Norms and Institutions. Origins, Change and Effects

1
Norms and Institutions.Origins, Change
and Effects
Prof. Dr. Karl-Dieter Opp Universität Leipzig,
Institut für Soziologie (Emeritus) University of
Washington (Seattle), Dept. of Sociology
(Affiliate Professor) opp_at_sozio.uni-leipzig.de /
oppkd_at_u.washington.edu
2
Power point presentation can be downloaded from
http//www.uni-leipzig.de/sozio/mitarbeiter/m29/
content/dokumente/527/OppVorlesungNormen.ppt
Basic articles can be downloaded from
dropbox.com (You will get an email that allows
you to download the articles and book chapters
from a seminar about norms and institutions held
at the University of Washington (Seattle). The
folder includes the power point presentation.
3
Overview of Goals and Contents
This lecture is not intended as an inventory or
description of existing norms or institutions.
The goal is explanation

• explanation of the formation (or origins) of
  norms and institutions,
• explanation of the stability and change of norms
  and institutions, and
• (only briefly) explanation of the effects of
  norms and institutions.

Further, this is an introduction to, critical
analysis and extension of the most important
extant theories .
4
Content

• I. Introduction
• II. Basic Concepts
• III. The Measurement of the Norm Dimensions
• IV. Are Explanations of Norms and Institutions
  Theories?
• V. The Problem of Social Order The Classical
  Explanation
• of Norm Setting (Hobbes)
• VI. Peaceful Cooperation without a State
  Spontaneous Norm Emergence in the State of
  Nature and beyond (Axelrod)
• VII. Resolving Disputes in Shasta County Robert
  C. Ellicksons Theory of Spontaneous Norm
  Emergence
• VIII. Resolving Coordination Problems The
  Origins of Conventions
• IX. Externalities and Second-Order Public Goods
  Norms as Solutions of Collective Action Problems
  (Coleman)
• X. Private-Interest Sanctioning and the Emergence
  of Norms (Opp)

5

• XI. Other Mechanisms Some Suggestions
• XII. The Origin of Norms by Design Collective
  Norm Making
• XIII. The Effects of Norms and Institutions
• XIV. A Toolkit for Institutional Analysis
• XV. Summary An Inventory of the Mechanisms of
  Norm Emergence
• XVI. How to Get a Grade Requirements for
  Papers

Problem not clear how far we will come I will
probably drop some themes!
6
A Note The Curse of Forgetting
Prediction 90 of what you will have learned
until Thursday will be forgotten after say
half a year!
Here is the curve of forgetting (Hermann
Ebbinghaus)
7
What can you do to achieve a higher curve
forget less?

• Repeat what you have learned in the afternoon.
• Repeat everything next week maybe everyday part
  of the slides.
• Read the basic readings about the theories
  presented!
• Read more about norms!
• Write a paper!

Why might this lecture be didactically less
useful than a seminar?
Time between the different topics is short, so
there is no time for repetition in contrast to
weekly seminars (???).
8
I. Introduction
I.1. Norms in Everyday Life
Examples in everyday life what are norms in
this lecture?
What is not regulated in this room?
Some examples for spectacular norms?
Are there norms about sanctions?
Are there norms about what reward are acceptable?
How would you define the concept of norm?
9
I.2. Some Facts about Norms
The following facts should be explained by the
theories (or mechanisms) to be discussed

• Norms change over time examples?
• Norms differ across societies examples?
• Norms are conditional examples? (One must not
  lie correct?)
• There are relationships between norms e.g.
  constitutional law and specific laws.
• The specificity of what norms prescribe (
  normative content) varies across norms. E.g. the
  role of a father does not specify every detail of
  what a father has to do. Compare this with the
  detailed prescription of fines for the violation
  of traffic laws.
• The precision of norms varies across norms. See
  paragraphs of the constitutional law.

10
I.3. Questions about Norms

• Descriptive questions see the previous slide
  (what are "facts" about norms). These are
  descriptive statements about norms.
• Explanatory questions about
• origins
• stability and change, and
• effects of norms.
• Normative questions "validity" ("Geltung") of
  norms what "is" justice, how should/must or
  should not/must not one behave?
• Can such judgments be true?

11
I.4. What Social Science Deals with Norms and
how (Empirically or Normatively)?

• Sociology,
• political science (e.g. norms about international
  relations),
• anthropology,
• economics (property rights, law and economics )
• social psychology (internalization ),
• jurisprudence,
• philosophy.

12
I.5. Institutions

• How would you define the term institution?
• Examples for institutions?
• Some basic facts about institutions similar to
  those of norms
• Institutions change over time examples?
• institutions differ across societies and cultures
  examples?
• institutions (i.e. the norms constitutions are
  made up of) are conditional,
• there are relationships between the norms of an
  institution again constitutional law ...
• the specificity (normative content) of the norms
  of an institution differ across institutions
• the precision of the norms of an institution
  differ across institutions.

13
I.6. Questions about Institutions
These are the same questions as about norms

• Descriptive questions
• Explanatory questions
• Origins
• stability and change, and
• effects of norms.
• Normative questions.

14
I.7. What Social Science Deals with Institutions
and how (Empirically or Normatively)?
Same answer as for norms

• Sociology,
• political science (e.g. norms about international
  relations),
• anthropology,
• economics (property rights, law and economics )
• social psychology (internalization ),
• jurisprudence,
• philosophy.

15
II. Basic Concepts
II.1. Norms
The following list consists of some definitions
from the literature. There are many more
definitions!
Question What are the criteria used in the
definitions? (Perhaps make notes when we go
through the definitions.)
16

• (1) "A norm will be defined as an expectation
  about how one ought to act, enforced by the
  threat of sanctions or the promise of reward"
  (Kerr 1995 33).
• (2) "In this model social norm is a term for
  behavioral regularities that emerge as people
  interact with each other in pursuit of their
  everyday interests" (Posner 2000 26).
• (3) a social norm is (i) a behavioural
  regularity that is (ii) based on a socially
  shared belief how one ought to behave which
  triggers (III) the enforcement of the prescribed
  behaviours by informal social sanctions" (Gächter
  and Fehr 1997 276 italics in the paper).
• (4) R. H. McAdams (1997) summarizes the
  definitions of an extensive literature (see the
  quotations on pp. 339-340) "Roughly speaking, by
  norms this literature refers to informal social
  regularities that individuals feel obligated to
  follow because of an internalized sense of duty,
  because of a fear of external non-legal
  sanctions, or both" (340).

