International Journal of Sociology and Social Policy

THE MEANING OF SOCIAL POLICY AND SOCIAL STRUCTURE

W.L. Grichting
Article information:
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

To cite this document:

W.L. Grichting, (1984),"THE MEANING OF SOCIAL POLICY AND SOCIAL
STRUCTURE", International Journal of Sociology and Social Policy, Vol. 4 Iss 4 pp.
16 - 37
Permanent link to this document:
http://dx.doi.org/10.1108/eb012971
Downloaded on: 30 January 2016, At: 20:37 (PT)
References: this document contains references to 0 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 227 times since 2006*
Users who downloaded this article also downloaded:
Sally Bould, (1984),"DEVELOPMENT AND THE FAMILY: THIRD WORLD WOMEN
AND INEQUALITY", International Journal of Sociology and Social Policy, Vol. 4 Iss 4
pp. 38-51 http://dx.doi.org/10.1108/eb012972
Gil Richard Musolf, (2003),"Social structure, human agency, and social policy",
International Journal of Sociology and Social Policy, Vol. 23 Iss 6/7 pp. 1-12 http://
dx.doi.org/10.1108/01443330310790570

Access to this document was granted through an Emerald subscription provided by

emerald-srm:545847 []
For Authors
If you would like to write for this, or any other Emerald publication, then please
use our Emerald for Authors service information about how to choose which
publication to write for and submission guidelines are available for all. Please visit
www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of
society. The company manages a portfolio of more than 290 journals and over
 2,350 books and book series volumes, as well as providing an extensive range of
online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a
partner of the Committee on Publication Ethics (COPE) and also works with Portico
and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct

at time of download.
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)
 THE MEANING OF SOCIAL POLICY AND SOCIAL STRUCTURE

W.L. Grichting, Department of Behavioural Sciences,

James Cook University of North Queensland.

During the past decade we have witnessed renewed interest in the study and analysis
of both social policy and social structure. Journalists, administrators, politicians,
welfare personnel and social scientists have extensively been engaged in the interpre-
tation and assessment of the impact of social policy on social structure and vice versa.
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

It would appear that the return of liberal governments in the United States, the
United Kingdom and Australia had much to do with this time-honoured revival. In
the United States this activity can be largely attributed to the promise of the "Great
Society" - an attempt by the Johnson administration (1963-1968) to eliminate pov-
erty, create employment, improve medical care and education and revitalise cities
and rural areas. In the United Kingdom the Labour Governments of Wilson (1964-
1970; 1974-1976) and Callaghan (1976-1979) held out similar promises for a more
equitable society. Predictably the literature on social policy and social structure
increased accordingly.

In Australia, a similar connection may be observed: the sudden increase in articles,

monographs and books on social policy and social structure in this country revolves
around the return of a Labour Government in 1972. With the arrival of the new
Government ambitious and far-reaching changes in almost all areas of social legisla-
tion were to usher in a new era of greater equality, enhanced civil liberties and greater
community participation in decision-making. Less than a year after Labour came to
power the Priorities Review Staff detailed the prevailing concerns of the new Govern-
ment. They were:

-the degree of equality between various members of the community;

-the quality of Life;
-the stable growth of prosperity;
-the preservation and expansion of civil rights. (Priorities Review Staff, 1973. p.2.)

In general, incremental social change endorsed by the previous Liberal Govern-

ments was to give way to a societal overhaul. In particular, the residual redistribution
welfare model was to be replaced by the institutional redistribution welfare model.
Titmuss (1974) equates the residual welfare model of social policy with the notion
that the family and private market are the natural and primary agencies responsible
for meeting individual needs. Consequently social welfare institutions in this model
are relegated to a subsidiary position and limited to temporary intervention. The
residual welfare model is thus a good example of the subsidiarity principle, a concept
extensively used by European social philosophers. Peacock (1960) summarises the
residual model when he states "the true object of the Welfare State is to teach people
how to do without it" (p.11). The institutional redistribution model, on the other
hand, views the provision of social welfare as a central institution in society. It largely
bases the welfare notion on the quality of life, frequently irrespective of an expressed
need by a potential beneficiary of such a service.

16
 The implicit identification of social policy with welfare politics may not have
been intended by the architects of the new Fabian Society. Nevertheless most
Australian theoretical treatises and research publications quite explicitly equate
social policy with a truncated notion of welfare politics and oftentimes narrow the
broad field of social policy to the small albeit important topics of wealth and income
redistribution. In so doing they largely emulate their British and American
colleagues who were equally explicit in limiting the study of social policy to the
examination of distributive justice. Thus T.H. Marshall (1965), the doyen of British
social policy spokesmen, defines social policy as the "policy of governments with
regard to action having a direct impact on the welfare of citizens, by providing them
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

with services or income. The central core consists, therefore, of social insurance,
public (or national) assistance, the health and welfare services, and housing policy"
(p.7). Martin Rein (1970), the American doppelgaenger of T.H. Marshall, views
social policy analysis in a similar vein when he reduces social policy to the planning
for social externalities, and the equitable distribution of social benefits, especially
social services. Graycar (1977), on the other hand, offers a more encompassing
definition of social policy when he first defines it as the result "of a desire to pursue
social justice, to achieve equality, to effect redistribution or to ensure the efficient
functioning of society" (p.3 - emphasis added). In his more recent work, however, he
too suggests that social policy is mainly to be equated with welfare politics (1978,
1979). More to the point yet: a national conference on social policy in the 1980's
(Canberra, 1982) almost exclusively featured speakers who took it for granted that
social policy is to be equated with welfare distribution and administration.

Despite our assertion of a temporal and conceptual affinity between the notions
of social policy and social structure a search of the literature reveals a peculiar
hiatus between these two concepts. The gap is two-fold: first, definitions of social
policy typically do not make explicit reference to social structure; second, computer
searches of Books in Print, Sociological Absracts and Psychological Information
Databases produced 1,245 entries on social policy, 4,300 entries on social structure,
and a grand total of 26 entries where both concepts were indexed together. Yet
publications on social policy and social structure wax and wane together as Figure 1
demonstrates.

