Relational Sociology: A New Paradigm for the Social Sciences
Relational Sociology: A New Paradigm for the Social Sciences
‘Simultaneous invention’ has become commonplace in the natural sciences, but is still virtually unknown within the sphere of social science. The convergence of two highly compatible versions of Critical Realism from two independent sources is a striking exception. Pierpaolo Donati’s Relational Sociology develops ‘upwards’ from sociology into a Realist meta-theory, unlike Roy Baskhar’s philosophy of science that works ‘downwards’ and ‘underlabours’ for the social sciences.
This book systematically introduces Donati’s Relational Sociology to an English readership for the first time since he began to advance his approach thirty years ago. In this eagerly awaited book, Pierpaolo Donati shifts the focus of sociological theory onto the relational order at all levels. He argues that society is constituted by the relations people create with one another, their emergent properties and powers, and internal and external causal effects.
Relational Sociology provides a distinctive variant upon the Realist theoretical conspectus, especially because of its ability to account for social integration. It will stimulate debate amongst realists themselves and, of course, with the adversaries of realism. It is a valuable new resource for students of social theory and practising social theorists.
تحميل (pdf, 1.76 MB)
اقرأ الكتاب مباشرة
Send-to-Kindle or Email
You may be interested in
PDF, 1.54 MB
‘Simultaneous invention’ has become commonplace in the natural sciences, but is
still virtually unknown within the sphere of social science. The convergence of two
highly compatible versions of critical realism from two independent sources is a
striking exception. Pierpaolo Donati’s Relational Sociology develops ‘upwards’
from sociology into a realist meta-theory, unlike Roy Baskhar’s philosophy of
science, which works ‘downwards’ and ‘underlabours’ for the social sciences.
This book systematically introduces Donati’s relational sociology to an English
readership for the first time since he began to advance his approach thirty years
ago. In this eagerly awaited book, Pierpaolo Donati shifts the focus of sociological
theory onto the relational order at all levels. He argues that society is constituted
by the relations people create with one another, their emergent properties and
powers, and internal and external causal effects.
Relational Sociology provides a distinctive variant upon the realist theoretical
conspectus, especially because of its ability to account for social integration.
It will stimulate debate amongst realists themselves and, of course, with the
adversaries of realism. It is a valuable new resource for students of social theory
and practising social theorists.
Pierpaolo Donati is Professor of Sociology at the University of Bologna (Italy).
Past-President of the Italian Sociological Association, he is known as the founder
of ‘relational sociology’ or the ‘relational theory of society’, an independently
developed form of critical realism. He has published more than 600 works (see:
Building a Relational Theory of Society: A Sociological Journey, in M. Deflem
ed., Sociologists in a Global Age. Biographical Perspectives, Ashgate, Aldershot,
Other titles in this series:
From One ‘Empire’ to the Next
Science for Humanism
The recovery of human agency
Charles R. Varela
Philosophical Problems of Sustainability
Taking sustainability forward with a critical realist approach
Dialectic and Difference
Dialectical critical realism and the grounds of justice
Interdisciplinarity and Climate Change
Transforming knowledge and practice for our global future
Edited by Roy Bhaskar, Cheryl Frank, Karl Georg Høyer,
Petter Naess and Jenneth Parker
Conversations about Reflexivity
Edited by Margaret S Archer
A new paradigm for the social sciences
A new paradigm for
the social sciences
First published 2011
2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN
Simultaneously published in the USA and Canada
270 Madison Avenue, New York, NY 10016
Routledge is an imprint of the Taylor & Francis Group, an informa business
This edition published in the Taylor & Francis e-Library, 2010.
To purchase your own copy of this or any of Taylor & Francis or Routledge’s
collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.
© 2011 Pierpaolo Donati
The right of Pierpaolo Donati to be identified as the author of this work
has been asserted by him in accordance with the Copyright, Designs and
Patent Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in
any information storage or retrieval system, without permission in writing
from the publishers.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging in Publication Data
Relational sociology : a new paradigm for the social sciences / edited by
1. Sociology. I. Donati, Pierpaolo, 1946–
ISBN 0-203-86028-4 Master e-book ISBN
ISBN: 978–0–415–56748–0 (hbk)
ISBN: 978–0–203–86028–1 (ebk)
To my parents and brother
who taught me the meaning of relations
List of figures
List of tables
Introduction: prospects for a ‘relational sociology’
The relational paradigm: its implications for the
understanding and organization of society
Society as a relation
Critical realism as viewed by relational sociology
Observing and thinking relationally: the premises of
the relational theory of society
Social change in the light of relational sociology
Reflexivity after modernity: from the viewpoint of
Doing sociology in the age of globalization
1.1 The conceptualization of the human person as someone who
develops between nature, practice, social interaction and
1.2 The epistemic triangle of critical realism
2.1 The place of relations in the semantics of identity
2.2 The components of social relations according to the AGIL scheme
2.3 The relational observation is that made by O (a third observer,
different from agents A and B) who examines the relations between
A and B and their emergent effect (Y)
3.1 Epistemic triangle and epistemic quadrangle
3.2 The components of sociology as a knowledge system
3.3 The articulation of critical realist theory according to relational
5.1 Theories of social change and modes of observation
5.2 The scheme of social change according to the holistic paradigm
5.3 The scheme of social change according to the individualist paradigm
5.4 The scheme of social change according to the relational paradigm
5.5 The three registers of time implied in social change
6.1 The basic schema for the social role of reflexivity in social change
6.2 The basic schema extended according to the relational theory of
6.3 The place of reflexivity in the morphogenetic cycle
6.4 The temporal sequence of emergence of a social structure through
personal, social and system reflexivity
7.1 Modern, post-modern and trans-modern sociologies
7.2 The basic scheme of the constitution of an actor’s (A) social
identity in trans-modern society
4.1 Three versions of the AGIL scheme
6.1 Forms of societal differentiation and their related types of
reflexivity, principles of identity, and associated empirical
spheres in society
7.1 The forms of social differentiation
‘In the beginning is the relation’. Pierpaolo Donati makes this his motto and
repeats it several times in the following chapters. It intentionally echoes the
opening of St John’s Gospel: ‘In the beginning was the word’ – ′Εν άρχή ήν ό
λόγος – because he wants to shock his readers into recognizing his foundational
statement for sociology as being exactly that. Historically, social scientists have
been immunized against taking social relations seriously. In the canon of sociology
they are synonymous either with social interaction, such as queuing for a bus or
shopping, which leave the individuals involved unchanged; they remain mere
aggregates. Alternatively, in the holistic version, agents are moved around the
board as collective träger, who are robotically responsive to systemic forces but
whose social relations are a matter of indifference; even ‘mobilization’ is a nonrelational term.
