck season had indeed begun. W

hatever category a particular speech act falls w ithin, particular speech

acts im ply nothing about future situations. W

e start all over again w hen

deer season begins. A speaker m

ay assert the existence of som e state of

affairs and others m ay agree, or m

ay request som ething and others m

ay com

ply, or m ay m

ake a com m

itm ent that others accept, w

ithout any neces- sary consequences in the long run.

If, how ever, speakers frequently repeat a particular speech act w

ith the sam

e general effect, everyone involved begins to think that the repetition becom

es significant. W e end up hunting w

ith each other all the tim e be-

cause w e go through the sam

e cycle of speech acts w henever hunting sea-

son begins. C onstantly repeated, the sam

e old speech acts turn into conventionas everyone com

es to believe that the w ords them

selves, and not the speakers m

outhing them , are responsible for w

hat happens. H unting

together is w hat w

e do at certain tim es, w

hether any of us even have to say anything m

uch about it anym ore.

C onventions com

e close to being rules. R ecall that rules tell agents w

hat they should do. A

convention rem inds agents w

hat they have alw ays done.

C O

N ST

R U

C T

IV ISM

: A U

SE R

'S M .L

T he bordcrl_ine bcm

1ecn kno"'.ing that W C

have alw ays done som

ething and probably w

,11 continue to do 1t, and believing that w e should do it because

w e have al~ays done it, is exceedingly fuzzy. If a convention prom

pts agents to think t~at t~e~ should do som

ething that they have alw ays done,

then the convention 1s indeed a nile. W e should consider the rule in qucs-

t10n a w eak rule because it is norm

ative,w hich m

eans that agents accept the "should" elem

ent, only to the extent that the regular pattern of conduct (such as hunting together) continues.

A _s agents begin to realize that they should act as they alw

ays have, and not Just because they alw

ays have acted that w ay, the convention gains

stren_gthas a rule. R ules keep the form

of a speech act by generalizing the relat1~n betw

een speaker and hearer. W ithin the general form

of a speech act, given rules rna~e hearers into agents to w

hom those rules apply. Fi-

nally, agents recognize that they should follow the rules in question because

they are rules and for no other reason. R

ules can take the general form of speech acts in each of the three

catego_ric_spresented above: assertive speech acts, directive speech acts, and cornrn1ss1vespeech acts. R

ules in the form of assertive speech acts inform

agents about the w

orld---!he w ay things are, the w

ay it w orks--and inform

them

w ha! consequences are likely to follow

if they disregard this inform a-

tion. T ~e m

f~rm ation contained in such rules m

ay be stated in very general term

s, _m w

hich _case w e m

ight call it a principle.T he principle of sover-

eignty 1s a conspicuous exam ple.

A t_the other end of the spectrum

of possibilities, rules in the form of

assert,:'e speech acts m ay be stated in very specific term

s. Instructions for oper~tlng appliances, filling com

m ittee seats, or presenting diplom

atic cre- dentials are useful ex~rnples. W

herever rules in this fonn fall on the spec- trui:n, they arc m

s11,1ct1on-n,/es. Providing inform

ation is not norm ative, but

telling agents w hat they should do w

ith that inform ation is. A

gents alw ays

know w

hat they should do because the rule tells them som

ething useful about their relation to the w

orld. D

irective speech acts are recognizable as im peratives. If the speaker says

that you n:iust do s_om ething,the speaker w

ants you to believe that you should ~o it. R

ules 1?the fonn of directive speech acts, directive-rules,are em

phatically nonnative. B y telling agents w

hat they m ust do (no hunting')

thes~ rul_es leave _no doubt as to w hat they should do. D

irective-rules oft~~ prov_1demfo~aho_n about th_e ~o~sequences 'for disregarding them

. H aving

this_in~orm at'.o? (sixty days m

Jail!) helps rational agents to m ake the right

choiceIn dec1dm g whether to follow

these rules or not. C

om rnissive spcec~ a~ts involve prom

ises. Speakers m ake prom

ises that hearers accept. C

ornrn1ss1ve speech acts give form to rules w

hen hearers as '

69 68

N IC

IIO L

A S O

N U

F -

. h ises of their ow

n. O nce these w

ebs of prom ises

speakers, respond w it

prom ,. d

d nonnative in their ow n term

s, they Suffi ciently genera ,ze

an .

h I

becom e

t likely to recognize t ese ru es becom

e co111111it111ent-ru/es. d d

that agents know they

A gents ar~ m

os .