Figure 1 presents the bi-ennial count of entries of social policy and social structure
as a percentage of all entries in Sociological Abstracts from 1964 to 1981. Two
features stand out: the seventies are characterised by a continuous increase in the
relative importance of studies in social policy and social structure as publication
material. In addition, the two are highly correlated over time (Pearson's r = +.73).
How can these divergent observations be reconciled? Two explanations come to
mind: first, there is little agreement concerning the definitions of both social policy
and social structure among the many writers on these topics. Second, the concept of
social policy has disproportionately often been limited to the study of a truncated
notion of the social welfare system. Consequently it appeared irrelevant to many
aspects of social and societal structure. As a result, the concepts of social policy and
social structure are only rarely linked explicitly in most publications (hence the few
entries which index both). However, the two concepts are intrinsically ordained
toward each other. This may account for the simultaneous growth and decline of
publications in both areas. At this point we are not prepared to offer an

17
 Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

18
 interpretation of the most recent decrease in publishing activity although it could be
suggested that it mirrors the disenchantment of social engineers because of the
economic crisis and the victory of conservative governments in Australia, the United
Kingdom and the United States.

The purpose of this article is to broaden and specify the concepts of social policy
and social structure. This is attempted by providing definitions of social policy and
social structure and highlighting how these two concepts are interrelated. The
emphasis rests on the specification and elaboration of the component parts of these
two crucial concepts.
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

Defining Social Policy

In our view, limiting social policy to welfare is extremely unfortunate and totally
unacceptable. It is unfortunate in as much as it deprives the study of social policy of
the broader context of social justice and societal interaction. By restricting social
policy to welfare administration one perpetuates the practice of excluding public
subsidies to the middle class and the rich from discussions of the social welfare
system. For instance, social policy analysis restricted to welfare administration
would typically fail to consider such issues as free tertiary education for the middle
class and rich in Australia. University fees were abolished in 1974 in an attempt to
enable the poor and underprivileged to attend tertiary studies. Little has been
accomplished except that the poor now finance the education of the well-to-do.
(See Anderson and Vervoorn (1983).) It is apposite to note that the university
graduate commands an above average salary after four years (each of 30 weeks) of
study which is not only tuition-free but also subsidised for everyday living expenses.
This subsidy amounts to about 3,000 per annum. A similar situation prevails
concerning the construction of superhighways through most American cities for the
convenience of the relatively well-to-do suburbanite irrespective of the cost to the
dwellers in the inner city. The inner city predictably turned into slum, partly
because of a federal policy not to grant any housing assistance for these less
desirable living quarters. The reduced scope of social policy as the distribution of
welfare benefits in the traditional sense not only dismisses justice as the relevant
criterion but also fails to consider many of the less obvious welfare payments to
privileged classes.

To limit social policy to welfare is unacceptable because it loses sight of the

precarious balance between social security and civil liberty. If there is still some
truth to the old adage that "the purpose of social action is not the power of the state
but the welfare of the people" then social analysts have every reason to suspect a
Trojan horse in the populist proliferation of social welfare services and to remind
us wryly that it will take an awful lot of government intervention to restore civil
liberty.

For these reasons we prefer to view the study of social policy as the critical
examination of the planning and implementation of social justice. By social justice
we refer to the moral imperative that equals be treated equally and unequals
according to their relevant inequalities. Social justice thus implies equitable rather

19
 than equal treatment. It also puts the onus of proof for justifying differential
treatment on whoever wants to make distinctions. This line of reasoning stems
directly from Plato's Republic and Aristotle's Nichomachean Ethics.

The relationship between equality and equity is complex. Is it possible to have

equitable inequalities? If so, are they inequalities of consideration, opportunity,
provision or outcome? May they be enduring or only ephemeral? How many
people must be affected by a certain inequality before it is seen as inequitable? -
There is little doubt that certain inequalities are perceived as equitable - such as in
competitive sport and championship. Few would presumably suggest that
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

competing teams be reshuffled to ensure that a "fair" play will result. Thus the
concept of "fairness" in sports - "that's not fair," "that's not cricket," "fair go ya
mug" - does not refer to the redistribution of basic team assets but rather to equal
treatment of all players according to the rules of the game. It is noteworthy that in
sports, rules precede fairness; similarly, in social policy, rights precede justice - at
least according to Thomas Aquinas (Summa Contra Gentes, 2,28). In addition fair
rules vary from one sport to the next. Should one expect rules of fairness and equity
to vary from one policy area to the next? Still, when does inequality become
inequitable? We suggest that inequality is inequitable in terms of social policy
formulation whenever the following two conditions apply:

-inequality affects a significant number of people (or a number of significant

people?). It should be remembered that hard cases arc apt to introduce bad law
(Rolfe in Winterbottom V Wright (1842)). Such a stipulation may sound harsh and
somewhat inequitable. Its justification depends on the clarification of the relative
size and centrality of the inequality under discussion as well as on the definition of
what constitutes a "significant" number. Clearly the two concepts are negatively
related: the larger the inequality the smaller the number required to constitute a
significant group. In general, a number is deemed significant whenever the
inequality is caused by or causes a structural misalignment in society. In other
words, inequalities based on personal choices and preferences require remedies
other than through social policy enactment.

-it is believed that something can be done about this perceived inequality.
Without the potential for improvement, social policy efforts remain doomed to
failure. This implicitly challenges the profoundly held article of faith in social
administration, namely, that all social problems are amenable to solution, or at
least to amelioration (Pinker, 1982).

Given that equity and justice have different denotations in different policy areas,
it is helpful to differentiate among three types of social policy as Thomas Aquinas
does in his discussion of justice (Summa Theologiae, II, II). Aquinas notes two
essential characteristics of justice. First, justice always deals with a relationship
between at least two actors and, second, the distinguishing mark of justice is that
some debt is to be paid. The nature of the debt to be paid varies depending on the
actors engaged in the relationship under discussion. On this basis Aquinas suggests
three types of social justice as diagrammed in Figure 2.

20
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

Commutative or exchange justice (also known as restitutive justice) obtains

between two individuals or aggregates of individuals. Buying and selling of goods
and services, that is, the market economy, falls under the principles of
commutative justice. Commutative justice can roughly be equated with the
area governed by civil law. Civil law is frequently characterised by the assertion
that the state or community at large does not have a direct interest in the
transactions between two individuals. To the degree that this is so. commutative
justice is radically different from both legal (or restributive) and distributive
justice. In the latter two situations the exchange occurs between the social whole
and an individual in as much as the latter belongs to that social whole. In other
words, restributive and distributive justice forms are qualitatively different from
commutative justice because the individual as part of the whole is one of the actors
in the transaction relationship. Philosophers disagree whether this differentiation is
based on an ontological difference or is simply the result of a convenient
epistemological distinction. Be this as it may. In either case it is a helpful analytical
juxtaposition and extensively honoured in jurisprudence. One thing is clear; there
cannot be distributive justice without restributive justice and vice versa. The ability
of the social whole to meet the expectations of distributive justice is predicated on
its success in implementing restributive justice in its broadest denotation.
Distributive justice is frequently equated with social welfare; restributive justice,
on the other hand, may be said to prevail "whenever a criminal is accused and
judged in the same way by all those he has not injured as he would be by the person
to whom he has done some injury" (Plutarch, Banquet, chp. 11).