Donati needs to use the same shock tactics again and again. Thus, for example,
he reverses the first Durkheimian rule that the subject matter of sociology is ‘social
facts’ with the corrective that ‘social facts’ are ‘social relations’. In other words,
there is no question of subtracting the individuals and leaving behind a residue
that can be dubbed ‘the social’, because nothing at all would remain. Any
relational good is indivisible in the sense that an attempt to divide it destroys the
very social relationship that generates it. Relational goods cannot be parcelled out
amongst those producing them. Thus, a divorcing couple cannot take away their
private shares of the marriage; they can only divide up their possessions and even
their children between them. The marriage itself is dissolved and thus qua talis no
longer generates anything because the relationship no longer exists. Relational
goods reside in the relationships that link or bond the members concerned. Hence,
no-one can take away part of the orchestra or the football game as their personal
property; they can only take themselves away from the orchestra or the team. In
short, the social order is a relational entity, sui generis because emergent in kind.
Substantively, social science has become so engrossed by market exchange
relations and political command relations that social relations have been squeezed
out. Yet, in fact, the former are merely interactions – procedural transactions that
proceed by instrumental rationality. They do not generate the relational good that
is characteristic of friendship, which belongs to neither of the friends but is shared
and valued by both. Hence, for a quarter of a century, Donati has been fulminating
against the lib/lab oscillation in politics and economics alike because commodification and command (increasing bureaucratic regulation) both generate relational
‘bads’, destructive of human relations.
Instead, what he champions is the French Revolutionary slogan that quickly
dropped off the European and is now off the Global agenda – fraternity. Its
promotion is the pursuit of the common good in society, not the Total Good or the
greatest good of the greatest number, because the latter rises if the well-being of
some is discounted or discarded. Contractual agreements and electoral systems are
of practical use, but they have nothing to do with generating the common good. In
other words, relational sociology fosters its own emancipatory project, based upon
valorizing ‘free-giving’, reciprocal relations rather than exchange or command
relations, the ‘social–private’ (third sector), and the complex of friends and family
(fourth sector) as the planks building up a robust civil society.
These two strands are closely interwoven: first the bold assertion that ‘social
reality is social relationality’ and second that, by corollary, society as an entity
with an emergent relational ontology cannot be theorized on the basis of the
atomistic ‘homo economicus’ or ‘homo sociologicus’ but only in terms of ‘homo
relatus’. Both his endorsement of a stratified social ontology and of an emancipatory project serve to make Donati’s relational sociology and critical realism very
close together indeed.
In fact, it is a very rare occurrence in sociology, unlike the natural sciences, to
have the ‘simultaneous invention’ of a remarkably similar approach from entirely
independent sources. The independent development of these two perspectives,
over much the same period, is because of more than the regrettable linguistic
barrier that still impedes Anglo–Italian exchanges, although that has played its
part. Equally importantly, the development of each was rooted in entirely different
parts of the history of social theorizing – they come from different stables. The
first generation of critical realists came from Marxism or conflict theory and found
their common enemy in empiricism. Donati made his way along a very different
path from Simmel, by way of an engaged critique of Parsons and later of
Luhmann. Theoretical starting points and the first adversaries that one’s new
perspective confronts are background influences that continue to remain salient.
Personally, I regard relational sociology and critical realism as fully complementary, both as meta-theories and as explanatory programmes. But that does not
preclude noting their differences because this is precisely where the two can enrich
To begin with – and this is not inconsequential – British readers (more so than
North Americans) will not encounter their familiar array of ‘heroes and villains’
amongst the theorists Donati takes most seriously. Instead, in this book, they are
invited to engage as seriously with Parsons and his critics as with Luhmann and
systems theory in general. Donati himself provides a radical critique of these
sources, but on the basis of a deep conversancy with this thinking. What is
considerably more important is that he takes the central Parsonian concern with
social order very seriously indeed, whilst providing his own critique of normative
integration and ever more extreme forms of functional differentiation.
Conversely, critical realists have frankly never been very concerned or convinced about social order (I include myself in this censure, until my exchanges
with Pierpaolo increased over the last decade). Such attention as we AngloAmerican realists have given it has accentuated power relations and necessary
compromises and concessions between interest groups etc. This, in my view, is
because we have – up to the new millennium – taken a negative stance towards
social order as the barrier to social transformation. There are ‘old’ exceptions
(nostalgia for the traditional working class community and its values) and there
are ‘new’ exceptions (the recent and surprising enthusiasm of some for both habits
and habitus). Nevertheless, the fact remains that critical realism does not have a
theory of social integration: when it exists we admit it, but why it exists – other
than as a tense balance of opposing interests and accompanying ideational
manipulation – we have seriously failed to theorize.
This is, perhaps, the main ground on which Donati’s relational sociology contains much that we can learn from, about relations, relational goods, and relationality (relations between relations). Given the complementarity of the two
perspectives, this can be assimilated without compromising our commitment to
ontological emergence, to epistemological relativity, or to judgemental rationality
because the ‘three pillars’ of realism are upheld in common. How much better,
more fruitful, and vastly more sophisticated this can be than some of the forced
‘reconciliations’ with antipathetic approaches that have recently been advocated.
First and last, relational sociologists and critical realists care deeply about the
human capacity for fulfilment and the human liability to multifarious forms of
suffering. As it becomes increasingly popular to blur the human/non-human
distinction in social theory, nothing could be more welcome than to find that in
relational sociology we encounter not just an abstract theoretical convergence but
a shared commitment to the promotion of human thriving.
Margaret S. Archer
Kenilworth, March 2010
There is an invisible world generated by human beings, but that human beings do
not see or come to see very rarely. This is the world of social relations. They ‘act’
this world, they live in it, but they do so with very little awareness. They take if for
granted, as they do the air they breathe. They become aware of its existence only
when they feel that it is lacking in some way or when it becomes so distorted as to
make them feel very bad towards themselves.
Scholars from different disciplines frequently refer to ‘social relations’, but the
latter are treated as being ‘derived from’, or as by-products of something else. For
most of them (the ‘Weberians’) social relations are a projection of individual/s.
More or less unquestioningly, they assume that the individual is the only real entity
they confront (the individual as the ens realissimum). For others (Marxists or
‘Durkheimians’), social relations are a product of conditioning by social structures
and systems. Everyone speaks of social relations, as do all social theories. But the
fact of the matter is that most people, like most social theorists, think of social
relations as a product of the Self or as an external constraint impinging upon it.
What really constitutes the ‘social relationality’ of our world remains latent,
hidden, unspoken, represented in a biased form, ignored or nullified.
This book has been written to shed a new light on this world. In order to see it,
we need an adequate framework. In my opinion, sociology to date has failed to
explore it adequately or to develop the means appropriate to its exploration. In this
work, I propose a new way to perceive, to conceptualize and to deal with it. I put
forward a ‘relational paradigm’ for sociology which depends upon the social
ontology of critical realism (understood as the more general approach). Within
such a framework, I suggest a way of analysing the configuration of any social
relation as an interlacing of subjective and objective elements, which exists in a
complex environment, so as to describe, understand and explain social phenomena
as ‘relational facts’ (not as ‘things’ as Durkheim claimed). To my mind, society
does not host relations, it is not a space–time where relations happen, it is relations.