T h

effects are the rights an utres

in their effects. ese

. agent's rights constitute

s w ith respect to other agents. A

ny given posses

. rty-no

hunting!). duties for other agents (pnvate prope

. them

to specific benefits. R ights

R ights m

ay entitle the agents possessm dg other agents in specific w

ays. agents to act tow

ar 1

m ay a so em

pow er

. .

o ers tum

people into agents. M ore

O bviously, pow

ers and 1tm 1ts on ~e ~nto agents by defining opportunities

generally, right and duties tum ,r~p struction-ru!es and directive-rules also

for them to act upon the w

or .

n .

ts for exactly the sam e reason.

tum people into agen

. b cause they perform

different Speech acts fall in_to three c~tegsor;:~er: and hearers together in three,

fi111ctions-they get things done hor p t gon· es hold for ni!es because they h

ays T he sam

e t ree ca e b

1 and on y t ree, w

·

h ch acts do-they

get things done Y

w

ork in the sam e three w

ays t a_t ~pee nts A

s observers w e see rules in

· d. e t·ng and com

m itting age

· '

h instructing, ,r c,

, . different functions for society. Q

uite~ few sc o_- 1

each category perform ing

d ..

lo y have w orked out variattons on this

ars in such fields as law an hsoc10 g

' sed all three of these categories,

functional schem e, but they

ave nev_eru and just these categori~s, ?tt~e s;i~:r~:e~chem

e for categorizing rules, and Philosophers have

evise a

d d it O

n functional grounds, a num

ber of constructivist scholars have ?tuotpvtee rul~s and regulative rules.

cries of rules· const1 1

there are tw o categ

d" .

f cial construction. R

egulative rules C

onstitutive rules are the m e lllm

o so

are the m edium

of social_ cttrol. to be constructivist it is actually a source

W hile this schem

e m ig t seem

. . t

f view ' all rules are alw

ays of confusion. From

a construct1v1st poi? o B

y definition rules regulate .

. d

u!ative at the sam e t1m

e. '

const1tut1vean reg

ative--they tell agents w hat

the conduct of agents because rules al r~ nnorm t·

conduct constitutes the w orld

F rtherm ore the regu at10 o

. l

Id d

they s 1ou o.

u '

w hether agents intend this conse-

1 w

ithin w hich such conduct takes Pace,

. h

Id often as an .

. the w

orld m eans acting on t e w

or '

quence or not. A cting m

.

. ht be a useful w ay to categorize

· tended consequence. Intentions m 1g

. . unm

d

. . b

1 ·s for categonzm

g rues. acts, but they are never a

ec1S1ve as, .

I rule serve only to regulate

E ven w

hen ag~nts intend that a pa::u ~~w

art by choosing, for exam -

conduct (an intention that other ag~nts . y

estion w ill have the effect of

pie, to disregard the ru!e),c~~o~~~o ~~::;g:

it) w eakening the rule. In the

strengthening or (,fagents .

I t be constitutive w

ill have to affect a rule that agents intcnc c,

. I

sam e w

ay, ' '

O "

ts intend rules to be sim ultaneous y

conduct if it is to succeed. ,ten agen

C O

N ST

R U

C T

IV ISM

: A U

SE R

'S-U A

L

constitutive and regulative. T o give an obvious exam

ple, w hen agents called

players take turns in playing a gam e, the rule instructing them

to do so constitutes the gam

e as one in w hich players regularly take turns.

A s w

e have seen, rules serve three possible functions. A gents m

ake rules and use them

for instruction, direction, and com m

itm ent. W

ithin each of these three functional categories, rules differ in the extent to w

hich they have been fonnalizcd. R

ules are form al if agents encounter them

as fixed and unavoidable features of their w

orld. R ules also differ in the extent

w hich they are linked to other rules. A

gents often discover that particular rules are linked to other rules telling other agents w

hat to do in the event that the relevant agents disregard the particular rules in question. Form

al rules that are effectively backed up by other rules are legal.