21
 The complementarity of distributive and restributive justice is acknowledged by
contemporary writers as well. Rawls (1972) in his A Theory of Justice develops two
principals of justice. His first principle links restributive (legal) justice with civil
liberty when it stipulates that "each person is to have an equal right to the most
extensive total system of equal basic liberties comparable with a similar system for
all" (p.302). Basic liberties refer to freedom of thought and conscience as well as
liberty of speech and assembly. The second principle asserts "that social and
economic inequalities are to be arranged so that they are both: (a) to the greatest
benefit of the least advantaged, consistent with the just savings principle, and (b)
attached to offices and positions open to all under conditions of fair equality of
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

opportunity" (p.302). In the second principle one may note an overlap between
distributive justice and social security. Brennan's (1983) incisive comment on
Rawls' two principles further links social security and distributive justice when he
offers a first preliminary principle: "Each person is to have the right to necessaries
of life, including food, shelter, and other things required for survival, and the right
to security so those necessaries may be retained and used" (p.203). Brennan's
second preliminary principle connects restributive justice with civil liberty when it
states that "every person is to have the right to live under a just (or nearly just)
constitution. If such a constitution is to endure, it must be assured of a background
of relative stability and of substantial compliance by all citizens with their natural
duties and political and moral obligations" (203).

It would be tempting to review the history of equality as an ideal but that is

clearly beyond the scope of the present exercise. Still we may point out that such a
review would essentially identify two major periods: the time before and after the
Enlightenment. Thinkers in the first era stressed the equality of all men (Plato's
political equality of the sexes, Aristotle's concern for all free citizens, The New
Testament's emphasis on the equality of all souls in the sight of God (Gal. 3:26-29))
but capitulated before the existing structure of social inequalities as illustrated by
their general endorsement of slavery. Philosophers since the Enlightenment
(Hobbes, Locke, Condillac, Helvétius, Rousseau, Kant to mention but a few) not
only endorsed and elaborated on the principle of equality but also argued the case
in the political arena. However, egalitarianism as conceived by these more recent
thinkers does not so much assert equality as deny the justice of some existing
inequality of treatment on the basis of some allegedly irrelevant differences of
quality or circumstance. From this vantage point the distinction between equality
and equity begins to fade into irrelevance. (For a detailed discussion of the concept
of equity and justice in social behaviour see Greenberg and Cohen, 1982).

Still the concept of equity has come under fierce attack particularly by the radical
left because equity implies that - depending on the policy area - equitable treatment
may not be equal and vice versa. Stilwell (1980) in a rhetorical outburst dismisses
the notion of equity as part of the lexicon of "snake oil merchants who try to justify
the cutbacks in social welfare and the widening of recession." Our understanding of
social welfare suggests that concern for equity rather than equality permits
extensive and effective redress of social injustice. For instance, the argument for
progressive taxation where the individual capacity to pay is taken as a ground for
discrimination presupposes a decision in favour of equity rather than equality. Lest
there be a misunderstanding, equity subsumes and treats as a special case the
22
 concern for equality. This is already implied by our definition of justice. It also may
explain why a good deal of the traditional equality literature effectively amounts to
a proposition for equitable rather than equal treatment.

Substituting equality for equity rarely amounts to treating everyone alike;

typically it means getting rid of one system of distinctions and replacing it with
another. R.H. Tawney explicitly criticises this substitution when he observes that
"the more anxiously a society endeavours to secure equality of consideration for all
its members, the greater will be the differentiation of treatment which, when once
the common human needs have been met, it accords to the special needs of
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

different groups and individuals among them" (1952, p.39). Pinker (1982) is
equally adamant when he observes that "radical change in social policy does not
abolish the criteria by which we allocate welfare goods and services; it simply
changes them." He is yet more challenging when he suggests that "social workers
today may identify themselves with the minority of deprived citizens who are
victims of change in the mixed capitalist society, but if processes of radical change
eventually give rise to a more egalitarian society, they are likely to find themselves
ministering to the needs of a dispossessed and alienated bourgeoisie" (p. 13).
Clearly poverty is not overcome by abolishing wealth. Finally, affirmative action
programmes typically rest on the concept of equity rather than equality. There
would be no room for reverse or positive discrimination (most welfare program-
mes) if treatment were to be based on equality rather than equity.

The concepts of social justice and equity point to the need for the study of social
change and invite a broader definition of social policy. Peter Townsend's (1975)
definition of social policy is sufficiently broad and specific for our purpose. He
defines social policy as those mechanisms and processes "whereby societies
prevent, postpone, introduce and manage changes in structure" (p.2). This
definition is broad enough to permit the study and examination of all three sides of
the justice triangle. Still we wish to extend this definition by substituting "social
systems" for "societies" in order to include both social organisations and entire
societies in our analysis. In so doing we do not subscribe to the distinction between
social and societal policy. Ferge (1978) defines social policy as "handling essentially
central redistribution" and societal policy as "aiming deliberately at the trans-
formation of structural social relations in accordance with socialist aims" (p.82).
We imply that any redistribution results in a realignment of social structure and vice
versa.

Our notion of social policy is not limited to the traditional concerns of welfare
politics (health, education, housing and income maintenance). It also includes the
frequently neglected areas of civil liberty (freedom of thought, conscience, speech
and assembly as well as the right to privacy, property and protection). Further
social control, that is, problems of system boundary permeability (migration and
international trade policies) and system regulation (e.g. civil duties, control of
substance use, gambling and fair trade practices) are core areas of social policy.
Our intention is not to deal with any of these social policy areas in detail but rather
to broaden our definition of social policy and to alert the reader to the fact that
modification of any policy issue is likely to have implications for other policy areas.

23
 Our definition of social policy also implies that social policy may be based on
social action or inaction, that is, changes in structure may be brought about by
commission as well as omission. Laissez faire is therefore not viewed as the absence
of policy (as the literal meaning suggests) but rather as the endorsement of a
specific policy model. Similarly, mechanisms which hinder rather than help
structural change are core elements in the study of social policy. As the proposed
definition leaves empirically and ideologically problematic the desirability of
structural change, it permits detached examination of the startling situation which
frequently results when social policy cross-cuts ideology. Ideology, of course,
results whenever we mistakenly take our preference for a given theory as the final
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

truth. Despite all our commitments to the creation of a more equitable and just
world our theories should be treated as conjectures and our methods of scientific
enquiry should consist of ingenious and severe attempts to refute them (Popper,
1981, pp.80-81). As a result it is "vitally important for the subject of social policy
and administration to remain answerable to the academic community, notwith-
standing its association with politics and policy-making" (Pinker, 1982, p.8).