17
The definitional characteristics are

• expectations of third parties that an action
  ought to be performed (1) (NOTE this is not a
  "cognitive" expectation prediction)
• shared normative expectations (oughtnes) of
  third parties (3)
• sanctioning (positive or negative) of norm
  violation (1)(3)
• regularity of a behavior (2)(3) no reference to
  beliefs
• internalization of norm (feel obligated) (4)
• felt obligation of the performance of an action
  ( internalization) that is not formally
  prescribed (4)
• fear of sanctioning (4) and one could add
  expectation of rewards (positive sanctioning).
  PERHAPS IDENTICAL WITH SANCTIONING (see before)?

Some definitions consist of several criteria (4).
18
NOTE no author provides any reason why he or she
chooses the definition he or she suggests!
NOTE some definitions include causal statements
(2, 3, 4). Or are these separate empirical
propositions (2)?
Two questions arise
(1) Is a simple or complex definition (one
consisting of several criteria) preferable?
(2) What is the most useful dimension (property)
or what are the most useful dimensions to be
included in a definition?
19
(No Transcript)
20
What dimensions are most useful for a definition
of norms?
Suggestion oughtness should be the central
criterion. Reason This is rarely addressed in
the literature, as we will see, and
its explanation is of interest.
Thus
A norm could (and should!) be defined as a
statement that one should behave in certain
situations in a certain way or that one should
have certain attitudes, preferences or cognitive
beliefs.
Acceptance of a norm means the extent to which a
statement describing a norm (see definition
before) is approved of. (Internalization refers
to a relatively strong acceptance.)
21
Some implications

• All other definitional characteristics could be
  dependent or independent variables.
• Conditionality is taken account of in the
  definition.
• The norm does not only refer to behavior. Also
  cognitive beliefs or attitudes or preferences may
  be subject to oughtness see religious beliefs.

22
II.2. Institutions
Some examples for definitions

• (1) "We define institutions as a system of
  human-made, nonphysical elements norms,
  beliefs, organizations, and rules exogenous to
  each individual whose behavior it influences that
  generates behavioral regularities." (Greif and
  Laitin 2004 635)
• (2) "In general, historical institutionalists
  work with a definition of institutions that
  include both formal organizations and informal
  rules and procedures that structure conduct."
  (Thelen and Steinmo 1992 2).
• (3) "Institutions are the rules of the game in a
  society or, more formally, are the humanly
  devised constraints that shape human interaction"
  (North 1990 3).

23
Note that

• the definitions are very broad see the first
  definition institutions refers to almost
  everything!
• It is not clear why the definitions are so broad.
• Organizations and norms are denoted as
  institutions why? Is this theoretically
  meaningful?

Suggestion
We understand by institution phenomena for which
there are so far few convincing theories these
are institutions as rules of the game
(Definition 3 by D. North). Thus
Institutions are, by definition, systems of
norms.
24
Examples for institutions
Constitution of Germany, the US, and other
basic laws
Inheritance laws of a country
Voting rights
Institution of the market
Health care system
Institutions for regulating common pool resources
System of rules for getting a Ph.D. or, in
general, a doctorate
25
II.3. Values
There are different definitions as well. A
frequent definition reads values are relatively
general norms such as human rights.
Why do we need a concept that denotes relatively
general norms if it is not clear how general a
norm must be in order to be called a value?
Why not distinguish simply between more or less
general norms?
Sometimes values are defined as attitudes
(evaluations) here the term attitude/evaluation
is clearer!
Thus we dont need the concept of value!
However, using it in the above sense does not
hurt either.
26
II.4. Sanctions
There are different definitions as well, e.g.

• (1) "Sanctions refer to any kind of reactions to
  a behavior, positive or negative, that attempt to
  alter the behavior, or decrease or increase its
  frequency" (Labovitz and Hagedorn 1973 284).
• (2) "... people receive physical or psychological
  rewards and punishments for their behavior, which
  encourages or discourages them to conduct
  themselves in a similar way in the future"
  (Labovitz and Hagedorn 1973 284).
• (3) "The terms 'sanction' and 'effective
  sanction' will be used interchangeably,
  indicating ... an action on the part of a norm
  beneficiary that has some effect in moving the
  focal action in the direction intended by the
  sanctioner" (Coleman 1990 40).

27
The definitional criteria are

• Reactions to a norm violation with the goal to
  influence future behavior (1) intention of the
  sanctioner relevant
• reward/punishment from the perspective of the
  sanctionee the person who is the target of the
  sanction for following a norm or for refraining
  from norm violation (2)
• reaction to norm violation that changes the
  sanctioned behavior in the future (3 also
  second part of definition 1) effect of the
  reaction relevant.

28
Which definition is to be preferred?
We can thus distinguish the following definitions

• intention definition (first definition before),
• reward-punishment definition (second definition
  before),
• effect definition (third definition before),
• combination of the definitions.