Form ality strengthens a rule by m

aking its nonnative character clearer, in the process separating it from

rules that are norm atively m

ore am biguous

(conventions, for exam ple). A

rule supporting another rule strengthens the latter by increasing the chances that agents w

ill choose to follow the latter

rule. T he m

ore frequently agents follow a rule, the stronger the rule w

ill be, norm

atively (and the easier it w ill be to m

ake it form al). For exam

ple, the principle of sovereignty is a highly form

al instruction-rule constituting the society of states. It is supported by com

m itm

ent-rules em pow

ering states, as agents, to bring new

m em

bers into this society. T hese supporting rules,

w hich w

e know as rules of recognition, are supported by instruction-rules

that spell out a num ber of social and the m

aterial conditions that m ust be

satisfied before statehood is possible. A

gents are inclined to m ake rules legal and to follow

them if they are

legal because they know w

hat the rules are, how m

uch they m atter to other

agents, and w hat consequences they can expect from

not follow ing them

. W

hen agents find them selves in a legal environm

ent, it is rational for them

to follow rules as a general proposition. It costs them

less than careless conduct w

ill. International relations is a peculiar environm ent in this re-

spect, but still a legal environm ent. W

hile there are very few fonnal direc-

tive-rules to be found, there are large num bers of other, quite form

al rules intricately linked in support of each other. R

elevant agents are perfectly aw

are of the situation and proceed accordingly.

R ules Form

Institutions, Institutions Form Societies

R ules are linked to each other in content as w

ell as function-both by w

hat they say and by w

hat they do. Standing back, agents can easily identify the w

ays that rules reinforce each other in w hat they say and do. Speaking

figuratively,w e m

ight say that rules com e in fam

ilies, and that som e fam

i-

70 .L

A S

O N

U F

lies of rules com e w

ith rules docum enting the fam

ily pedigree. O ther fam

i- lies of rules depend on observers to docum

ent fam ily resem

blances. T hese

and m any other practices help to give farr.ilies of rules their distinguishing

features. R ules and related practices arc alm

ost im possible to separate in

practice, because every tim e agents respond to rules, w

hether by m aking

choices or by observing the choices that other agents m ake, they have an

effect on those rules and on their places in fam ilies of rules.

13y recent convention, scholarly observers of international relations call these fam

ilies of rules and related practices "regim es." A

t an earlier tim e,

they called them "institutions," and this rem

ains the usual term for m

ost scholars w

ho devote their attentions to social relations. In practice, the. tw o

tenns are indistinguishable. International regim es are said to consist of

principles, rules, nom 1s, and procedures. B

y w hatever nam

e, these are all categories of rules. Principles and procedures anchor the tw

o ends of a spectrum

of possibilities distinguishable by how general they are in content.

R ules and nonns are distinguishable by how

fonnal they are, norm s being

sufrtciently infonnal that observers are not alw ays sure that they are rules

until they see how other agents respond to them

. International regim

es differ in size. T hey have rules that w

ork in differ- ent w

ays (assertive-, directive-, and com m

itm ent-rules) in different propor-

tions. A dditionally, regim

es differ in the extent to w hich they have rules

backing up other rules. Institutions differ in exactly the sam e w

ays. T hey

are m ade up of rules that vary, not just in generality and form

ality but also in num

ber and arrangem ent.

Som e sim

ple institutions consist of a sm all num

ber of rules w hose con-

tent m akes them

a fam ily, even if the rules seem

to give little support to each other, and to get little support from

other institutions to w hich they are

connected. In the w orld of international relations, the balance of pow

er is an exam

ple of such an institution. Instruction-rules constitute, and regulate, the balance of pow

er. T hese rules tell the great pow

ers w hat to expect w

hen they choose allies and go to w

ar. Y et even the balance of pow

er, as an institution, is not as sim

ple as it seem s. T

reaties give allies rights and duties. R

ules lim iting the conduct of w

ar help to keep the balance from being

perm anently upset.

In the context of international relations, spheres of influence are also sim

ple institutions m ade up of inform

al directive-rules. These rules direct w

eak states w ithin the sphere to carry out a m

uch stronger state's w ishes.