This framework leads me to elaborate new concepts, such as those of relational
goods, relational differentiation, relational reason and so on. As a matter of fact, a
relational sociology redefines all the concepts commonly advanced and employed
in the social sciences, and adds new ones. It operates in synergy with the morphogenetic approach developed by Archer from the viewpoint of critical realism, with
which relational sociology is closely convergent. In sum, I will try to show that
there exists a sui generis reality which I will call ‘the order of relations’. This
expression resembles what sociologists know as the ‘order of interaction’, but is
different from it. The former precedes and exceeds the latter.
My hope is that this book can help the social sciences, as well as ordinary
people, to learn to read the ‘world as a social relation’ and, thus, to gain an
increased awareness of the reasons why social relations, as a sui generis reality,
can make society better or worse, happy or sad, just or unjust.
Relational sociology aims at disclosing the fact that every human being is
relationally constituted as a person, and the same holds true for any social
institution. According to this paradigm, a social formation is human insofar as the
social relations constituting it are produced by subjects who orient themselves
reciprocally towards one another on the basis of a meaning that surpasses functional
requirements. Even when we talk to our Self in solitude or isolation, social relations
are at stake. We can forget them, we can ignore them, we can banish them. But they
are still there. Modernity has tried to immunize human individuals against social
relations, and continues to do so. It is precisely for that reason that modernity is
now at its end. We are what we care about, and if we do not relate to significant
others, we are nothing, we become nothing. We are our ‘relational concerns’, as
individuals as well as social agents/actors, since we necessarily live in many
different contexts that are social circles (like a family, a network of friends, maybe
a civil association, up to a nation) which imply a collective identity.
There is a certain correspondence between personal identity (‘Who I am is what
I care about”) and collective identity (‘Who we are is what we care about’). For
relational sociology, this correspondence does not mean that we – as individual
persons – are subjugated or subordinated by any ‘holistic’ entity whatsoever. The
correspondence is due to the fact that the link between personal and social identity
is a relational matter: we are what we care about not because we (as a group,
network or any collective entity) think in the same way, or because we share
external commitments, or because we have mutual intentionality, or because we
are conditioned by the same structures, but because we are in a special relation, and
that relation is what makes us reflexive in a social, instead of an individual way.
Individual relations are a projection (a performance) of an individual mind. Social
relations, on the contrary, reflect the performance of an emergent reality between
two or more people, groups, and even institutions when they act as social subjects.
To say that ‘we are what we care about’ is not reducible to a kind of ‘resonance’
between what ego cares about and what alter cares about. There is ‘something
different’ that lies in between, and this is the social relation. We need a relational
reflexivity to catch it. Such reflexivity consists in the subjects orienting themselves
to the reality emerging from their interactions by taking into consideration how
this reality is able (has its own powers) to feed back onto the subjects (agents/
actors), since it exceeds their individual as well as their aggregate contribution to
it by virtue of their personal powers. The book is about this reality.
Bologna, March 2010
Although this book contains the fruit of many years of lonely work, it could not
have appeared without the help of a lot of friends and colleagues whose scientific
curiosity has supported me along the way. My first thanks are to Margaret Archer
who greatly encouraged and assisted me in presenting my theory to an international audience. Without her unending hunger for knowing more about my
relational sociology, many ideas were bound to have remained implicit or
underdeveloped. Her involvement has brought about a remarkable encounter
between the Italian relational sociologists and the international group of critical
realists, which is reflected in this book. My gratitude also goes to the network of
relational sociologists, in particular Pablo García Ruiz, Giovanna Rossi, Riccardo
Prandini, Andrea Maccarini, Gabriel Chalmeta, Emmanuele Morandi, Paolo
Terenzi and Sergio Belardinelli for their support in following this itinerary. The
Reflexivity Forum that met at the University of Warwick in September 2008 was
a further prompt to pursue the project of introducing the relational theory of society
to a broader public. Lastly, my thanks to all those friends, like Bob Constable and
Doug Porpora, who have encouraged my research and helped to make this book
Prospects for a ‘relational sociology’
By way of preface, I wish to reply positively to the central question ‘does the
prospect of a general sociological theory still mean anything today, in the time of
globalization?’ The road map I will follow is drawn, on one hand, upon a revision
of classical theory, and on the other hand, on the construction of an epistemological
and ontological framework which is seen as a necessary condition for the building
up of a new general sociological theory appropriate to the current era.
My key point is that what is common to the classical sociological tradition is
implicit and not yet still well understood today: i.e. that the nature of ‘social facts’
is a relational matter.
All the classics tried to identify and define what a social fact is, from their
different perspectives (either ‘social action’ or social structure or ‘the system’),
which they usually conceived of as being opposed, but in fact are complementary
to one another. Not one of the classics succeeded in grasping what a social relation
is, despite their great achievements in understanding single aspects of what a social
The most significant attempt to reconcile and unify the classical approaches
and traditions, that is, Parsons’ ‘general theory’, failed precisely because it tried to
unify ‘action theory’ and ‘system theory’ without having a generalized theory of
social relations, i.e. without arriving at a ‘relational theory’ of society, freed from
the assumptions of modernity.
Certainly, many other disciplines are interested in social relations, such as
philosophy (from the metaphysical point of view), psychology (from the psychic
point of view), economics (from the resource perspective), law (control by rule),
and even biology (bioethics). What is proper to sociology is its distinctive way of
looking at social relations, i.e. in terms of a theory which holds all the dimensions
of social relations together and develops an understanding of their differentiation
as well as their conflict and integration on an empirical basis. Only sociology has
the task of conceptualizing social relations in their ‘relationality’.
Contemporary sociology is, of course, both in continuity and discontinuity with
classical sociological theory, in many different ways. However, what I want to
emphasize here is precisely the fact that contemporary sociology is still caught up
in the main divide between action and system approaches. To my mind, contemporary theories are at variance with one another not because they are irreconcilable
(at least in principle), but because they do not confront each other directly on the
terrain of social relations by taking these as their subject matter. Quite simply, they
look at different aspects of social relations (symbols, forms of communication,
structure, agency, one function or another, etc.), and their disputes are about what
aspect or dimension can be thought of as ‘the’ major factor (variable, explanans,
etc.) in order to account for what is happening (explanandum). In other words, the
continuity with the classics is basically given by the fact that contemporary
sociology has not changed its basic assumptions about what a social fact is.