One question social scientists are concerned with
is to explain when and how individuals react to
norm violations. This speaks in favor of the
intention definition.
29
Another interesting question is the extent to
which behaviors intended to change behavior
actually reach their goal, i.e. when are
sanctions effective? See questions of deterrence
or effectiveness of laws.
This speaks in favor of the effect definition and
the reward defi- nition (rewards are conditions
for effectiveness).
Conclusion
Each of the three definitions refers to
theoretically interesting phenomena either their
origins or effects or both are the subject of
explanations
Important when one speaks of sanctions one
should always tell what concept is used!
30
Note the following implication of the
definitions reactions to a norm violations that
are only intended to hurt the norm violator
(revenge) are not sanctions! Useful? (Example
instructor gives low grade!)
Perhaps the intention definition could be
extended
Sanctions df. reactions to a norm violation
with the goal to influence future behavior or to
hurt the violator (i.e. to impose costs on the
norm violator).
31
Question Do the definitions imply that a
sanctioner may be Interested in influencing the
behavior of non-violators (general deterrence)?
Definitions seem to refer only to the target of
the sanctions. Whether there is general
deterrence has to be determined empirically.
Alternative extend definitions
Question Does our theme explaining the origin
and effects of norms violate the postulate of
a value-free social science?
32
II.5. Cited Literature in II
Coleman, James S. 1990. "The Emergence of Norms."
Pp. 35-60 in Social Institutions. Their
Emergence, Maintenance and Effects, edited by
Michael Hechter, Karl-Dieter Opp, and Reinhard
Wippler. New York Aldine de Gruyter. Gächter,
Simon, and Ernst Fehr. 1997. "Social Norms as a
Social Exchange." Swiss Journal of Economics and
Statistics 133275-292. Greif, Avner, and David
D. Laitin. 2004. "A Theory of Endogenous
Institutional Change." American Political Science
Review 98633-652. Kerr, Norbert L. 1995. "Norms
in Social Dilemmas." in Social Dilemmas
Perspectives on Individual Groups, edited by
David A. Schroeder. Westport, Conn.
Praeger. Kuran, Timur. 1998. "Moral Overload and
Its Alleviation." Pp. 231-266 in Economics,
Values, and Organization, edited by Avner Ben-Ner
and Louis Putterman. Cambridge Cambridge
University Press. Labovitz, Sanford, and Robert
Hagedorn. 1973. "Measuring Social Norms." Pacific
Sociological Review 16283-303.
33
McAdams, Richard H. 1997. "The Origin,
Development, and Regulation of Norms." Michigan
Law Review 96338-433. North, Douglass C. 1990.
Institutions, Institutional Change and Economic
Performance. Cambridge Cambridge University
Press. Opp, Karl-Dieter. 2000. "Die Entstehung
sozialer Normen als geplanter und spontaner
Prozeß." Pp. 35-64 in Normen und Institutionen
Ursachen und Wirkungen, edited by Regina Metze,
Kurt Mühler, and Karl-Dieter Opp. Leipzig
Leipziger Universitätsverlag. Posner, Eric A.
2000. Law and Social Norms. Cambridge, Mass.
Harvard University Press.
34
III. The Measurement of the Norm Dimensions
In order to test hypotheses about norms (or
institutions) one needs to measure norms and
institutions.
Question Can norms be measured at all?
Interactionists claim that norms form in the
process of interaction. For example, whether a
defendant has committed a crime such as theft
(242 in the German StGB see next slide), i.e.
has violated a norm, is negotiated in the
court. Is the argument correct? What is
negotiated in the court?
How are decisions made if norms are vague such
as one should not harm others? Assume, you
mention this norm if somebody lights a cigarette
next to you, and the smoker thinks that this norm
does not apply in this situation!
35
242 Diebstahl (1) Wer eine fremde bewegliche
Sache einem anderen in der Absicht wegnimmt, die
Sache sich oder einem Dritten rechtswidrig
zuzueignen, wird mit Freiheitsstrafe bis zu fünf
Jahren oder mit Geldstrafe bestraft. (2) Der
Versuch ist strafbar.
36
In order to measure norms, the following
phenomena are to be measured all are components
or dimensions of the existing norms definitions

• behaviors regular behavior and sanctions as
  reactions to norm violations and, thus,
  behaviors
• attitudes and motives internalization norm
  conformity is an intrinsic motive ( acceptance
  of a norm)
• kinds of oughtness (should/must/is allowed to)
  and degree of oughtness
• cognitive beliefs about the conditions for a norm
  to hold (conditionality) and about the
  probability of sanctioning
• intention to punish or reward behavior.

Thus, the measurement refers to phenomena
external to an individual (behavior) or internal
(all the rest).
37
Which methods are most appropriate to measure the
norms dimensions mentioned before and what are
the problems of these methods for measuring norms?

• Interview?
• Problems?
• Observation (participant or non-participant)?
• Problems? (E.g. measurement of sanctions or
  attitudes? SEE SLIDE 39.)
• Analysis of documents?
• Problems? (What do people include in documents
  always the truth? Important possibility of
  access to events of the distant past! SEE NEXT
  SLIDE.)
• Qualitative or quantitative methods?

Thus each method has its problems. In measuring
norms, each method should be checked in regard to
its usefulness.
38
Question What are the problems of measuring
norms if documents are analyzed such as diaries
or minutes of a parliamentary session where norms
are reported?
Hint writing a text is a behavior (or a sequence
of behaviors). What could be the motives or
incentives to write a text in a certain way? Is
it always a goal of the writer(s) to tell the
truth? What about errors (misperception)
perhaps actors want to tell the truth but they
dont know it!
39
Question To what extent can sanctions be
measured by observation?

• If sanctions are defined as actions with the
  intention to punish a norm violation, then
  subjective phenomena must be ascertained which
  cannot be directly observed.
• This holds also for sanctions which are defined
  as behaviors that are rewarding for the targets
  of a sanction.
• If sanctions are simply defined as reactions to
  norm violation then causality must be determined.
  For this one needs a theory or experimental
  design!

40
Question Are there possibilities to avoid the
measurement of the norms dimensions because
measuring subjective phenomena is difficult?
(Think of revealed preferences!)
One possibility might be to infer the existence
of norms from behavior. The assumption could be
the more frequent a behavior is performed by the
members of a group, the more likely it is that
the behavior is normatively demanded. Correct?
Are there examples for regularly performed
behavior where no oughtness is involved?
Taking a certain route to the bus every morning,
taking the bus instead of a tram, buying
regularly the same yoghurt
41
Contents of the rest of this section

• Discussion of some conventional measurement,
  illustrated with the procedure of J.M. Jackson
• Jackson, Jay M. 1966. "Structural Characteristics
  of Norms." Pp. 113-125 in Role Theory. Concepts
  and Research, edited by Bruce J. Biddle and Edwin
  J. Thomas. New York Wiley.
• Some examples from existing research.
• Discussion of the factorial survey, based on
• Beck, Michael, and Karl-Dieter Opp. 2001. "Der
  faktorielle Survey und die Messung von Normen."
  Kölner Zeitschrift für Soziologie und
  Sozialpsychologie 53283-306,
• Opp, Karl-Dieter. 2002b. "When Do Norms Emerge by
  Human Design and When by the Unintended
  Consequences of Human Action? The Example of the
  No-Smoking Norm." Rationality Society
  14131-158 (see also Opp 2000).