W hen these rules are backed up by principles justifying such arrangem

ents, the sphere of influence is no longer quite so sim

ple an institution. A s fonnal

equals, states m ay also adopt treaties distributing rights and duties that have

unequal consequences w ithin the sphere. T

reaties are them selves sim

ple

C O

N ST

R U

C T

IV ISM

: A U

SE R

'S --MAN

U A

L

71

institutions m inim

ally consisting of fonnal com m

itm ent-rules that apply

o~ly_ to the states adopting such treaties. T he principle that treaties are

bm dm

g, and therefore legal, autom atically provides them

w ith support from

other, highly form

al rules. Institutions such as the balance of pow

er, spheres of influence and trea- ties are sim

ple only because observers can easily pick them ~ut of an

institutional environm ent characterized by a large num

ber of linked rules and. rel_ated practices. A

gents act as observers w hen they recognize any

m st1tut1onas such, no m

atter how com

plex it is. Scholars often think of international regim

es as som ething that they alone can see, w

hile agents can see only the sim

pler institutions m aking up the regim

e. Y et observers be-

com e agents, and regim

es becom e institutions, w

hen other agents !cam

w hat observers have to say.

. l~te:national regim

es are hard to see because the rules connecting the m

stitut1ons that m ake them

up tend to be infonnal. A gents take them

for granted. Form

al rules m ake things clearer, and agents need not stand back.

For a long tim e in the context of international relations, agents have had

access to a legal institution, conventionally know n as the sources of interna-

tional law , through w

hich they can m ake legal rules and thus institutions

w hose existence no one can doubt. T

reaties are one such institution thanks to the legal principle that treaties are binding on the states adopting them

: A

gents respond to rules w ith goals in m

ind; institutions serve their inter- ests .. A

s a general, ~atter, sim ple institutions have a m

ore straightforw ard

relation to agents interests than do m

ore com plex and m

ore difficult to '.ecog~i~e instituti_ons. We think of relatively sim

ple institutions as perfonn- 1ng distinct fun~t1ons for agents and for other institutions. D

epending on w

hat these relatively sim ple institutions do, they give priority to rules in one

of the functional categories that w e have already identified.

W hen instruction-rules are m

ost in evidence, agents are situated in net- w

orks of ruies and related practices. 'rhe balance of pow er is an exam

ple. Its rules assign ~n elevated status to a few

great pow ers (ideally five states)

that m ust act as 1fthey are roughly equal in the resources available to them

. If state~' ~gents act as, instructed, the consequences are supposed to be an ever-sh1ftm

g and re!attvely peaceful balance of alliances am ong the great

po~ers, w hatever the im

m ediate intentions of their agents m

ight be. R ecog-

nizing the balance of pow er as an institution w

hose function suits their interests, agents intentionally foster those sam

e consequences in the nam e

of the balance. W

hen directive-rules are m ost in evidence, agents are situated in a chain

of com m

and, a firm , or an organization.A

sphere ofinnucnce is a rudim en-

tary institution of this sort. Its very inform al rules assign each agent to an

72 -NIC IIO

L A

S O N

U F

office, as w e w

ould call it in a m ore form

al organization. O fficers rep_ort up

the chain of com m

and and carry out orders that com e dow

n the chain. B y

this \ogic, the top officer decides w hat the organization's function is. In

practice, m ost organizations are m

ore com plex tha~ this. N

_everthe!ess, _a sphere of influence is so rudim

entary in organization that its function 1s nothing m

ore than to fulfill the w ishes of a leading pow

er, as top officer, over the w

eaker states w ithin the sphere.

Finally, w hen com

m itm

ent-rules are m ost in eviden:e, ~ge~ts end up in

partnerships, or associations,w it_h other agents. I~ the 1nst1tut1onal

cont~xt of international relations, the pnnc1plc of sovereignty and the supporting rules of recognition m

ake states into form al equals. W

hen tw o or m

ore states adopt a treaty, they act as m

em bers of an associatio~ giving them

at least som

e rights in com m

on, including the right to com m

it t~~m sel v~s to

each other. U nder the term

s of the treaty, all parties take on add1t10nal rights and duties w

ith respect to the others. In this situation, states are form ally

equal because they all have the sam e _role.T

he_ function of any association is to distribute roles to agents through its com

m ttm

ent-ru\es. O

nly states (and the associations that _they have cre~ted_ by tr~aty) can adopt treaties, because there is a corrum

tm ent-rule assigning thrs :ale to

them exclusively. T

o return to an earher exam ple, m

arkets function by assigning agents either of tw

o roles-they are either sellers or _buyers.