Our definition is also specific in as much as it relates social policy to changes in

social structure. Two points are important here. First, change denotes a difference
across either space or time or both. Second, this change is a change in structure and
therefore requires a concise definition of social structure.

Differences, by definition, are identifiable and measurable. That is to say,

differences which cannot be identified either at the level of causality or
consequence are not differences. To the degree that they can be identified they can
also be measured, by which we mean that we are able to assign symbols to aspects
of objects according to specified rules. This definition rejects the notion that
measurement equals quantification. Rather, measurement as understood here,
refers to the examination of differences in both quantity and quality. While
traditional analysis - particularly in the context of economic and social indicators -
has greatly focused on quantitative differences, contemporary social science
analysis permits the study of qualitative differences. Loglinear and logit analyses as
well as a plethora of multidimensional scaling procedures come readily to mind as
one attempts to do justice to the analysis and interpretation of qualitative or
categorical data.

Denning Social Structure

The concept of social structure is as ubiquitous as it is vague. Many writers take it

for granted that there is an agreed upon definition of social structure or that the
readers will intuitively understand the meaning and significance of this crucial
concept. However, even a perfunctory review of the many denotations and
connotations of this central notion make perfectly clear that it means whatever it
happens to mean in the context of other ill-defined parameters under discussion.

A casual perusal of the literature suggests at least four increasingly complex

definitions of social structure. They all have a common denominator in as much as
they rest on the etymological concept of the word "structure". The word is derived

24
 from the Latin struere meaning "to heap up," "to build." Thus "something" is used
as building blocks to produce a "structure". What are these building blocks? The
authors diverge. The first group treats as structural building blocks the relative
frequencies in each of several categories of certain variables of interest. Thus
demographers speak of the occupational structure of a society and refer to the
proportion of the labour force found in various occupational slots, e.g.,
professionals, graziers, clerical, operatives, etc. Population pyramids - often
referred to as graphic displays of population structures - rest on a similar notion of
categorical frequencies of the age and sex distributions. Usually little effort is made
to show how these categories relate to each other and how these relationships affect
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

social interaction.

A second type of building blocks for structural analysis is individual elements in

groups, organisations and society at large. Treating individual elements as the
building blocks of social structure tends to result in a reductionistic approach to the
study of social structure. Here social structures are little more than the sum total of
the individual components. As a result the emergent qualities of social structure are
frequently ignored. Social policy studies based on this definition of social structure
are frequently myopic and focus disproportionately on the fleeting well-being of the
here and now. To the degree that social actors are substituted for individual
elements the problem is somewhat remedied if we adopt Weber's notion that social
actors take into account other actors' behaviour.

Another kind of building blocks of social structure is suggested by cultural

anthropology. Anthropologists who define culture by emphasising the pattern or
organisation of meaning elements, habits, customs and responses (Willey, 1929;
Coutu, 1949; Gillin, 1948) overcome the nominalistic approach to the study of
social structure but separate the building blocks from their concrete carriers or
actors. Culture and structure thus become conceptual models or "ideal types" and
are removed from actual behaviour in social interaction. This approach renders
difficult the observation and measurement of social change as it abstracts and
thereby removes the building blocks from the overt, observable uniformities of
behaviour.

A fourth kind of building block for structural analysis combines the building
blocks of the second and third approaches and focuses on relationships of social
actors. Social structure is thus defined as a pervasive pattern or network of
relationships among social positions. This is the definition of social structure most
suitable for the study of social policy. The proposed definition emphasises both the
social and concrete dimensions of the concept "structure". Concreteness is assured
because the focus is on actual relationships rather than on subjective meaning
elements; and the social aspect is preserved because the concern is with positions
rather than with incumbents of positions or individual elements. The distinction
between positions and incumbents is important in this context because a given
incumbent may occupy several positions, that is, play various roles. Our definition
of social structure has been suggested by Laumann and Pappi (1976) who
specifically identify the unit of structural analysis as the relationship between any
pair of actors. The concept of social actor leaves as empirically problematic the
level of analysis and suggests that individuals, groups, organisations and societies

25
 may be used as the end points (or nodes) of a specific relationship. Given our notion
that social policy result from commission or omission it is evident that the absence of a
relationship between two actors is as important as its presence.

One can further highlight the importance of the presence or absence of a

relationship between different actors by differentiating among various types of
relationships which are deemed appropriate in a social system. In order to assess
social policy in the context of social structure as a pattern of relationships among
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

social positions it is necessary to identify the actually observed relationships and to

compare this set against the set of normatively expected relationships. For instance,
the policy analyst may have to ask whether personal entrepreneurship of public
servants is predicated on a set of conflicting relationships which require more
extensive scrutiny than what traditional legislation provides. For instance, should
members of public decision-making bodies be required to make public their private
interests in economic ventures? In a recent study, Grichting (1983) found that a
significant majority (58 percent) felt that financial interest of decision makers are
incompatible with the possession of regulatory power of an official nature unless
such interests are made public. Public knowledge alone may not sufficiently protect
the commonweal. The commonweal may require officials to put in trust private
interests as a condition for public office. Specifically stated, should a minister for
main roads be able to operate a large quarry or a minister for gambling be
permitted to own a major race track? (See Queensland Politics in 1984 which are
rather tellingly indicted by Brennan's (1983) study, entitled 'Too much order with too
little law.') The above examples suggest potential conflict of interest. Such
potential conflict is pervasive in any social system and its prevention and resolution
require a hard-nosed analysis of actual and ideal relationships among social actors,
that is, of social policy as a modifier of social structure.

Perhaps the most comprehensive scheme for classifying various relationships of

social actors may be found in Parsons' (1937) pattern variables. This scheme
resulted from the critique of Toennies' Gemeinschaft-Gesellschaft dichotomy and its
treatment as an ideal type. Parsons' desire to be functionally more specific than
Toennies and to be able to compare vastly different social systems in terms of his
action frame of reference resulted in the formulation of six independent
dichotomies, two of which were subsequently removed from the presumably
exhaustive scheme. (The latter two were: collectivity versus self-orientation and
short-run versus long-run focus of valuation (Parsons, 1953).)