Discontinuities are relative to the different ways in which they account for social
phenomena (such as globalization) that are different from or novel in respect to
In the interests of parsimony, four main rival schools or general sociological
theories can be distinguished: (1) action theory (including rational choice theory)
in various and often opposed versions; (2) system functionalism (taking many
different forms, but still sharing common background assumptions); (3) critical
theory (again of different kinds); (4) narrative theory (including its ethnomethodological and linguistic strands). In my view, they differ because they are
interested in different apects of social relations, which they try to make more
significant than those treated by contending theories. This does not mean that there
are no other ‘schools’ of importance. For instance, the realist social theory of
morphogenesis set forth by Margaret Archer is a good example of an outstanding
theory which is fully compatible with the perspective I have in mind. Here, what
I seek to underline are the main cleavages or rivalries among paradigms that I
believe can be better understood within a more general framework, which I term
the ‘relational theory of society’.1
Classical sociology has clearly used a concept of society that is heavily
dependent upon the modern idea of the nation-state, which was undoubtedly
inherited by Parsons. Of course, the subsequent processes of globalization cannot
but nullify this idea. However, it would mean repeating the same mistake as the
classics to think of a new sociological theory (or paradigm or research program) as
being required by ‘global society’. Such an attitude would lead us to reify society
once again, as the classics did, in one way or another. ‘Global society’, or globalization as a dynamic historical process, can better be understood not as an ‘object’
in itself, but as a way of configuring social relations in a mode radically different
from the past. Otherwise we end up saying that globalization is something which
has always existed, at least from the Roman Empire onwards, which may be an
interesting perspective for historians to take but would put sociology on the wrong
track. What sociology should investigate is precisely this different mode, in its
diverse expressions. It cannot be the same as (or even similar to) the modes existing in the past, for a number of reasons (those of time and space besides cultural,
See P. Donati (1991) Teoria Relazionale della Società, FrancoAngeli, Milan.
communicative, technological and even biological reasons). To me, ‘global
sociology’ cannot be sociology simply because it is not the object targeted for
investigation which makes the difference in sociological theorizing. To speak of
‘global sociology’ can lead us to fail in responding to the ‘core challenge’ that
sociology still has to confront.
To my mind, the crucial elements making for the identity of sociology are (i) its
directive-distinction, i.e. the relationality of social relations, and (ii) the way it
deals with the processes of differentiation, conflict and integration of the components of social relations. No other discipline can do that. This contribution tries
to make these statements a little more explicit.
Social relations as the object of sociology
From its beginning the focus of sociology has rested on ‘social relations’ and this
continues to be the case. Yet it often appears that social relations are not well
understood. Sometimes they are simply ‘presupposed’, sometimes they are treated
as ‘structures’ or expressions of actions, sometimes they just happen, sometimes
they are ‘events’ or ‘communications’. In most cases, sociologists treat them not
as the main focus of analysis, but only as a resultant. It is as if sociology observes
social relations, so to speak, by instinct and intuition, deriving them from other
‘factors’ (individual and/or collective). When more rigorous theories and methods
are introduced, it quite often happens that relations remain a derivative category
and the relatedness of social phenomena remains a background problem. Thus, it
is not infrequent that relations appear opaque and distorted, and consequently are
treated in a reductionistic way.
When relations are assumed as a general presupposition,2 rarely do sociologists
observe them within an epistemology capable of avoiding a form of radical
relativism.3 To avoid relativism (as a form of cultural and scientific contingentism),
both common sense thinking and the bulk of sociology itself still continue to think
in terms of an actor A who is for or against B, and vice versa, and more generally
of individual or collective actors ‘relating’ to each other in the same manner.
Precisely for fear of succumbing to full indeterminacy, the focus is put on A or B
or both and their link, but not on the (qualitative) relationality of their relation. To
speak of ‘structures impinging upon them’ is not a substitute for that.
Relations are quite often only a formal way of describing the ‘state of affairs of
the system’. In other circumstances relations are a means of understanding qualities of particular actors. Alternatively, they refer to functions and to operations of
subsystems and, in general, to characteristics which play a role in a given field. In
I am here referring to the meaning of ‘presupposition’ as suggested by J.C. Alexander (1982)
Theoretical Logic in Sociology, vol. I: Positivism, Presuppositions, and Current Controversies,
University of California Press, Berkeley-Los Angeles.
3 The perspective put forward by Mustafa Emirbayer (1997) is clearly caught up in full relativism
(‘Manifesto for a relational sociology’, American Journal of Sociology, 103(2), September,
certain cases sociology considers the ‘society’ which surrounds us (instead of
being ‘within us’) and which conditions us in such a way that social relations
become an indeterminate force, often perceived in undifferentiated terms and as
something almost mysterious. Conversely, relations become tough ‘structures’
which serve to make society ‘substantial’, as in the covering-law model.
In all of these manoeuvres, sociological thought avoids being confronted with its
own proper object. When different actors in a field are observed, they sometimes
come to be placed in acccord, sometimes in antithesis, and at still other times are
placed on a continuum according to the strength of their attraction and repulsion.
Ambivalence in actors’ (or systems’) identities is the concept which ‘deals with’
these latter phenomena. But a focus on relations as such is absent more often than
not. When relations are evoked, they are presented as a way of describing the
situation, rather than as the ‘foundational’ element of the situation itself. They are
considered as an ‘additional’ element which serves to make the subjects comprehensible. Very often relations remain a derivative element, a subproduct, and a
means of comparison. They are or become only a connection established by the
observer within the ‘real’ terms of phenomena (be they ‘subjects’ or ‘structures’).
Only in certain circumstances and with certain authors do relations become the
object and sometimes even the appropriate and specific focal point of sociological
reflection, in a strict sense. Only then are theorists capable of avoiding reductionism. Only then can they avoid their own contradictions and self-deceptions
with regard to concrete social reality.
Sociology draws from its rooted assumptions to discuss social relations. All the
greats of sociological thought have provided a fundamental contribution to the
understanding of social relations. Nevertheless, today’s situation is characterized
by the fact that social relations still remain the ‘unknown object’, virtually the
terra incognita, of theory and of empirical research, not to mention of practical
(applied, clinical) sociology. And yet the very stuff of sociology, as well as the
historical trajectory which it has taken in and through modernity, is made up first
of all of social relations. The passages from premodern to modern and then to
postmodern eras are intimately related, indeed they are marked by turning points
in the relational sense. These points have yet to be fully understood and elaborated
Ironically, thought becomes more and more ‘relational’, but at the same time,
social relations themselves are lost. This is a paradox, and behind this paradox the
phenomenon of the increasing reflexive differentiation of social relations as such
is hidden. Today, under conditions of globalization, social relations are spoken of
more and more precisely because ‘society’ becomes more problematic, more
uncertain, and more unstable. Yet, the more social relations become contingent
(their existence and their forms imploding and losing their internal as well as
external boundaries), the more sociology seems to enter into a state of recurrent
crisis. That is why ‘relational sociology’ seems to be the best answer to the crisis
of sociology in times of globalization.
The starting point I assume here is that the object of sociology is neither the socalled ‘subject’, nor the social system, nor equivalent couplets (structure and
agency, life-worlds and social system, and so forth), but is the social relation itself.
However obvious it may seem, this statement has far-reaching implications. For
instance, it implies that there is no point in theorizing an opposition between a
‘sociology of the social’ and a ‘sociology of association’ (as does Bruno Latour),
since the social is intrinsically associational (relational). Nor, from this vantage
point, it is plausible to think of society in terms of dualistic cultural codes (as does
Jeffrey Alexander), since culture is also a relational matter. To give a final
example, Boudon’s reading of Max Weber in terms of methodological individualism (rational choice) represents a deep misunderstanding of the Weberian
relational approach to thinking about society.