E very seller is form

ally equal in possessing the right to buy, a~d so 1s every buyer. N

ote, how ever, that neither sellers nor buyers have a nght to a fix:d

price. Form ally speaking, agents in these roles are free to com

pete w ith

e h other presum

ably for the good of every agent in the association. T he

ac ,

, h

function of this, or any, association is im plied by the com

m itm

ents t at agents have m

ade to a given distribution of roles.. . ,

, It is im

portant to note, how ever, that an assoc1at1on s roles are not gener·

ally equal in the rights and duties that they create. T hink, for exam

ple, of the roles that m

em bers of m

ost households have. For that m atter, agents

holding the sam e status (for exam

ple, w hite m

ales) are equal to each oth~r w

ithin the term s of that status, even if different statuses are unequal in

relation to each other. T his is no less true for agents holding the s~m

e or sim

ilar offices (for exam ple, foreign m

inisters). N evertheless, com

m 1tm

ent- ru\es are especially useful for m

aking large num bers of agents form

ally equal for lim

ited purposes. .

. .

A gency consists of statuses, offices, and roles. D

epending on the m st1tu-

tional context, every agent m ust have a status, an office, or a role. M

ost, perhaps all, agents have all three in som

e com bination. !hi~ is ?ecause m

ost people are agents in a variety of institutions, and m

any 1nst1tut10ns com

bine features of netw

orks, organizations, and associations.

C O

N ST

R U

C T

IV ISM

: A U

SE R

'S -

M A

N U

A L

71

Institutions such as these arc com plex in function and structure. Instruc-

tion-, directive-, and com m

itm ent-rules arc all present, even if the propor-

tions differ from institution to institution. O

bservers usually have no difficulty in picking out the pattern of rules, because institutions are social arrangem

ents that alw ays reflect agents' interests. From

an observer's point of view

, institutions have purposes. It seem s this w

ay even if the observer is an interested agent.

A com

plex institution w ill have general instruction-rules, or principles,

telling agents w hat the purposes of that institution are. D

etailed instruction- rules m

ay provide support for these principles by spelling out all relevant statuses. D

irective-rules m ay also repeat and elaborate on w

hat these princi- ples have to say and then support them

by dem anding that officers do w

hat these rules say that they should. In situations w

here there are no conspicu- ous instruction-rules or directive-rules supporting principles, com

m itm

ent- rules create roles for agents that have, from

any one agent's point of view ,

the unintended effect of supporting the institution's principles. R

ules in all three categories

often w ork together

to support an

institution's principles. Som etim

es, how ever, institutions develop in such a

w ay that rules from

one or even tw o categories are scarce or not to be found

at all. If w e consider international relations as taking place w

ithin a single, overarching institution, its rules constitute a conspicuously lopsided ar- rangem

ent. T hanks to the principle of sovereignty, there are few

if any form

al directive-rules. O bservers w

ill discover inform ar directive-rules in

practice, even if som e agents routinely deny that such rules exist.

C onsidered as a com

plex institution, international relations takes place in a context w

here agents and observers find a large num ber of

form al com

m itm

ent-rules (rules of international law ), behind w

hich there is an even larger num

ber of instruction-rules. T hese latter rules

differ enorm ously in form

ality (quite a few are legal rules), detail, and

the degree to w hich they are linked to each other. T

hey support the principle of sovereignty and a few

other principles m ore or less directly

and effectively. T hanks again to the principle of sovereignty, states are

com plex institutions w

ithin w hich form

al directive-rules allow agents to

act on behalf of states in their relations. T

he context w ithin w

hich any institution functions as an agent is itself an institution. Society is a com

plex institution w ithin w

hich m any other related

institutions are to be found. A gents are likely to act as if their society's

boundaries are clear and accepted, even if observers, including agents, have a hard tim

e specifying those boundaries to anyone's satisfaction. States are societies that have exceptionally clear boundaries as w

ell as highly devel- oped institutions for conducting relations w

ith other states.

-74 NICIIO L

A _S O

N U

F

T he com

plex institution w ithin w

hich states function as relatively self- contained societies is itself a society. W

ithin international society, states function as prim

ary agents sim ply by conducting relations w

ith each other. International society includes m

any other, m ore or less self-contained insti-

tutions. Som e of them

add secondary agents, such as officers of interna- tional organizations, to that society. T

he sum total of institutions and their

relations add up to a society of staggering com plexity and constant change,

even though its large patterns seem at least to som

e observers to call for generalization.