Parsons claims that the remaining four dichotomies constitute an exhaustive

classificatory scheme for defining normative expectations about social relationships.
Little or nothing is implied about actual role performance. Neither is there any
notion about the type of normative expectations until the definition of the situation
has been tied to a specific frame of reference. By so doing, Parsons implies that
different systems have different role expectations concerning the interaction of any
two social actors (be they individuals or collectivities). For example, business
transactions in the East and the West are defined differently; the meaning and
bestowal of citizenship varies greatly from country to country and the student-

26
 teacher interaction is very much subject to situational definition by different action
systems. More than that, Parsons observes that the role expectations of two actors
may not be symmetrical. Thus the patient may expect his physician to provide
emotional nurturance while the professional orientation of the physician suggests
emotional detachment. Parsons' concern then is to provide normative templates of
abstractly defined social relationships which permit the researcher to compare them
against the observed reality and to specify the degree to which the latter
approximates the former. The nomenclature of the pattern variables implies
dichotomies. However there is no apparent reason why the pattern variables could
not be measured in a more refined fashion, particularly since the underlying
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

construct of each of the four variables is clearly of a continuous nature.

A frequent difficulty with the pattern variables is their apparent resistance to

adequate operationalisation. However since they are relationally and hence
structurally extremely fertile constructs they merit more than passing mention in
this context. The pattern variables are grouped into two sets: those dealing with the
actor's attitudinal orientation toward social objects and those pertaining to modes
of categorising social objects. The actor's attitudinal orientations are defined in
terms of his affectivity versus affective neutrality toward another actor, and in
terms of the degree of functional specificity versus diffuseness, i.e. the expected
scope of the relationship under discussion.

As the word indicates, a relationship is expected to be affectively neutral if there

is no legitimate room for the display of emotions in the interaction. The
relationship between customer and shop assistant would be defined as affectively
neutral, whereas the relationship between mother and child is typically seen as
affective. Since the amount of (cither positive or negative) emotion invested in a
relationship modifies both it and the resulting structure extensively, researchers can
ill afford to neglect this aspect of social interaction. Scales which attempt to
measure liking and disliking or sympathetic understanding (e.g. Bogardus scale) as
well as desired proximity or distance between actors are appropriate for measuring
affectivity and affective neutrality in the Parsonian sense. Another criterion which
Parsons suggests to measure the first pattern variable requires knowledge of
whether the actor is expected to get immediate gratification out of the interaction
or is supposed to defer it by applying some discipline. Since deferred gratification is
typically associated with task-oriented situations (i.e., instrumental rather than
consummatory activities) it may be possible to assess the degree of affective
neutrality by asking how much the interaction is supposed to be used for some
other purpose (neutrality) rather than to be enjoyed in its own right (affectivity). In
the context of social policy affectively neutral relationships are typically formal,
rational and legalistic whereas affective relationships tend to be informal and
subject to multiple interpretations. Sim (undated) in studying government grants to
voluntary organisations discusses different types of grants in terms of Parsons' first
pattern variable without making specific reference to this distinction.

The second attitudinal dimension in orientation toward social objects is the

functional specificity-diffuseness variable. This distinction measures the scope or
inclusiveness of the relationship under study. Parsons has suggested a useful
mechanism to measure the degree to which a relationship is diffuse. He introduces

27
 the notion of "burden of proof in clarifying conflict situations about role
obligations: if two actors cannot agree whether a given function is normatively
included in the role definition, the relationship is said to be specific if the actor
requesting the service must prove its legitimacy within the role expectation; it is
said to be diffuse if the actor refusing the service is expected to justify such a
refusal. For example, it is perfectly appropriate for a parent to inquire from a
teenage daughter as to her whereabouts this morning at 3 a.m. The same inquiry
from the milkman would presumably be judged inappropriate. Refusal of the
requested information in the former instance would have to be justified by the
daughter (diffuse relationsip) whereas in the latter situation any claim to an answer
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

would have to be substantiated by the milkman (specific relationship).

Using this criterion it is readily possible to determine the degree of role

specificity in a given relationship. A battery of actual or contrived task disputes
deemed relevant to the relationship should permit the investigator to ascertain role
diffuseness as well as perceived role symmetry. Since many role disputes seem to
centre about property issues of one sort or another, questions about actual and
expected access to, use and alienability of, some valued possession, promise to be
particularly helpful in assessing the degree to which a relationship is or is expected
to be functionally specific or diffuse. Answers provided to such questions could
then be entered in Figure 3 to compute the amount of role specificity in any
relationship of interest. The specificity-diffuseness distinction affects the social
policy debate whenever the issue concerns an actor's rights versus obligations.
When civil liberty is challenged specificity would tend to be the preferred modality
of the relationship whereas diffuseness may be invoked when a given actor feels
entitled to provision of social security from another (typically higher-order) actor.

The remaining two pattern variables deal with ways judged appropriate for
treating role actors. The first focuses on the universalism-particularism distinction.
It can perhaps be conceptualised by asking how appropriate it is to substitute one
social actor for another. Thus the question is whether an actor is to be judged in
terms of some universal standard or in terms of some particular frame of reference
in which ego is personally involved. For instance, bank tellers, taxi drivers and
lecturers are supposed to be chosen in terms of some universally agreed-upon
standard rather than in terms of personal consideration such as whether a given
actor happens to be a cousin, neighbour or personal acquaintance. Thus criteria
employed for the selection of such personnel permit one to determine the degree to
which a relationship is universalistic. Inasmuch as the occupancy of a given position
by a specific incumbent results in a role transformation, the role may be judged
particularistic. Alternatively, one might enquire of participants in the system
whether or not different incumbents of a specific position (e.g., teacher, student,
business partner) are treated alike. To the extent that identical behaviours of all
incumbents are reported by all (or the majority of) participants, the role may be
deemed universal. Specifically one may want to gauge the degree of particularism
by computing a weighted average of how much relationship-specific criteria are
applied in the selection or treatment of social actors. Clearly, if nepotism instead of
competence is the decisive criterion for treating social actors in normatively
universal relationships, systematic deviance results. Social policy based on equality
largely favours the universalistically defined social relationships whereas the equity

28
 argument presupposes a particularistic definition of the relationship as it permits
the differential treatment of actors on situation-specific criteria. However it should
be noted that this very distinction depends on the definition of system boundaries:
particularism tends to give way to universalism as the system grows.