The category of the social relation in modern sociology
Social relations, as an explicit cognitive category, were born with ‘modernity’. In
sociology they have their own ‘discipline’, as far as the study of their empirical,
factual and phenomenal aspects is concerned.
The characteristic mode of treating social relations in the ‘modern manner’
consists in treating them as an expression of the human individual or collective
subject when acting in roles and in social institutions. The relation is seen as an
emanation, an expression, an establishment of something that remains within. In
this ‘the modern’ remains chained to the anchor of tradition. The presupposition is
that the human person ‘needs’ to have relationships, as its ‘manner of being’, to
regard them as ‘real’, to ‘develop’ them and so forth. Conversely, postmodernity
becomes characterized by the elimination of this field, i.e. it lies beyond the
horizon of cultural traditions.
We can say that a sociological school of thought is still modern to the extent to
which, in spite of all of the radicalizations of modernity (rationalism, empiricism,
relativism, etc.), it preserves the above-mentioned characteristic which makes the
social relation the product or emanation of a ‘subject’. Modernity, however,
‘knows’ that the relation is little by little going to ‘consume’ its own subject. In this
it has a tragic consciousness and yet it remains faithful to itself as soon as and
insofar as it moves on to the assumption that ‘something must happen’, through
which the subject, whether it be social class for Marx, the charismatic personality
for Max Weber, or the corporation for Durkheim, will be able to maintain control
of social relations. Everyone acknowledges that sociology is born and still persists
on the basis of its appeal to legitimacy as the science for the anchorage, the
integration, in brief, the regulation and control of social relations.
No attempt is made to recall all or a selection of the principal authors of relevance
here. On the bases of the criteria mentioned at the start, all who think of social
relations as the product of some ‘factor’ (e.g. the material infrastructure of society,
the division of labour, culture as collective consciousness or individuals
themselves), are ‘modern’. What is implicit in this analysis is that, when one factor
is chosen over others, this implies an unacceptable reductionism. Thus, any
sociology having recourse in the last instance to material factors, such as Marx’s,
loses the subject. Any sociology having recourse in the last instance to individual
factors, such as Weber’s, loses the autonomous non-individual dynamics of action
with their internal (non-subjective) determinations, their collective logic and order,
and unintended consequences. Any sociology which has recourse in the last
instance to functional factors (such as Durkheim’s division of labour or functional
differentiation) loses subjective meanings, while any sociology having recourse in
the last instance to cultural factors (such as the late Durkheim) loses those structural
aspects which are independent of human representation or consciousness.
Modernity had ‘freed’ the social relation, but in the early phase of sociology it
remained, so to speak, ‘anchored’ to various nonrelational factors, for example to
the labour theory of value, or to utility corresponding to needs, or to reason, or to
basic social rules. As classical sociology matured, the more it was realized that
such an anchorage was difficult if not impossible to maintain. Having long
remained implicit and almost, as it were, in gestation, the relational watershed was
affirmed as the central theme in the sociology of Georg Simmel.
Georg Simmel’s ‘relational turning-point’ can be considered the very beginning
of a proper relational theory in sociology, although it cannot stand alone and must
be revised and supplemented with other relational visions, above all the Maussian
theory of exchange.4 In the course of sociological thought from modern to
postmodern, there are thinkers who have expressed a particularly profound and
even dramatic consciousness of the relational turning points which were going to
consume them. Simmel is one of these, perhaps the greatest. With him it was
understood for the first time that the reality of what we call ‘social’ is intimately
relational in a sense which, in principle, could not be reduced to the partial
determinations of the above paradigms (Marxian, Weberian, Durkheimian).
Unfortunately, however, Simmel elaborated a theoretical approach that was
predominantly formal in response to his intuition that society is (not ‘has’)
relations. As is well known, the fundamental theoretical category for Simmel was
the social relation. According to him this category should be understood as
interaction. The importance of Simmel’s theory, in which this great thinker
expressed the tragic consciousness of the drama of modernity, still remains to be
For Simmel, a phenomenon is social to the extent that it expresses a particular
and sui generis character that is its ‘constituent’: namely, the fact of being an interrelation, or inter-dependence, or better still a reciprocal effect or an ‘effect of
reciprocity’. All these terms are included in the concept of Wechselwirkung. For
Simmel the social phenomenon is neither a priori an emanation from a subject,
nor a product of an abstract system. The social is the relational as such, that is,
reciprocal action, inasmuch as it produces interaction, is incorporated and
manifested in something that, even though non-observable, has its own solidity.
Unfortunately, in Simmel this ‘solidity’ is not sufficiently well clarified. Instead,
the category of Wechselwirkung becomes a metaphysical principle.
See Alain Caillé (1993) La Démission des Clercs. La Crise des Sciences Sociales et L’oubli du
Politique, La Découverte, Paris.
From my own point of view the fundamental problem that Simmel’s relational
sociology raises is the fact that it treats relations as ‘events’ or pure ‘emergent
phenomena’. In addition, he lacks a critical distance, when, for instance, he sees
money as the formal, substantial and functional prototype of generalized social
relations. Nevertheless, Simmel carries out an operation of enormous significance
and importance when he elaborates a relational theory of society, building it on
the definition of social objects as ‘substantialized’ social relations by using the
standard of money as its basic paradigm. In fact, with this, he elevates to the
foremost general presupposition of sociology the fact that modernity coincides
with the transformation of concrete entities, of any entity, into purely abstract and
Such a perspective, despite their different frameworks and orientations, was to
be taken up later by Parsons, with his theory of money as the paradigm of the
generalized symbolic means of interchange, and later still by Luhmann with his
theory of the primacy of systems-with-one-function.5 Parsons and Luhmann,
willingly or not, are followers of Simmel precisely on this point. They take up the
Simmelian assumptions in very different manners, but nevertheless decisively,
and they are yet more radical than Simmel in seeing modern society as being
‘relational’ with money as its symbolic medium.
Simmel still looked for a ‘logic (or a grammar) of social relations’ internal to the
social domain, thinking that social differentiation could make them more and more
autonomous. His formalism must be understood, not so much or merely as a
separation between the form and content of the relation, but rather as a heuristic
instrument with which to understand that the ‘game’ of social relations has rules
and nonarbitrary dynamics.6 Money, as a ‘substantialized social relation’ is only
one of its paradigms.
The Parsonian attempt to ‘systematize’ social relations.
The apex of what I have called modern thought on social relations can be found in
Parsons. He attempted to synthesize a great part of the principal acquisitions from
the classical sociologists and at the same time to channel the germs of dissolution
threatening modern society. In performing this operation Parsons ‘closes’ the
social relation in the sense of making it into an element of the general system of
social action, intentionally leaving aside the thought of Simmel and the problems
From the beginning, in developing his theory, Parsons associated himself with
those whom he selectively designated as the ‘classics’ in sociology because they
could usefully be presented as ‘converging’upon his theory of action.7 Because of
this, Parsons condemned himself to ground social relations on presuppositions and
Cf. N. Luhmann (1987) ‘The evolutionary differentiation between society and interaction’, in J.C.