R ules Y

ield R ule

W e have seen that institutions consist of related rules and practices. It is

possible to think of a single rule as an institution. A s a practical m

atter, w e

never find a single rule standing by itself. E very rule gives the agents to

w hom

it applies the choice of follow ing the rule, or not, w

ith m ore or less

predictable consequences. M

ost of the tim e, agents choose to follow

the rule. T he pattern of agents'

choices has a general consequence, w hether or not it is intended by particu-

lar agents--it has the effect of distributing m

aterial and social. benefits am

ong agents. A n extrem

ely im portant category of such benefits 1s control

over resources and control over other agents and their activities. Som e

agents benefit m ore than other agents. O

ver tim e, institutions w

ork to the advantage of som

e agents at the expense of other agents. A

s rational beings, those agents w ho benefit the m

ost from the rules that

apply to them are the m

ost inclined to follow those rules. ~gents w

ho benefit less are still inclined to follow

the rules because doing so still benefits them

m ore than not doing so. N

evertheless, agents m ay proceed to

break any given rule after w eighing the consequences of either choice for

them selves. A

s a general consequence, rule breaking is likely to involve a loss of benefits to other agents.

A gents w

ho are negatively affected by the breaking of a rule also have a choice. They m

ay accept the consequences (including a w eakened faith in

the broken rule and a greater chance of its being broken again). A lterna-

tively, they m ay choose to follow

a rule that has the consequence of pre- senting the rule breaker w

ith a loss of benefits, w hich the rule breaker ·is

either prepared to accept or had thought w ould not be likely to occur. T

he second choice, w

hich w e think of as enforcing the rule, involves using

resources that m ight otherw

ise have been put to beneficial use. T his loss of

benefits is still less than the loss that com es from

not enforcing the rule. . --~·~

.,,t.,.. A

" ..,,...t h""""fitfrf'lm

fn11nW ·

C O

N ST

R U

C T

IV ISM

: A U SE

R 'S - M

A N

U A

L

75

ing it m ay choose to use w

hatever resources are needed to change that nile, and thus to change the distribution of benefits that results from

the rule's existence. If som

e agents try to change the rule, other agents w ho w

ould benefit less from

the changes m ay choose to use the necessary resources to

keep the rule from changing. Furtherm

ore, those agents w ho benefit 1hc

m ost from

a given rule w ill probably have to use few

er of the resources available to them

to keep the rule.from changing than w

ill agents w ho w

ant to change the rule. C

learly, rules say w hat they say, and institutions are

slow to change, because agents m

ake rational choices in circum stances that

alw ays give the advantage to som

e agents over others. T

he general consequence of agents' responding to rules w ith the re-

sources available to them is that som

e agents exercise greater control over the content of those rules, and over their success in being follow

ed, than other agents do. In other w

ords, rules yield rule. B y m

aking agents and society w

hat they are, rules m ake rule inevitable. R

ule is som ething that

agents do to, and for, other agents, and they do it by follow ing rules. R

ule is som

ething that happens to agents w hen they follow

rules or w hen they

suffer the consequences ofnot follow ing rules.

Specific institutions m ay form

alize rule by seem ing to lim

it its exercise to a particular agent or set of agents-lo

rulers. Just because w e can identify

rulers, w e should not conclude that they alone do the ruling. W

herever there are inform

al rules (w hich is everyw

here), there is inform al rule, either sup-

porting or undercutting form al institutions of rule, or both (probably in a

com plex and hard to observe pattern), E

ven if the form alities of rule are

now here to be found, rule rem

ains a pervasive condition for that society. L

oaded w ith rules but lacking rulers, international society is a case in point.

R ules in different functional categories yield different form

s of rule. W

here instruction-rules are param ount and status is a defining feature of

society, ideas and beliefs seem to do the ruling. D

espite appearances, agents actually do the ruling by getting other agents to accept their ideas and beliefs. T

hey do so by exam ple and by indoctrination. R

ule in this form is

hegem ony.