42
III.1. The Procedure of J.M. Jackson
Jackson measures

• behaviors that are performed more or less
  frequently, and
• evaluations of the behaviors approval,
  attitudes, normative expectations ( oughtness).

Those dimensions can be measured in a
two-dimensional coordinate-system
43
Example frequency of raising your hand per
session in this lecture (or hours per week
reading).
E.g. evaluation of one other student or average
of evaluation of class
overachiever (Streber)
e.g. frequency of hand raising of a student
(may range from 0 to 8 per session empirical
maximum value)
slug (Faulpelz)
What are other possible curves e.g. curve of a
deviant?
44
Questions

• What does the curve mean? (Only oughtness?)
• Which dimensions could be represented on the
  y-axis and the x-axis?
• Could the curve also describe the norm of groups?
  (Example students of this class evaluate
  different frequencies of attendance and the
  average is computed. What different forms of
  curves could exist?)
• Is it possible to construct measures which
  describe different properties of a norm?
• Is there always a value of y for each value of x?
• Can conditionality be represented in the
  coordinate system?

45

• Can the change of norms over time be depicted in
  the diagram?
• Jackson constructs group properties, i.e. a group
  norm, on the basis of individual properties.
  Could a group norm be defined independently of an
  individual distribution of norms? IMPORTANT such
  measures are based on individual properties!!!
• Can the coordinate system be used for theoretical
  purposes? (What hypotheses could be formulated?
  E.g. could network density in a class affect the
  shape of the curve?)
• How could these properties be measured in a
  questionnaire?

46
Critique?
There is no theory so far! But Jackson formulates
some hypotheses, such as
the greater the consequence behavior has for a
group and its central concerns, the narrower will
be the range of tolerable behavior (p. 115).
The hypotheses are often not very clear (how can
the consequence and the central concern for a
group be measured?), not very informative, and
they are formulated ad hoc, i.e. without applying
a theory!
There is so far no systematic attempt to use
these measures in a theory.
Thus this is an example for a classification and
measurement, that may be used for descriptive
purposes. But the theoretical relevance is so far
not clear.
47
III.2. Some Examples from Research
From the DFG-project on the East German
revolution for details see
Opp, Karl-Dieter, Peter Voss, and Christiane
Gern. 1995. The Origins of a Spontaneous
Revolution. East Germany 1989. Ann Arbor
Michigan University Press. This is a translation
from German Die volkseigene Revolution (1993).
Items for the measurement of the acceptance of a
protest norm
Politics should be left to the elected
representatives of the people. A citizen should
become politically active only if he or she
believes that he or she can achieve
something. Violence against persons to achieve
political goals can be morally justified. If a
state oppresses free expression and other basic
rights of the citizens then also violence of the
citizens is justified. Five answer
categories, from fully disagree to fully
agree.
48
CONTINUATION
What do you think about the following
statement(s)
I think that violence against objects is morally
justified. Does this hold never, rarely,
sometimes, most of the time or always?
What are the situations when yourself would be
willing to use violence against objects
If some highway is constructed near your
apartment. If the right to demonstrate will be
restricted. If the police uses firearms against
demonstrators. . Answer categories yes or no
49
THE FOLLOWING IS FROM Opp, K.-D., K.
Burow-Auffarth, P. Hartmann, Th. von Witzleben,
V. Pöhls, and Th. Spitzley. 1984. Soziale
Probleme und Protestverhalten. Eine empirische
Konfrontierung des Modells rationalen Verhaltens
mit soziologischen Hypothesen am Beispiel von
Atomkraftgegnern. Wiesbaden Westdeutscher Verlag
(124-127)
If I wouldnt do anything against nuclear power
plants I would have a bad conscience.
If I do something against the construction of
nuclear power plants I sometimes feel that I do
something that I actually should not do.
Five answer categories fully agree, agree,
indifferent, disagree, fully disagree.
50
(No Transcript)
51
III.3. The Factorial Survey
The measurement procedures discussed so far are
not appropriate to capture the complex conditions
under which norms hold (i.e. their
conditionality). This is the strength of the
factorial survey also called vignette analysis.
Basic idea possible situations are described
where a norm could hold. Respondents are then
asked to tell to what extent a certain behavior
should be performed or to what extent an attitude
or cognitive belief should exist in the
respective situation (evaluation dimension).
52
Example you want to find out when it is allowed
to smoke in a restaurant. (Note the research was
conducted before there existed a ban on smoking
in Germany by law.)
First step specify under what circumstances this
norm may hold (these are hypotheses), e.g.