Alexander et al. (eds.), The Micro–Macro Link, University of California Press, Berkeley, p. 116.
6 Cf. S. Moscovici (1988) La Machine à Faire des Dieux, Fayard, Paris.
7 Cf. T. Parsons (1937) The Structure of Social Action, Free Press, New York.
concepts inappropriate to the changes that were taking place within the society of
the twentieth century.
Let me synthetically recall the nodal points of Parsonian thought on social
relations. For Parsons the social system is a patterned collection of status-roles
which imply certain courses of action. The unit-act, not the relation among atoms,
is the basic element from which more and more complex social organisms are
formed. As is widely recognized, organic, biological and chemical–physical
analogies are ineluctable in interpretating the thought of this author. This means
that from the sociological point of view the theory of social relations becomes a
product of the theory of action and a correlate of the theory of status-roles8 within
an epistemological framework strongly conditioned by both neo-Kantian and
positivistic presuppositions. In the end, for Parsons the social relation is the
reciprocal action of actors in a social system, but, insofar as the actor disappears
from the scene,9 the relation becomes the product of the AGIL scheme, i.e. a
product of social structure.
Although Parsons, as a matter of principle, would have been able to adopt the
point of view through which roles are created by actions, he never began to get to
the very bottom of this principle, since he would have opened up for himself an
unacceptable circularity. For Parsons, status-roles have a Durkheimian flavour,
that is, they descend from and are products of collective consciousness as
expressed in the division of social labour. It is by means of this approach that the
social relations that constitute the system come to have a predominantly normative
connotation in Parsons. Relations are the reciprocal actions of socialized
individuals in given modes according to determinate status-roles.
Parsons’ analytic schemes of both social action (with its components: actor,
purpose, means, conditions and norms) and of social systems (the AGIL or LIGA
scheme: latency, integration, goal-attainment, adaptation)10 can be considered as
heuristic instruments of exceptional importance for the relational perspective.
Clearly I cannot recapitulate or summarize the vast literature which has been
written on these themes. What I want to assert is that through these schemes and
their more or less implicit ‘theoretical logic’, it becomes possible to understand
how social relations can be involved within a single action (unit act) in co-relation
with what occurs in the broader social system encompassing it. This possibility is
exactly what Parsons opened-up, but at the same time failed to explore in depth.
The fact of not having problematized the association between the relations
internal to a single unit-act (or system of action) and the relations among the
8 See R. Ruddock (1969) Roles and Relationships, Routledge and Kegan Paul, London; S. P. Savage
(1981) The Theories of Talcott Parsons: The Social Relations of Action, Macmillan, London. The
fact is worth noting that if one glances through the subject indices of the work of Parsons and of
his commentators one almost never finds an explicit reference to social relation as such.
9 See P. Donati, Teoria Relazionale della Società, op. cit., ch. 4.
10 For a simultaneously critical and sympathetic presentation of such schemes, see J.C. Alexander
(1984) Theoretical Logic in Sociology, Routledge and Kegan Paul, London, vol. IV; J.C. Alexander
(1987) Twenty Lectures, Columbia University Press, New York, chapters 2–6.
different elements (or sub-systems) that compose the unit-act (or system of action),
not to mention their transcendent effect,11 that leads Parsons to neglect their
interactions (i.e. relations as ‘emergent effects’). Consequently, in Parsonian
theory the relations internal to the unit-act (or social system) are seen in a strongly
normative way. They are viewed as secondary and even derivative in relation to
the ‘system’, which includes them at a higher cybernetic level. In brief, not having
developed the theme of the significance of social relations as such, leads Parsons
to fall into both reductionism and (downwards) conflation. Later on these were
brought to light by anti- and post-Parsonian sociologists.
The Parsonian theoretical framework is unbalanced in the direction of
‘systematizing’ social relations to the point of rigidifying the relational logic as
such.12 To go beyond his bold venture, it is not enough to introduce more contingency into the social world, i.e. its two-tiered reality of order and social action.13
The idea of merely enlarging social contingencies is too simple once one considers
the growth, albeit in appearance, of the ‘de-normativization’ of the social within
The failure of Parsonian theory is found precisely in the fact that he attempted
to ground modernity’s success on the institutionalization of social relations. This
is something that cannot be within the limits of the conception, or symbolic code,
of these very relations within modernity itself.
The category of social relations in postmodern sociology
From a theoretical point of view, postmodern sociology begins when social
relations are considered as products of themselves, or else as an end in themselves.
To me, such an approach is no longer relational, but can more properly be depicted
Marx is undoubtedly a forerunner of this position, but only in some aspects, and
always within the framework of his materialistic utopia, which envisages capitalist
relationships as the way out of ‘prehistory’, tending towards the end of the division
of labour and the revelation of ‘true’ history in the communist society: a framework which contrasts empirically with the post-Marxian historical experience.14
A position that is equally ‘relationistic’ can be found in all those modern and
11 Cf. J. Bartelson (2000) ‘Three concepts of globalization’, in G. Therborn (ed.), Globalizations are
Plural, International Sociology, special issue, 15(2), June, 180–96.
12 In this I distance myself from theorists, such as J. Alexander, who take up Parsonian theory only
in the sense of increasing the contingency of the social world (order and action) without focussing
on the central analytic theme of social relations as such.
13 See J.C. Alexander, Twenty Lectures.
14 It goes without saying that, in building social theory, Marx used not one but three heuristic models:
base–superstructure, organic totality and dialectical development (see: L.J.D. Wacquant (1985)
‘Heuristic models in Marxian theory’, Social Forces, 64(1), September). The epistemological status
of what I call ‘social relation’ is of course different in each of them (see P. Donati (1998) Lezioni
di sociologia, Cedam, Padova, ch. 1).
contemporary sociologisms which have made the social relation an eschaton (or
ultimate end) in itself, but always with little awareness of the epistemological
premises and of the consequences that a radical relational assumption is going to
Today this position can be found amongst those scholars15 who hold that the
relation would become an end in itself insofar as, in their opinion, it can dispense
with the subject, all the more or less institutionalized collective entities, as well as
normative rationality. But this is not the point. For the majority of postmodernists,
neither society nor theory can eliminate the subject. On the contrary, to them,
postmodern society emphasizes the irreducibility of the subject as ‘mind’ or
‘consciousness’ or the ‘environment’ of action and of the social system. In my
view, on the whole, postmodern thought signifies the contemporary exaltation and
dissolution of social relations. However, this opens new horizons for their
understanding, as the key concept for the explanation of the subject and of social
In the postmodern, the dialectic between subject and institution, or between
agency and structure or the like, follows the simple unfolding of relations as such,
in general conceived of as a flow of communication. The symbolic code which
interprets this process is predominantly communicative: relations are subsumed
and, indeed, reduced to communications and only to communications.