A ny society w

here principles get m ost of their support from

detailed instruction-rules is hegem

onically ruled. C aste societies are exam

ples. E ach

hegem onically ruled caste has clear boundaries and a fixed position in the

netw ork of castes constituting the society, M

em bership in a caste gives

agents so m uch of their identity, defined as a set of ideas about self and

position in society, that caste identity seem s to rule the society as a w

hole. H

egem onically ruled institutions exist in societies w

here other sorts of insti- tutions and a m

ixed form of rule can be identified. The professions offer an

exam nle. D

etailed instruction-rules, ordinarily learned through a long an-

-76 NIC I IO

LA S

O N

U F

prenticcship, support professional standards and rule agents to their advan· tagc in their relations w

ith clients needing their professional services. In institutions w

here directive-rules are param ount and office is a defining

feature of society, offices are vertically organized in a chain of com m

and. O

fficers at each position in the chain use resources that their offices m ake

available to them to carry out the rules that their offices require them

to carry out. From

top to bottom , such an arrangem

ent of offices is called a hierar- chy, and so m

ight w e call the form

of rule that results w hen officers carry out

directive-rules. T he state as a legal order exem

plifies hierarchical rule. W

hen directive-rules are legal, hierarchy is form al. D

espite the m inim

al description of the state as a legal order, form

al hierarchies rarely stand alone. 1legcm

onical ideas typically reinforce form al hierarchy. T

he result is a,1thority, conventionally defined as legitim

ate control. M ilitary officers

possess authority according to their rank, w hich is their status and office

form ally joined together in m

utual reinforcem ent. Finally, inform

al hierar- chy m

ay reinforce hegem ony that has achieved a relatively high level of

form ality. A

fter W orld W

ar ll, the so-calledpax A m

ericana m ay be thought

of as a condition of rule in w hich the U

nited States ruled, in the nam e of

freedom and prosperity, by intervening w

henever and w herever it chose.

Proclaim ing principles had the effect (perhaps initially unintended) of for-

m alizing the status of the U

nited States as leader of"the free w orld," w

hile acting on those principles gave it an inform

al office. W

here com m

itm ent-rules are param

ount and role is a defining feature of society, agents hold a variety of roles that are defined by reference to the roles that other agents hold. N

o one role, or institution, even com es close to

m aking particular agents into rulers. O

n the contrary, form al com

m itm

ent- rules m

ostly seem to reinforce form

al hierarchy. T hey do so by granting

officers w ell-defined pow

ers to help them issue orders and carry them

out, and by granting agents w

ell-defined rights to help protect them from

offi- cers abusing their pow

ers. T he result is a constitutional state, in w

hich the constitution form

alizes com m

itm ent-rules that lim

it the governm ent of the

state and m ake it responsible.

T aken as a w

hole, roles m ay yield rule on their ow

n, and not just because they reinforce other form

s of rule. A gents in association are the rulers--all

of them together-even

if none of them have the status or office to m

ake them

rulers. R uled by association, agents do not see rule in their roles. A

s agents, they are m

ostly concerned w ith their roles and w

hat they are free to do w

ithin them . T

o return once m ore to the exam

ple of a m arket, agents

participating in it generally have the sense that this is an institution free of rule. A

s sellers and buyers, they are nevertheless ruled as an unintended ·····•·•

~r ,ho overri<e of their right to buy and sell. A dam

Sm ith's

C O

N S

TR U

C TIV

IS M

: A U

S E

R 'S

M A

N U

A L

77 -

invisible hand is a hand that rules and it agents over others.

. ' rules to the advanlagc of som

e

A s w

e saw , quite a few

sch 1 d

. .

anarchical. A n anarchy is rule bo a;s

csc:tbe to_tcrnational relations as everyone in association

as y . o one to particular, and therefore by

, an unintended conseq

f h .

uncoordinated acts. R ecall that a

ucnce o t etr m

any, 1

unintended consequences can n rnts

w 10 o~serve a ge_neral paltern of

consequences even if the o anger be said to act w

ithout intending intend to be 'ruled

for g~ c~nttoue to act as they had been acting. T hey

O

reasons and if they d'd

h reasons, they w

ould m ake other choice's.

, not

ave good If anarchy is a condition of rule un

1 then international

relations . re ated to any agent's intentions

. .

1s no anarchy W

e need 1

' todtcate the form

of rule in w h' h

: anot 1cr term

to w

hat seem to be unintended

tc agents totend that they be ruled by

consequences of excrc. . h .

. eteronom

y is a better term A

t ,sing t etr rights.