1. Class of the restaurant pub, other
2. number of persons in the restaurant who smoke.

Thus, two dimensions ( variables) may be
important. These are situational characteristics
which could be relevant for the norm to hold.
Each of these dimensions has certain values. E.g.
class of restaurant may consist of pub/other
0/1 (two values) number of persons who smoke
possible values none, few, many, almost all,
(four values).
53
To describe all possible situations, the values
of each dimension (variable) are combined. That
is to say, the Cartesian product could be
computed. In this case 2 x 4 8 Thus, there
are eight possible situations. Examples
There is a pub where nobody smokes.
There is a restaurant where few people smoke.
These descriptions of situations are called
vignettes.
54
Here are the dimensions and their values from the
research reported in Opp 2002 (reference see
below)
55
Here is an example of a vignette from the study
Mr. Müller goes to a restaurant. This is a top
class restaurant, in which smoking is prohibited.
There is no one in the restaurant who smokes. Mr.
Müller stays only for a short time in the
restaurant in order to have a beer. He smokes
most of the time more than a package of
cigarettes per day.
56
Second step the rating scale (evaluation
dimension) has to be specified which answers the
question
To what extent does a non-smoking norm hold in
these situations?
To answer these questions respondents are
presented with the following scale for each
vignette on which they could mark a value
It is in any case allowed to smoke.
It is not at all allowed to smoke
It is allowed to smoke
It is not allowed to smoke
57
Thus in applying the factorial survey to measure
norms the first step is to devise descriptions of
situations ( vignettes) combinations of values
of the dimensions. These are the norm-relevant
situations.
Second, for each vignette a rating scale is
presented in order to find out to what extent a
norm holds.
Third, a certain number of the possible (and
meaningful see later) situational descriptions
are assigned to each respondent by chance. Each
respondent thus gets a set of vignettes.
The size of the sets depends, among other things,
on the difficulty of answering the vignettes and
on the length of the questionnaire.
58
Practical problems
How many vignettes should a respondent be
presented with?
How many dimensions are appropriate in a vignette?
How many values should each dimension have?
Example number of persons. One could present
percentages from 0 to 100 with a distance of 1
(1, 2, 3...) this would result in a great
number of vignettes see next slide. Do
respondents discriminate between these values?
In any event the dimensions and their values
must be theoretically useful!
59
Are all combinations of the values of the
dimensions meaningful?
Example dimensions are age and occupation. Let
one occupation be university professor, and let
the age range from 16 to 65. One possible
vignette would include a professor with age 16.
Meaningful?
Statistical analysis How is a factorial survey
analyzed statistically? Here is the data matrix
for the non-smoking norm (Opp 2002)
60
(For the meaning of the codes see slide 54.)
Acceptance of the non-smoking norm is the average
of the judgments for the two situations per
respondent.t
61
Other examples for norms that could be measured
by a factorial survey?

• Conditions for self-defence (Notwehr),
• conditions for lying or telling the truth white
  lie (Notlüge) ...,
• when should one keep promises?
• When is it allowed to break a law or when should
  or must a law be broken such as participating
  in an illegal demonstration, blocking streets,
  occupying buildings (like university offices)?
• When do people think they should get a divorce?
  (See Diefenbach and Opp 2007 reference next
  page)
• What is sexual harassment?
• What sanctions are allowed for a given behavior?

Possibility for term paper!
62
What are the strengths and weaknesses of the
factorial survey to measure norms?

• Problem the situations are hypothetical. Often
  answers may not be very reliable. Pretest is
  important. Among other things, one should explore
  to what extent the situations in the vignette
  resemble real situations for the respondents.
• Strength 1 the factorial survey allows the
  detailed measurement of the conditionality of
  norms.
• Strength 2 it is possible to determine the
  weight of the dimensions, i.e. how strong their
  effect is on the rating of the oughtness.
• Other problems or strengths?

63
References for the Measurement of Norms with the
Factorial Survey
Beck, Michael, and Karl-Dieter Opp. 2001. "Der
faktorielle Survey und die Messung von Normen."
Kölner Zeitschrift für Soziologie und
Sozialpsychologie 53283-306.
Diefenbach, Heike, and Karl-Dieter Opp. 2007.
"When and Why Do People Think There Should Be a
Divorce? An Application of the Factorial Survey."
Rationality Society 19485-517.
Opp, Karl-Dieter. 2002. "When Do Norms Emerge by
Human Design and When by the Unintended
Consequences of Human Action? The Example of the
No-Smoking Norm." Rationality Society
14131-158.
Opp, Karl-Dieter. 2003. "Die Evolution sozialer
Normen." S. 139-205 in Evolution in den Natur-,
Sozial- und Geisteswissenschaften, hrsgg.
Von Andreas Diekmann and Rupert Moser. Bern Paul
Haupt. (Ausführlichere Fassung von Opp 2002
siehe vorangegangenen Literaturhinweis.)
Jasso, Guillermina. 2006. "Factorial Survey
Methods for Studying Beliefs and Judgments."
Sociological Methods Research 34334-423.
64
Recent contribution on the factorial survey in
general with further references
Auspurg, Katrin, Thomas Hinz, and Stefan Liebig.
2009. "Komplexität von Vignetten, Lerneffekte und
Plausibilität im Faktoriellen Survey." Methoden -
Daten - Analysen 359-96.
65
IV. Are Explanations of Norms and Institutions
Theories?
We will discuss the major theories about the
emergence of norms and institutions. Question
What is the relationship between these theories?
Do they contradict each other? Are some derivable
from others?
Thesis The theories describe different
processes of norm emergence ( mechanisms). In
other words, the theories describe different
constellations of conditions under which
different processes of norm emergence arise.
66
In order to explain the different processes a
general theory of action must be applied. This
ensures that causal statements are not ad hoc.
Most existing explanations of norms and
institutions apply often implicitly a wide
version of the theory of rational action.
We should ask for each theory of norm emergence
to be discussed what the background theory is
that is applied!
An important question that we will address
throughout this lecture is under what conditions
which processes obtain. This is not discussed in
the literature!!!
67
In general, what are the theories that could be
applied in explaining processes of norm emergence?

• Rational choice theory includes game theory
• important distinguish wide and narrow version
• functionalism,
• power theories,
• marxist hypotheses ...

For a discussion of some of the theories see
Mahoney, James, and Dietrich Rueschemeyer. 2003.
Comparative Historical Analysis in the Social
Sciences. Cambridge Cambridge University Press.
68
Some basics (perhaps you know this?)
How to explain singular events or relationships
between singular events the logic of explanation.
Example Why did smoking decrease in Germany in
2008?
Explanans Law The higher the costs of smoking,
the less people smoke. Initial conditions The
anti-smoking law in Germany in 2008 increased the
costs of smoking. Explanandum Smoking decreased
in Germany in 2008.
When is an explanation adequate?
69
V. The Problem of Social Order The Classical
Explanation of Norm Setting
A good starting point for an introduction to the
explanation of norms and institution is a general
question What are the origins of social order?
If this question is answered and if (!) social
order refers to a system of norms, then all
questions about the emergence of norms are
answered!
70
V.1. The Solution of the Problem of Social Order
by Thomas Hobbes
The most famous answer to the question of the
origin(s) of social order or how social order is
possible is by Thomas Hobbes (1588 to 1679)
Leviathan Or the Matter, Forme and Power of a
Commonwealth Ecclesiasticall and Civil, first
1651.
See in particular chapters 13 and 17.
71
The initial situation is a state of nature which
is characterized in the following way

• there are no internalized norms,
• all individuals are pure egoists,
• individuals are in a conflict situation, i.e.
  there is competition for goods
• this leads to hostility and the attempt to
  subjugate others
• there is no organization or state that prevents
  violence or issues laws
• individuals behave rationally, i.e. they try to
  maximize their utility by enhancing their
  material welfare.