It is obviously not possible to provide a detailed account here of these processes
and of the thought of many authors in terms of ‘relationism’, that is, of absolutizing
and radicalizing social relations within their own self-referential circularity, which
usually ends up with the reduction of the reality contained in social relations to a
mere abstraction. In order to give the reader some examples of what I mean, two
versions of ‘relationism’ could be singled out as particularly representative or ideal
typical of this sociological attitude: (i) J. Baudrillard, or the implosion of social
relations and their exaltation as pure simulation; (ii) N. Luhmann, or the exaltation
of social relations as pure communication.
Therefore social relations come to be characterized by the following features.
From the point of view of the social order they are subject to radical contingency.
They are subject to the systemic code of more and more radical functional
differentiation with the loss of reference to institutions and social structures.
From the point of view of action they are reduced to communication and only
to that, which assumes the irrelevance of every cultural tradition.
15 I am thinking, for example, of Marshall Berman (who has adopted the slogan of Marx’s Communist
Manifesto: see All That is Solid Melts into Air. The Experience of Modernity, Simon and Schuster,
New York, 1988 and Mustafa Emirbayer’s ‘Manifesto for a relational sociology’ (American
Journal of Sociology, 103(2), September 1997, 281–317), which reduces the category of relation
to a mere transaction, so that, within this perspective, the units of analysis derive their meaning,
significance and identity from the changing functional roles they play within a given transaction.
In one respect it is necessary to recognize that the sociology of Baudrillard, as
much as that of Luhmann, contributes powerfully to the proper illumination of the
autonomous object of sociology, that is, the increasing autonomy of social
relations in society and the effects of such a trend. However, it is also necessary to
recognize that such insights are partial and, as subtle and seductive as they may
seem, in the end they furnish an image of social reality that can brutally distort the
meaning of the human condition, even within postmodern society. For example,
there is no doubt that today generalized social relations are becoming full of
pretence and that they autonomize the communicative aspect beyond any threshold
ever known. Yet, social reality is not only this or is not properly this.
About ‘reconstructive’ attempts
After the classical theorists and within the great postmodern river there exists a
broad, even if very heterogeneous, sociological current aiming at reconstructing a
theoretical framework in which and through which one could achieve a positive
view of the social relation. I will only mention a few instances with the sole
purpose of outlining and labelling paradigmatic lines of study (Habermas, or the
normative components of relations; Giddens, or the ‘structuration’ of social
relations; Alexander, or the hermeneutic components of social relations; Archer, or
the morphogenesis of social relations which is a critical view of the other three).
However, all the evidence suggests that they are, so to speak, riveted by a
particular aspect of relations and thus remain ‘on this side of’ the postmodern
challenge, which was anticipated by Simmel and literally ‘exploded’ with
If one truly wants to accept the challenge and thence to understand the
‘relational society’, which is being born under our eyes it is necessary to place
relationality at the level of a first general presupposition in the metaphysical
environment of theory.
Luhmann grasped the point, since he understood the need for a relational
perspective, but unfortunately he stopped there: at the paradox according to which
present society removes the social relation as such from the scene and therefore
one has to abandon its explanatory power.16 Where I believe he is mistaken is in
the act of reducing the concept of the social relation to a mere (logical) reference
between elements in society, so as to render the social world contingent in an
absolute manner (according to him, the social relation is only a communication
which makes reference to another communication, whereas these events exist only
in relation to each other, without any residue). From this, extremely reductionist
consequences follow, amongst the most important being the radicalization of social
differentiation as the engine of history and the location of the functional method
(which is only a method) at the presuppositional level (i.e. at the boundary with the
16 See G. Teubner (2001) ‘The economics of gift – positivity of justice: the mutual paranoia of
Jacques Derrida and Niklas Luhmann’, Theory, Culture & Society, 18(1), February, 29–47.
metaphysical environment of the theory). This has disastrous effects on theory and
on the capacity for understanding empirical instances.
Nevertheless, it is otherwise clear that, unless the assumption of bringing
relations to the presuppositional level (in the epistemological sense) is adopted,
the complexity of society which emerges behind the thrust of new symbolic and
relational communicative codes cannot be treated adequately.17 For example, it is
clear by now that the reasons for the emergence of so much social pathology today,
and so much cultural regression as well, cannot be understood and explained if we
maintain that the ‘modern system’ is substantially stable and progressive in terms
of its symbolic and communication codes. Today such a system is capable of
meeting requests to regulate generalized social relations, to confer subjective
meaning upon social structures and institutions, to satisfy needs for empathy, and
so on, only on the condition of developing ‘other’ relational codes which can deal
with such requests, needs and the social responses to them. The lack of these new
relational codes is the reason why there is less and less capacity for empathic or
hermeneutic meaning, or apparently even the need for it, whether it is or is not
shared. In this rests the apparent power of the ‘global system’ and of its own
The reasons are obviously multiple and complex, but in synthesis they are
related to the fact that ‘global society’ does not seem to be able to generate or
tolerate complex solutions and thence precedes collectively to ‘wreak havoc’
amongst those subjects or about those problems which are needful of complex
solutions which cannot be found. Only an adequate theory and relational praxis
can illuminate the situation in an uncynical manner.
‘Relational thinking’ in sociology: an epistemology (based
upon a social ontology of the relation), a paradigm (society as
a web of relations) and a pragmatic (networking)
In sociology what is implied here in the passage from a simple to a ‘complex paradigm’, as Morin defines it, is a radical change in the ontological, epistemological
and phenomenological status of the social relation in science and in society.
To speak in a relationally adequate manner means: (a) first of all, that social
relations should be accorded the status of an ontological ‘stratum’, that is a level
of sui generis reality embodied in ‘social facts’, and (b) that every sociological
object must be defined as a social relation in a sensible manner, since sociology is
17 In general, by ‘symbolic code’ I mean a set of ideas and/or images used to represent something the
recipient will recognize according to certain rules (see Chapter 3, footnote 11). A ‘communicative
code’ is a set of rules concerning the way communications can and must be treated. Luhmann
defines the code in a very strict and abstract way, as a ‘rule of duplication’ by which a term can be
correlated to a correspondent entity which is the negation of the first one (for instance, if I say ‘it
rains’, the code implies that I am denying the fact that ‘it does not rain’). As I will clarify later on,
we need to see ‘relational codes’, which mean symbolic and communicative codes that work
through symbols and rules that are relational (not binary etc.) in kind.
concerned with the social relation ‘embodied’ in the object itself, insofar as it is a
social phenomenon, i.e. arises, exists and is transformed socially. These points
must be discussed in greater detail in order to enter more deeply into ‘relational
The proper and sui generis reality of social relations
To maintain that social relations have a reality of their own (sui generis) means
saying that they are not simply derived from something else, but reflect an order of
reality of their own with internal dynamics that require theoretical–practical
conceptualization. This order of reality is not derivable from or reducible to this or
that particular factor or variable (for example, ‘power’, as Elias maintains), but
derives from the relatedness itself of the social order.