H

eronom ous agents cannot a~t

o~om ~u\agents

act freely, w hile het-

society. From a constructivist

rec y .. ot

term s refer to agents, not

autonom ous, but their autono:rs~:ct\ve,

ho"'.e".er, agents arc alw ays

autonom y of other agents

T h

y w

ays ltm tted by the (lim

ited) .

e exercise of autonom

k h

om y a social condition

w h' h

Y m

a es eteron-

' tc

agents accept as an a l

. ten ed consequence of their ind'

'd p_rarent y un1n-

d ~nt~m

ational society is heter~~:;~~~;utonom ous

choices. their todependence under the p .

. l f y ruled because states exercise

of com m

itm ent-rules grantin

~nctp o

sovereignty and under a num ber

other. O ne state's independc g

~m nl~h'.s and duties w

ith respect to each nee ts a

im tt on every othc '

d 1

agents accept the unintended rs,

an a 1states'

vidual choices. W ithin this gcc:c"selquend~e~

that result from their m

any indi- ra con 1t1on of rule arc to b

r, d

1 num

er of institutions contributin to rul .

. '

e oun a arge

b not just states'

agents) constantf e to a vanety of w

ays. A gents (and

w ithin them

. D espite their n

b Y

~ork _on these institutions and w ork

relations, they are arranged a~~h er an vanety, and the com

plexity of their . .

ey are on purpose by age t , · .

serve th etr toterests--including the' h

d. , .

. n s tntent1ons, to tr s are

tnterest m betog ruled.

N ote

K urt B

~rch, H arry G

ould, and V endulka K

ubal expos,tion of constructivism

as I h d d 1

~o_vapersuaded m e to w

rite a concise result is "A

C onstructivist M

anifes:o" ("c;~uor:i ti m 1¥_orldof O

ur M aA

ing ( 1989). The for a scholarly audience introduc·

97), w hich I w

rote in a telegraphic slyle 'I

I '

m g som

e new m

at . I

d I .

t e had planned m

y essay for th' b k

ena an eavm

g a great deal out W

h the .".M

anifesto" for a larger audie~ee o~ e~d:;entence-~y-sent~ncc reconstruction of add1t1onsond changes, and I deleted ali of.

I ~p m aking quite n few

substantive Its re atlvely few

citations. I am grateful to

m em

bers of the M in.m

i International Relations G roup for their questions and suggcs~

tions.

llibliograplly

· I I

G

d J989

W orld o,f O

ur M aking: R

ules and R ule in Social

O nuf N

ie 10 as rccnw

oo , ·

, C

I'

p T

l;eorv and International Relations. C olum

bia:U niversity of South a:o m

a re~~- --

'1997, "A C

onstructivist M anifesto." In C

onstituting lntem at1onal ~o/w

cal £ c:onom

y, .

· ed

K urt B

urch and R obert A

. D enem

ark, 7-17. B oulder, C

O . L

ynne R

ienner.

- 4

W hat Is at Stake in the

A gent-Structure D

ebate?

H arnJD

. G ould

Introduction

A s notedin the introductory chapter to this volum

e, the division of interna- tional relations (IR

) into a series of debates has becom e a conventional

device for m aking sense of the field. B

eyond the "great debates," at least three and possibly five in num

ber (W rever 1997, 12-25), tw

o debates arc of interest here: the "levels of analysis" debate touched off by D

avid Singer in I 961 and especially the "agent-structure" debate initiated by A

lexander W

endt in 1987. Sim

ply put, the conceptual problem at the heart of the agent-structure

debate is: H ow

arc agents nnd structures related? O ver the course of the

debate, this problem has disappeared from

view , or, m

ore to the point, the debate itself has becom

e problem atic. There are several problem

s: the posi- tions held by som

e of the participants have changed over tim e, the tenns of

debate have changed, and the serial subdebates have becom e further re-

m oved boih from

the core issue and from IR

's substantive concerns. T

he purpose of this paper is to put O nurs constructivism

into the context of the agent-structure debate, expanding on his contribution to this book, his book W

orld of O ur M

aking (1989), and several articles w hich have also

served to refine his position (O nuf 1994, 1995, I 996, 1997). This w

ill entail a brief review

of constructivism , as w

ell as a detailed analysis in w hich I

shall put constructivism into dialogue w

ith the various positions staked out during the debate. T

he concluding discussion w ill look at the levels of

79