The consequence is a war of all against all
(where every man is enemy to every man). Hobbes
characterizes the effects of this situation in
the following way
72
In such condition, there is no place for
industry because the fruit thereof is uncertain
and consequently no culture of the earth no
navigation, nor use of the commodities that may
be imported by sea no commodious building no
instruments of moving, and removing, such things
as require much force no knowledge of the face
of the earth no account of time no arts no
letters no society and which is worst of all,
continual fear, and danger of violent death and
the life of man, solitary, poor, nasty, brutish,
and short (chapter 13).
The last part after the (last) semicolon is a
famous text and not printed in bold in the
original!
Question why is there no infrastructure, culture
etc.? (For details see later.) Think of
function of property rights!
73
How does social order originate in such a
situation?
The basic idea is that a Leviathan (translation
is "sea monster"), i.e. a state, is created. How
does the state originate? Hobbess argument can
be reconstructed in the following way

• (1) Men love (their own) liberty and dominion
  over others (which is a condition for the war of
  all against all) ( motive 1)
• (2) men want their own preservation and a more
  contented life ( motive 2)
• (3) there is a belief ("foresight") that (only?)
  "restraint ( state) leads to realizing the
  second motive by imposing fear of punishment for
  pursuing the first motive
• (4) individuals are willing to transfer rights to
  a central authority ( motive 3).

74
Assumption seems to be that the second motive is
stronger than the first one otherwise people
would not be willing to sacrifice power in order
to gain a more contented life!

• (5) establishing order is not possible by
• covenants without a sword (why?), by
• "the joining together of a small number of men,"
  (why?) and
• being "governed, and directed by one judgment,
  for a limited time (why? think of stability.)
• (6) Consequence A Leviathan originates.

75
The causal structure of the argument thus is
Preference for domination Preference for
peace Belief that establishing a state is most
effective for realizing preference for
peace Readiness (preference) to transfer rights
to authority
-
(weak)
(strong)

Creation of a state (central authority)



Alternative means for order not promising
76
What could happen, if conditions 1 and 2
motives 1 and 2 (and perhaps other conditions?)
are given? Could there be other solutions than
the origin or creation of a state?

• Nothing the state of war remains stable
• small groups form who loot and kill others the
  physically strongest gain power (similar to
  alternative 2 by Hobbes)
• small groups unite to form a sort of
  authoritarian organi-zation like a Mafia (similar
  to alternative 2 by Hobbes) etc.
• neighborhoods form an association to protect each
  other (like neighborhood watch in the US)
• "spontaneous" cooperation e.g. somebody begins
  to "cooperate" in the sense that he or she builds
  up a peaceful relationship with his neighbor who
  reciprocate etc. (like the Axelrod model see
  later)

CONTINUED NEXT PAGE
77

• individuals hire others and pay them for
  protection see countries with a high crime rate
  where guards are hired
• Hobbess solution individuals transfer rights to
  a central authority. THIS IS THUS ONLY ONE
  POSSIBLE SOLUTION!

78
What are the central problems of Hobbess
solution?

• It is not clear what the "mechanism, i.e. the
  process, of the emergence of a state is. Assume
  there is a large group. Who initiates state
  foundation? Who gets which position in a
  government etc.?
• Under what conditions will people be willing to
  transfer rights, and act to found a state? Think
  of the problem of collective action (free rider
  problem). Common interest of the members of a
  large group does not suffice to realize the goals
  (Olson). It is not clear under what conditions
  certain passions (motives) and reason
  (beliefs) lead to peace.
• Who controls the state?
• Hobbess argument implies that there are no
  stateless societies. Empirical research does not
  confirm this implication.

79

• It is assumed that the state realizes the common
  good. See common good fiction in public choice
  theory.
• Is only external punishment for order relevant?
  Think of internalization of social norms.
• Hobbess model includes beliefs. Can beliefs be
  biased ( wrong)? For example, citizens may
  believe that others will reciprocate cooperation
  (see later Axelrod). Or they may not believe in
  the common good fiction (i.e. they do not
  believe that the state acts in the interest of
  the citizens). This might lead to unwillingness
  to transfer rights to a "Leviathan."

80
What is meant by social order?

• Social order may refer to cooperation in a
  prisoners dilemma will be discussed later.
  Intuitively refraining from behavior that hurts
  others like fraud, stealing, using violence.
  Thus, order means a certain class of regular
  behaviors.
• Taylor 1982, Community , p. 44 In the most
  restrictive of its common usage, social order
  refers to an absence, more or less complete, of
  violence, a state of affairs in which people are
  relatively safe from physical attack. On a
  somewhat broader view, social order is security
  of property (against theft and damage at the
  hands of other individuals) as well as of
  persons. This is the order Hobbes was concerned
  with and he called it Peace. (emphasis not in
  the original) refraining from certain behaviors.

81

• Implication order in the above sense is a
  quantitative variable that is dichotomized. Why
  not distinguish degrees of order more or less
  orderliness?