Finally Parsons introduces the quality-performance or ascription-achievement

distinction in an effort to render exhaustive the pattern variables scheme. The
notion here is whether primary consideration in defining a social relationship is
given to "what people are" as against "what people do". If considerations of
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

physical attractiveness, age, sex and racial origin prevail in the selection and
treatment of social actors a relationship is termed quality-oriented; conversely, if
effort, performance and excellence are prime considerations, the relationship is
said to be achievement-oriented. Thus the person prepared to scold a coed is
quality-oriented if he addresses his insults to the lady while the person who directs
his comments mainly to the student is performance-oriented. In evaluating a
specific relationship one might draw up a list of suitable statements, perhaps of a
Likert-type, and compute an overall agreement score to compare actual rela-
tionships with each other or against a normatively defined situation. Titmuss (1974)
recognises the quality-performance distinction when he adds to his residual and
institutional redistribution models of social policy a third model: the industrial
achievement-performance model. He does not have much to say about this model,
because its assumed ideology of laissez faire is discussed in the context of the
residual model. Traditional welfare politics have little use for the industrial
achievement-performance model. This model holds that social needs should be met
on the basis of merit, work performance and productivity. In social policy analysis,
however, it is of considerable importance, particularly in the debate of civil liberty
(private entrepreneurship) versus social security (shared social responsibility).

While a given relationship can, of course, be described and evaluated in terms of

any one of the four pattern variables, the power of the scheme is best revealed by
generating multidimensional profiles. An intuitively appealing way of doing this is
to crosstabulate the four pattern variables as shown in Figure 3.

Figure 3: SYNOPSIS OF PARSONS' PATTERN VARIABLES

Affective Neutrality Affectivity

Performance Quality Performance Quality

Universalism Bureaucracy
Specificity
Particularism

Universalism
Diffuseness
Particularism Parenthood

29
 Given the above contingency table, the researcher may want to enter into the
resulting cells normatively defined or actually observed scores computed on the
basis of the above suggested methods for the description, analysis and comparison
of any two relationships. Of particular interest may be difference scores between
normatively defined and actually observed relationships in order to ascertain the
amount of deviance present in the system. These scores might then be used both to
measure strain perceived by social actors and to estimate the potential for
systematic change.

To recapitulate, we have defined social policy as the management of structural

Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

change in a system. Further, structure is taken to mean a relatively stable pattern of

relationships among social positions or simply a social network. Mitchell (1969) has
advanced a working definition of social network: "a specific set of linkages among a
set of persons, with the additional property that the characteristics of these linkages
as a whole may be used to interpret the social behaviour of the persons involved"
(p.2). Since social policy issues typically affect large aggregates of people it is
necessary to relax Mitchell's definition of a social network as a specific set of
linkages among a set of persons and to consider actors other than "real" persons in
defining a relationship. Thus we may treat corporate actors such as political parties
or socioeconomic groups - that is, groups sharing a particular attribute of interest -
as the basic nodes of a social network. This concession is necessary and justifiable.
It is necessary because even a small constituency - say of 10,000 people - affected by
a specific social policy results in far too many possible pairwise relationships
(almost 100 million) to permit analysis even in a large computer.

How can one justify this relaxation of Mitchell's definition if one wants to infer
the presence of links between nodes that are comprised of complex entities such as
political parties or ethnic groups? Laumann and Pappi (1976) suggest an appealing
solution to this problem. They propose that one may "consider the 'relationship'
between the two nodes consisting of complex entities as being indicated by the
differential likelihood that their constituent elements have the relationship in
question with one another" (p.19). For instance, Social Democrats may more
readily support Labour rather than Liberal party causes. Given such a tendency
one may suggest that the party of Social Democrats is closer to the Labour party
than to the Liberal party. Given this relaxation of the network notion, all sorts of
social relationships may justifiably be subjected to network analysis and become
part of a structural analysis in an effort to shed additional light on social policy
issues.

Lest our definition of structure be dismissed as too static and hence irrelevant for
the study of social policy, three postulates of structural analysis must be identified
(see Laumann and Pappi, 1976, p.7-8):

- Postulate I (relationship-specific structures) asserts that there exists a

multiplicity of social structures in any complex social system that arises out of the
many possible types of social relationships linking positions to one another. Thus
we do not agree with Nadel (1957) that structure can be ascertained independent of
context. For instance, societal structures based on the mode of production and
division of labour will be different from societal structures stemming from
30
 interactions among national groups, races, religious and other groups.

- Postulate II (distance-generating mechanism) asserts that, for any given

relationship-specific structure, there exists a principle of systematic bias in
channeling the formation of (or making more likely the) relationship between
certain kinds fo positions and the avoidance of such relationships between others.
The notion of coalition formation among specific interest groups is a concrete
application of this postulate. Pfeffer (1981) discusses this concept in considerable
detail.
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

- Postulate III (structural contradictions) holds that, given a plurality of

relationship-specific structures predicated on different principles of organisation,
structural contradictions are possible features of any complex system.

The emphasis on social policy as a modifier of social structure requires that

changes in social structure be identified and measured. The remainder of this
article is devoted to a brief discussion of the potential and limitation of social
indicators and their suitability as a linchpin between social structure and social
policy.

Social Indicators and Social Policy

The contemporary study of social policy is characterised by two opposite

orientations. On the one hand, the econometric approach to social policy analysis is
guided by highly abstract models and employs complex mathematical algorithms
which defy all but a small group of social researchers. As a result econometric
solutions to social policy issues are difficult to translate into social action
programmes and must largely be accepted on good faith. The philosophical
approach, on the other hand, cither avoids quantifiable empirical information
altogether or else distorts it by exclusively focusing on those empirical aspects
which support a stated value configuration or implied ideology.

The econometric and philosophical approaches nevertheless have one thing in

common: a preoccupation with social indicators. What arc social indicators? The
concept of social indicators is vague and elusive. Freeman and Sheldon (1977)
pointedly comment "The elusiveness of the social indicator concept stems from the
multitude of views on the relevance and purpose of organising statistics on the state
of affairs of the system of interest. The vagueness of the concept encourages
persons to advocate their own perspectives, further increasing the confusion about
the utility of social statistics in planning, programme development and scholarly
endeavours" (p.1351). Still the concern with social indicators has developed to the
point where one might speak of the field as a social movement (Duncan, 1969).
Armstrong (1981) provides a succinct definition of social indicators in diagramma-
tical form:

31
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

Despite this seemlingly clear-cut definition of social indicators the vagueness and
elusiveness of the concept remain whenever they are used in actual research. The
major problem with social indicators derives from their very nature: they are
measures and surrogates (see Carley. 1981). As such they are affected by

32
 measurement error and suffer from the less than perfect overlap with the reality
which they are to indicate. Even in situations where a strong correlation obtains
between the phenomenon (indicator) and the noumenon (the underlying reality of
interest), problems of both reliability and validity reign supreme. A relevant
example is provided by the legal presumption that blood alcohol concentration in
excess of .05 percent constitutes dangerous driving and the current debate of the
usefulness of random breathalyser tests for accident prevention. Incidentally, this
very debate neatly illustrates the precarious balance between providing social
security without unduly restricting civil liberty.
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

It would be difficult to produce a more telling example of the vagueness and

elusiveness of social indicators than Alford's (1975) review of Turk's attempt to
infuse social indicators with theoretical meaning. Alford accepts Turk's method
and data of a study of 130 cities in the United States but contrasts the author's
consensus interpretation of the social indicators with a conflict interpretation. The
social policy implications, depending on the adopted theoretical model, would
appear to be at loggerheads as Figure 5 clearly shows.

Figure 5: Two Theoretical Perspectives on the Same Empirical Indicators

of Characteristics of 130 United States Cities

Theoretical variable Theoretical variable

Empirical indicator (Turk) (Alford)

1 . Poverty dollars New network of Capacity for symbolic

per capita up to interorganizational response by local
June 1966 activity elites to potential
for social conflict
2. Percent of low- Demand Potential for social
Income families, non- conflict
whites, youths not in
school
3. Neighborhood Youth Complex interorga- Legitimating
Corps project spon- nizational network sponsorship
sored by local
federation
A. Number of head- Extralocal Location of national
quarters of national integration elites in the city
associations of all
kinds
5. Number of local Local integration Extent of legitimation
voluntary associations of rule of local
which do not represent elites
organized cleavages
6. Existence of Local integration Presence of structural
"reform government" devices for reducing
institutions the salience of
cleavages

33
 It would be futile to arbitrate between these two competing interpretations of
social indicators. Any indicator which can mean opposite things either is not an
indicator of an empirical referent or has arbitrarily been loaded with ideological
meaning elements foreign to it. It is at this point that social indicators become social
vindicators. We are not implying that data speak for themselves or even have an
objective existence of their own; rather data are information anchored in a
theoretical frame of reference. Thus the major problem with social indicators is not
the empirical observation but rather its interpretation. This points to two important
qualifications.
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

First any theoretical interpretation of social indicators must be tested against

alternative interpretations of the indicators after they have been disaggregated.
Unfortunately the theoretically relevant information to disaggregate social
indicators is frequently not available. When it exists, it is typically used to identify
main effects only. However the acid test of any theoretical interpretation of basic
data consists in the careful inspection of possible two-way and higher-order
interaction effects. Recent developments in log-linear and latent structure analysis
(Goodman, 1978) provide powerful techniques for the disaggregation of social
indicators. Lest there be a misunderstanding, social indicators have to be
aggregated to some degree in order both to avoid an inforamtion overload and to
render them useful for policy formulation and implementation. However,
aggregation criteria should be chosen on the basis of theoretically meaningful and
practically feasible considerations.

A second qualification pertains to the distinction between objective and

subjective measures of social indicators. Most measures are termed objective
because they result from easily counted units and do not require a personal
evaluation from the reporting individual. It should however be noted that they are
still subjective in the sense that the researcher has to rely on the veracity of the
reporting individual. Examples of "objective" social indicators are employment
figures, crime and victimisation rates as well as the extent of provision of social
services. Subjective indicators, on the other hand, measure the "perceived
conditions affecting a selected population and the subjective attitudes towards
those conditions held by persons in that population" (Campbell et al. 1976). For
instance, happiness, satisfaction with specific aspects of one's life (work, marriage,
family) or with quality of life in general are examples of subjective indicators
extensively used in contemporary research. Clearly the two types of social
indicators are complimentary for the understanding of social structures and for the
formulation, implementation and evaluation of social policy.

Several researchers doubt the complimentarity and even compatibility of

subjective and objective social indicators because of the frequently poor fit
between them (Schneider, 1975; Wasserman and Ghua, 1980; Kutz, 1978).
Conversely Campbell et al (1976) and Atkinson (19S2) conclude that there are
subjective quality of life indicators that react very sensitively to the circumstances
they should measure. Walter-Busch (1983) offers three qualifications which, if met,
will virtually guarantee a strong subjective-objective fit:"(l) the perceptual
measures applied should not try to reproduce more than simple judgements of
respondents about concrete aspects of their living conditions; (2) the number of
34
 respondents should be large enough; and (3) the range of variation of the living
contexts to be compared should be wide enough to allow for a fine-meshed, richly
faceted disaggregation of data"(p.384). Headey, Holmstrom and Wearing (1983)
offer considerable evidence for the reliability and validity of subjective social
indicators. Specifically, the average Cronbach's alpha for nine of their subjective
measures amounts to .83 (range .76 to .89) while a LISREL maximum likelihood
analysis (Andrews and McKennel, 1980) produced indices which yield about 60%
valid variance. This finding was further corroborated by Heady et al. when they
compared survey results and in-depth interviews by professional psychologists.
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

Summary and Conclusion

This article has attempted to identify the difference between traditional social
policy concepts and social policy as a modifier of social structure. It was noted that
a large number of writers have equated social policy with welfare politics. It was
also alleged that in doing so emphasis came to be placed on equal rather than
equitable treatment. To overcome this unduly narrow conceptualisation of social
policy it was argued that social policy should be predicated on the concepts of
equity and social justice. On this basis social policy was defined as those
mechanisms and processes whereby systems prevent, postpone, introduce and
manage structural change. This definition of social policy led to a definition of
structural change and its major components. The article concluded with some
observations about the usefulness and limitations of social indicators as a means of
examining social policy.

If social policy is supposed to create more equitable structures the orchestrated

pronouncement of various ideologies by politicians and sundry professional groups
will not do (Vinson, 1983). Solid research is required. Ideally it should be based on
a value consensus or at least value prevalence, Currently that does not exist.
Competent research which spells out the implications and consequences of various
policy alternatives can significantly contribute towards this goal. It would appear
that three major issues must systematically be addressed if we are to succeed. First,
the boundaries of both social policies and social structures must be identified
explicitly and thoughtfully at the planning stage lest they be determined implicitly
and haphazardly at the stage of implementation. Second, inequality of both actual
and ideal distributions of scarce resources (various manifestations of civil liberty
and social security) must be conceptualised and operationalised to permit informed
comparison of what is and what ought to be. Third, adequate measures and
methods for ascertaining social change must be developed to assess the impact of
social policy on social structure and vice versa. These issues are extensively
discussed in W.L. Grichting, Security vs. Liberty: Analyzing Social Structure and
Policy, Lanham, Md.; University Press of America, 1984. This article is based on
the first chapter of this publication.