What is the PROBLEM of social order?
(1) It could mean in general how does social
order originate?
(2) In regard to the state of nature (war of all
against all) it could mean

• How does order emerge in such a state of
  nature? (This is an explanatory question)
• How could a war of all against all be avoided?
  (This is a technological question not identical
  with explanation of order)

82
Does Hobbes address the EXPLANATION of norms and
institutions?
Transfer of (or assignment of) rights to a
central authority is addressed this is a change
of norms (rights kind of norms). Thus Citizens
are no longer allowed to do certain things such
as taking the law in ones own hand the state
sets restraints. Not clear what exactly these
norms are.
A central authority exists, and there are norms
specifying what the members of this authority are
allowed and not allowed to do. But these norms
are not addressed in detail!
Thus explaining of the emergence of a state is
equivalent to explaining norms. But what these
norms are is not specified.
83
If cooperation emerges then a regular behavior is
explained. This is one meaning of norm.
Sanctioning (of the state only?) is included in
Hobbess model as well this is another
dimension of the norm concept.
Thus, Hobbes addresses several dimensions of the
norms concept, but he does not provide detailed,
informative propositions.
Question Is it possible to explain the kind of
norms does Hobbess model imply certain
propositions?
Hobbes would probably predict that those norms
will be issued that are expected to solve the
problem of social order. Question what happens
if there are different expectations how are
they reconciled?
84
What can we learn from Hobbes about the emergence
of norms and institutions?

• The state of nature is a situation with strong
  externalities. This suggests that a central
  condition for the origin of norms and
  institutions might be externalities. As we will
  see later, this is a central variable in all
  theories of norm emergence.
• However, if there are externalities, there could
  be different reactions by those affected. It is
  not clear what exactly happens if there are
  externalities. Thus, the question is if there
  are externalities, what are the conditions for
  the emergence of which kind of norms?
• Hobbes applies a general background theory
  individual preferences like egoism, constraints
  like the behavior of others and maximization of
  utility are conditions for the emergence of
  norms. This theory could be applied to explain
  the origin and effects of norms in general.

85
V.2. The Normative Solution Talcott Parsons
The so-called normative solution is advanced
especially by Talcott Parsons. Here are some
quotations

• Essentially that men being reasonable ought
  to, and in general will in pursuit of their ends
  subordinate their actions, whatever these may be,
  to certain rules. The essential content of these
  rules is to respect the natural rights of others,
  to refrain from injuring them (Structure, 96).
• the basic condition on which an interaction
  system can be stabilized is for the interests of
  the actors to be bound to conformity with a
  shared system of value orientation standards
  (The Social System, 38)
• without the attachment to the constitutive
  common values the collectivity tends to
  dissolve (The Social System, 41)

86
Question What are the major problems of this
solution?

• There is no theory that explains when the
  required norms will emerge.
• The normative solution could be interpreted as a
  technological statement if there are norms then
  there will be peace.
• BUT is this technological statement correct?
• In general, norms are not sufficient to guarantee
  that the regulated behavior is performed.
• Not clear what the other conditions for
  conformity are.
• The free rider problem is not addressed it is
  well known (at least since M. Olsons theory of
  collective action) that especially in a large
  group reasonable behavior is not carried out
  (i.e. doing something to realize goals cannot
  simply be taken for granted). See as an example
  environmental pollution.

87
V.3. The General Mechanism of Norm Emergence in
Hobbess Leviathan the Transfer and
Centralization of Rights
We can describe the process (or mechanism)
outlined by Hobbes that leads to social order in
a general way individuals transfer rights to a
collectivity (or to a group of other
individuals). These individuals then make rules
and enforce them.
This happens not only in a state of nature but in
real societies as well. Examples?

• Foundation of clubs (e.g. a chess or tennis
  club), associations etc. (see the Vereinsrecht
  law of associations)
• informal groups friends may develop a division
  of labor so that certain tasks and rights
  organizing weekend trips or dinners in
  restaurants are transferred and centralized
• formal organizations like firms employees
  transfer rights to the firm (i.e. to managers,
  the board etc.).

88
Question Under what conditions will this
mechanism be chosen? Some brainstorming ..
Transfer of rights may occur spontaneously step
by step a member of a friendship group offers to
organize the next visit to a restaurant. The
other friends like the choice and assign the
task alternately. Again, this works well and is
repeated.
89
Conditions for voluntary transmission to a
central authority (spontaneously or by design)

• There exists a central authority (a group or an
  individual) rights may be transferred to.
• The central authority fulfills the needs of the
  members, and it is expected that this will happen
  in the future.
• Expected costs of failure are relatively low,
  compared to the benefits of the transfer of
  rights.
• Perceived alternatives to transferring rights are
  regarded as more costly, such as voting on
  restaurant proposals or voting on every decision
  in a sports club. These are transaction costs.

Note perception of alternative institutions or
norms is important see the creation of
constitutions such as the US constitution not
all alternatives were known.
90

• There are political entrepreneurs individuals
  who bear the costs of setting up the organization
  and run it. This is a condition for founding a
  central authority if there is none and for
  initiatives to transfer rights.
• Those kinds of rights will be transferred that
  are least likely to hurt members e.g. one will
  usually not transfer the right to use all of
  ones assets! However, this happens sometimes

Parliaments sometimes renounce their
constitutional rights. See the enabling law
(Ermächtigungsgesetz) under Hitler-Germany in
1933. (see http//en.wikipedia.org/wiki/Enabling_a
ct and the article in the German version). See in
particular Ivan Ermakoff. 2008 Ruling Oneself
Out. A Theory of Collective Abdications. Durham
Duke University Press.
91
Concluding note
This is the first time we ask the general
question
What are the conditions for a given mechanism of
norm emergence? The different theories or
mechanisms like the one by Hobbes outline
mechanisms but they never state the conditions
that are relevant for the origin of this and not
of alternative processes!
We will ask this question for each mechanism!
Another question will be what exactly are the
differences between the different mechanisms?
92
VI. Peaceful Cooperation Without a State
Spontaneous Norm Emergence in the State of Nature
and Beyond
An important extension of Hobbess approach is
Robert Axelrods The Evolution of Cooperation.
The starting point is that the state of nature is
a prisoners dilemma.
I will first explain what this means and then
turn to Axelrods theory.
93
VI.1. The State of Nature as a Prisoners Dilemma
(PD)
What is a prisoners dilemma? Here is the
original story
Behavioral alternatives of prisoner A Behavioral alternatives of prisoner B Not confess Confess Behavioral alternatives of prisoner B Not confess Confess
Not confess 1 year prison / 1 year prison 10 years prison / Free
Confess Free / 10 years prison 8 years prison / 8 years prison
First entry is payoff for A, second is payoff
for B
94
The general structure of the situation is the
following
Behavioral alternatives of person A Behavioral alternatives of person B Cooperate ( C ) Defect (D) Behavioral alternatives of person B Cooperate ( C ) Defect (D)
Cooperate (not confess) 3, 3 1, 4
Defect (confess) 4, 1 2, 2
A graph that again depicts the structure of the
payoffs
Lower payoff 3
Lowest payoff 1
Highest payoff 4
Lower payoff 2
95
What has all this to do with Hobbess problem of
social order?
Here is the PD matrix again
Behavioral alternatives of person A Behavioral alternatives of person B Cooperate (C) Defect (D) Behavioral alternatives of person B Cooperate (C) Defect (D)
Cooperate (not confess) 3, 3 1, 4
Defect (confess) 4, 1 2, 2

• Defect war
• Cooperate peace (any kind of making a contract,
  trust etc.)
• Warfare is the dominant alternative for everybody

96
Lets look again at the PD
Behavioral alternatives of person A Behavioral alternatives of person B Cooperate (C) Defect (D) Behavioral alternatives of person B Cooperate (C) Defect (D)
Cooperate (not confess) 3, 3 1, 4
Defect (confess) 4, 1 2, 2
Assume now the game is not played once, it is
repeated (iterated). Does this change the
situation? Example You need daily a yoghurt, and
a shop owner puts the yoghurt at a certain place
in front of your door you promise to bring the
money to the shop after work. Questions (1)
There is a one-time purchase, (2) you buy every
day. Would it pay to cheat?
97
Assume there is cooperation for some days both
get 3. Then you defect at day 10, whereas the
shop owner gives you the yoghurt (i.e. he
cooperates) You earn 1 additional point (4
instead of 3). But then the shop owner will
defect as well and your payoff is 2. Does it pay
to cooperate? If you cooperate say 5
days, you get five times 3 15. If you defect
immediately you get 4 (the first defection on
day 1) (4 x 2) (defection for four days), i.e.
a payoff of 12. Thus, in repeated PDs defection
is not always profitable.
98
Question Can you imagine situations where even
in an iterated PD defection pays?
(1) The payoff for defection is huge, compared to
cooperation (not 4, but say 4000). AND you
need the money
2, 2 0, 4000 4000, 0 1, 1
(2) The play is repeated only a finite number of
times, i.e. it is a finite game.
(3) You think that the partner is not trustworthy
(he will not sustain cooperation you heard bad
rumours about him).
Thus, in repeated PDs cooperation is not always
profitable.
99
VI.2. How Can Cooperation Emerge in a Prisoners
Dilemma Situation? The Theory of Robert Axelrod
The basic idea in Robert Axelrods theory (for
references see at the end of this section) is if
the prisoners dilemma is repeated, cooperation
will be likely under certain conditions.
Note how Axelrod formulates the problem at the
beginning of his book The Evolution of
Cooperation (1984 3)
Under what conditions will cooperation emerge in
a world of egoists without central authority?
What is the difference to Hobbes problem?
100
There is no Nirwana approach who is
interested in a utopian situation as Hobbes
describes it? Axelrod assumes egoistic actors,
there are no internalized norms and there is no
enforcement agency. This is the case in many real
situations so that Axelrods theory can be
applied.
Axelrods procedure
The question of how cooperation can emerge in a
world of egoists can be formulated in the
following way
What strategy of the actors is best to
generate cooperation in the long run?
For example one strategy could be if you meet
somebody, choose C and D by chance. Other
possibilities
101

• Play always C.
• Play always D.
• Begin with C and then by chance choose C and D.
• Begin always with C and then do what the other
  player did on the previous move. This is TIT FOR
  TAT.
• Begin with C if the opponent defects twice then
  defect as well TIT FOR TWO TATs.
• Play C as long as the other player plays C. Then
  play always D.

The payoffs of each of the two players can be
computed, based on the game matrix, e.g.
3, 3 1, 4 4, 1 2, 2
102
To answer the question of what strategy yields
the best results Axelrod invited game theorists
to a computer tournament and to submit programs.
The program that won was submitted by Anatol
Rapoport TIT FOR TAT

• choose C on the first move,
• imitate the player on the previous move. Thus, as
  long as the other player chooses C, play C. If
  he/she chooses D, choose D etc.

Then a second tournament was organized. Again,
TIT FOR TAT made it it was the strategy that in
general was most success- ful, i.e. it yielded
the best results in many different situations
(i.e. when it meets many different strategies).
103
Axelrods principles and theory
Based on the results of the tournament, Axelrod
formulates some principles for a successful
strategy, i.e. a strategy that yields
cooperation with a high probability.

• Be NICE, i.e. never defect before an opponent
  does. This implies always begin with C.
• Be FORGIVING, i.e. defect only once and never
  hold a grudge once you have vented your anger.
  Thus, be not too harsh.
• Be PROVOCABLE, i.e. one should get mad
  immediately after the defection of an opponent
  and not, e.g., continue with C several times.
• Be CLEAR, i.e. your opponent should recognize
  that you play C or D (clarity is often absent in
  situations with many players e.g. if you save
  energy nobody will probably notice it.)

104
Implication
Cooperation is possible without a central
authority!!!
105
Axelrod refers to pairs of players, i.e. dyads
(two-person game theory). The theory could thus
be formulated more generally
If there are two players who interact

• if a player never defects first (nice),
• if a player defects only once after the defection
  of the opponent (forgiving),
• if a player defects immediately after the
  defection of the opponent (provocable),
• if the opponent perceives the reaction of a
  player,

then cooperation emerges between the players.
Consequence if cooperation is to be predicted in
a large group, the overall cooperation rate
depends on what happens among dyads. (This is
actually how Axelrod proceeds!)
106
An even more general formulation could read
If there is a group of players

• the more often a player in an encounter with
  another player begins with C and defects only
  once after the defection of the o