35
 References
Alford, R.R. "Ideological filters and bureaucratic responses in interpreting research: Com-
munity planning and poverty." In Demerath, et al (Eds.), Social policy and
sociology. New York: Academic Press, 1975.
Anderson, D.S., and "Access to privilege: Patterns of participation in Australian post-secondary
Vervoom, A.E. education." Canberra: Australian National University, 1983.
Andrews, F.M., and "Measures of self-reported well-being: Their affective, cognitive, and other com-
McKcnnell, A.C. ponents." SocialIndicatorsResearch, 1980,8, 127-155.
Armstrong, A.F. "An introduction to social indicators." Proceedings of the National Social Indi-
cators Conference. Melbourne: Bureau of Statistics, 1981.
Atkinson, T. "The stability and validity of quality of life measures." Social Indicators Re-
search, 1982,10,113-132.
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

Campbell, A., The quality of American life: Perceptions, evaluations, and satisfactions. New
Converse, P., and York: Russell Sage, 1976.
Rodgers, W.
Brennan, F. Too much order with loo little law. Brisbane: University of Queensland Press,
19S3.
Carley, M. Social measurements and social indicators: Issues of policy and theory. London:
George Allen & Unwin, 1981.
Coutu, W. Emergent human nature, a symbolicfield interpretation. New York: A.A. Knopf,
1949.
Duncan, O.D. Toward social reporting: Nat steps. New York: Russell Sage Foundation, 1969.
Ferge, Z. "Hungary". In S.B. Kamerman & A.J. Kahn (Eds.), Family policy: Government
and families in fourteen countries. New York: Columbia University Press. 1978.
Frankel, M.R. Inference from survey samples. Ann Arbor: University of Michigan, 1971.
Freeman, H.E., and "Social indicators." In J.B. Turner et al (Eds.), Encyclopaedia of Social Work
Sheldon, E.B. (17th ed.). Washington: National Association of Social Workers, 1977, 2, 1350-
1354.
Gillin, J.P. The ways of man. New York: D. Appleton-Century Co., 1948.
Goodman, L.A. Analyzing qualitative/categorical data. Cambridge, Massachusetts: Abt Books,
1978.
Graycar, A. Social policy: An Australian introduction. Sydney: Macmillan, 1977.
Graycar, A. (Ed.) Perspectives in Australian social policy. Sydney: Macmillan, 1978.
Graycar, A. Welfare politics in Australia: A Study in policy analysis. Sydney: Macmillan, 1979.
Greenberg, J., and Equity and justice in social behaviour. New York: Academic Press, 1982.
Cohen, R.L.
Grichting, W.L. "Controlling alcohol abuse in Australia: From treatment to prevention." British
Journal of Addiction, 1983, 78, 37-50.
Heady, B., "Quality of life studies and the development of positive welfare programs." In J.
Holmstrom, E., and Dixon & D.L. Jayasuriya (Eds.), Social policy in the 1980s. Canberra: Canberra
Wearing, A. College of Advanced Education, 1983.
Kutz, T.J. "Quality of life, an objective and subjective variable analysis." Regional Studies,
1978, 12, 409-117.
Laumann, E.O., and Networks of collective action. New York: Academic Press, 1976.
Pappi. F.U.
Marshall, T.H. Social policy. London: Hutchinson, 1965.
Mitchell, J.D. (Ed.) Social networks in urban situations. Manchester, England: Manchester
University Press, 1969.
Nadel, S.F. The theory ofsocial structure. London: Cohen & West, 1957.
Parsons, T. The structure ofsocial action. New York: McGraw Hill, 1937.
Parsons, T. "A revised analytical approach to the theory of social stratification." In R.
Bendix & S.M. Lipset (Eds.), Class, status, and power: A reader in stratification.
New York: Free Press, 1953.
Peacock, A. The welfare society. London: Liberal Publication Department, 1960.
Pfeffer, J. Power in organisations. Marshfield, Massachusetts: Pitman, 1981.
Pinker, R. Theory, ideology and social policy. Sydney: Social Welfare Research Centre,
University of New South Wales, Report No. 26, 1982.
Popper, K.R. Objective knowledge: An evolutionary approach. Oxford: Clarendon Press, 1981.
Priorities Review Staff Interim report on goals and strategies. Canberra: AGPS, 1973. Parliamentary
Paper No. 26 of 1973.
Rawls, J. A theory ofjustice. London: Oxford University Press, 1972.
Rein, M. Social policy:Issuesof choice and change. New York: Random House, 1970.
Rolfe, B. In F.K.H. Maher, W. Waller, & D.P. Derham, Cases and materials on the legal
process (2nd ed.). Melbourne: Law Book Co.. 1971.

36
 Schneider, M. "The quality of life in large American cities: Objective and subjective social
indicators." Social Indicators Research, 1975, I, 495-509.
Sim, R.A. Government grants to voluntary organisation. Ottowa: Voluntary Action Com-
mittee of the Canadian Association of Adult Education through the Citizenship
Branch of the Department of Secretary of State, undated.
SthVell, F.J.B. Equality. In P. Weller and J. North (Eds.), Labor: Directions for the Eighties.
Sydney: Novak, 19S0.
Tawney, R.H. Equality (revised edition). London: Allen & Unvwin, 1952.
Thomas Aquinas Summa Theologiae, London: Blackfriars in conjunction with Eyre and Spot-
tiswoode, 1964-1976.
Titmuss, R.M. Social policy: An introduction. London: George Allen & Unwin, 1974.
Townsend, P. Sociology and social policy. Harmondsworth: Penguin Books, 1975.
Downloaded by The University of British Columbia Library At 20:37 30 January 2016 (PT)

Vinson, T. Keynote Address: Australian Association of Social Workers National Confer-

ence. Sydney,1983.
Walter-Busch, E. "Subjective and objective indicators of regional quality of life in Switzerland."
Social Indicators Research, 1983,12, 337-391.
Wasserman, I., and "Objective and subjective social indicators of the quality of life in American
Chua, L.A. SMSA's: A reanalysis." Social Indicators Research, 19S0,8,365-381.
Willey, M. "The validity of the culture concept." American Journal of Sociology, 1929,
204-219.

37