This relational ontology has, so to speak, an empirical foundation that can be
known, in part, experientially. In the system of organic reference (body), a person
cannot exist without air and without food, but the former cannot be reduced to the
latter. In the system of social reference (social forms) the human being cannot exist
without relations with others. These relations are what ‘constitutes’ his or her being
a person, and are like the air and food necessary for the body. Were the relation
with the other to be suspended, so too would be the relation with the self. This,
and nothing else, is what sociology deals with.
Certainly we do not see social relations, so to speak, ‘going for a walk’. Yet we
know that they exist and are relatively lasting, not simply because they are
concretized in movements and social institutions, but because we have direct and
interior experience of them. That the existence of any particular relation is a matter
of contingency is no good reason for saying that they have no reality. For example,
eye-colour is a contingent characteristic, but is nonetheless real. Indeed, sometimes
contingent characteristics are more decisive than fixed characteristics.
What is this reality? It is a reality of a relation constituting the object (conventionally called ‘the social fact’ in sociology), which is not merely psychic, not
merely in the mind of the observer, but a reality which requires an appropriate
theory for the process of observation insofar as this is capable of situating
itself within different systems of reference, some of which would withhold
ontological status from relational reality altogether (for example, a constructionist
The conclusion drawn from the above is that to adopt the relational outlook
means to be dealing with a non-observable but equally real level of reality, where
the relation is the tertium. One must be located in this system of reference if one
wants to avoid the possibility that epistemic relativity would be transformed into
relativism. This tertium has the same reality as that which constitutes our identity
as a community, plural and pragmatic. Although such realist arguments have
provoked much misunderstanding, relatedness exists not only at the social level,
but also in the interconnections among the other levels of reality (biological,
ethical, political, economic, etc.).
Sociology as the relational definition of objects
Relations are not only a medium. They are also the point of view from which
sociology can grasp its objects of study in their social nature. Every sociological
object can and must be defined in relational terms. It is inexact to say that
sociology studies relations among social realities (according to the expression
rendered classic by Pareto at the start of his Treaty of General Sociology). Rather,
it studies social realities as relations. To do this it must redefine its objects and
then its concepts as relations.
When, at the beginning of a study, we state the problem (‘why does it happen
that X?’), it must never be forgotten that the phenomenon under investigation is
derived from a relational context, is immersed in a relational context and brings
about a relational context. The greatest error of the phenomenological approach
is to place social relations in brackets, and then to try to reconstruct them as
products of intersubjectivity, something which can rarely be done.
The full acknowledgement of these relations as starting points is what distinguishes the sociological approach from the pure phenomenological approach.
The latter places social relations in parentheses in order to allow the operations of
the transcendental consciousness (or ego) to emerge. We need not deny the
fruitfulness of the phenomenological approach when it is used to examine a
suspended moment and as an indication of methodological recursivness.
However, the sociological point of view must hold tenaciously that it is the social
relation which provides the ‘key’ for entry into and exit from its subject matter.
In sociology, the phenomenon we want to know about is the social relation and
not the subject or ego as such. Elsewhere, I have sought to furnish some examples
of applying this procedure to themes such as health and chronic illness, the
family, the welfare state, citizenship, civil society, the third sector18 and free
In understanding social reality and subsequently intervening in it, it is important
to clarify what specific symbolic code is being used and what type and level of
relatedness this implies. In a general sense, to understand increasing social
complexity it is necessary to have codes which are capable of keeping this
complexity open to selectivity (i.e. choices, or even reductions) broader than that
which operates in terms of binary codes (of the type either. . .or). This has at least
the following requirements.
18 For more details on these studies see P. Donati (2000) ‘Freedom vs. control in post-modern society:
a relational approach’, in E.K. Scheuch and D. Sciulli (eds.), Societies, Corporations and the
Nation State, Brill, Leiden, The Annals of the International Institute of Sociology, 7, 47–76.
19 See P. Donati (2003a) ‘Giving and social relations: the culture of free giving and its differentiation
today’, International Review of Sociology, 2, 243–72. This paper has to be understood as a further
development of the basic paradigm of the gift put forward by Alain Caillé (1996) ‘Ni holism ni
individualisme méthodologiques. Marcel Mauss et le paradigme du don, in L’obligation de donner.
La découverte sociologique capitale de Marcel Mauss’, Revue du MAUSS, n. 8, 2° semestre,
At the epistemological level, it requires a symbolic code that not only refers
to the poles of the relation (relata), but concerns the relation in itself, as a
mediation or mediator that is irrreducible to the terms that are connected.
At the level of the explanatory paradigm and of pragmatics, it requires
appropriate analytical network models to undergird social interventions that
can show not only the contribution made by the particular elements of the
relation, considered in isolation, but also the contribution of the relation
(interaction) viewed as an ‘emergent effect’,20 in order to be able to investigate
and treat the social system in question as ‘condensing’ the underlying social
An example may help to clarify what is being affirmed here. Let us take the
common problem of knowing what the social reality of the family is. The only
possible way out of the current dissention and disorientation in sociology about
what the family ‘is’ depends upon understanding the proper and sui generis reality
of family relations, namely formulating them in fully relational terms. Such a
definition must be oriented to understanding them as a specific form of symbolic
exchange, operating simultaneously between genders and between generations.
However, in what does the ‘relational reality’ of the family consist? Do we see
the family ‘going out for a walk’? Certainly, we do not. However, to take a banal
example, we would obviously be able to see a man, a woman and children going
on a walk. If we already know them as the White family, we would say that it is the
White family who is going on a walk. If we do not know the White family, we can
think of a certain number of finite possibilities of intersecting biological and social
relationships among these persons who are walking together. The judgement as to
whether or not they constitute a family cannot be certain and must be suspended.
But, were they indeed to be family members, the problem remains – are they ‘the’
family? If another son, or the grandmother, or another person normally living
under the same roof is missing, what would or should we say?
In reality we see persons and, having posited or supposed certain relations
among them, we say that this is the White family which, complete or incomplete,
is going on a walk. The presupposition is therefore of the existence of a certain
relationship which connects the elements that we observe. We see persons, but we
think in terms of relationships and we speak on the supposition of relationships.
The word family indicates relationships: it is a relationship, not a place or
something like a locale. All the language that we adopt to describe what we ‘see’
beyond single individuals is essentially relational. Words have meaning only if
they refer to relations: in a sense, words are relations. All our thinking processes
20 See W. Cook and A. Dreyer (1984) ‘The social relations models’, Journal of Marriage and the
Family, 46(3), 679–87.
21 I have atttempted to clarify in what sense one can speak of ‘social systems as condensed in the
network’ in P. Donati (1988) ‘Zwischen ‘Gemeinschaft’ und ‘Gesellschaft’: Die informellen Netze
in der gegenwaertigen Gesellschaft’, Soziologisches Jahrbuch, 4(I), 249–76.
are relational: they connect and refer through relations, and to this extent they are
served by the mediation of language, which is a great collective, symbolic, and
On the other hand, it is also a matter of experience that language does not
resolve questions about reality. If I see John and Clara White and one of their
children going on a walk and say that the White family goes on a walk, this does
not imply that what I claim to be the White family is only or usually what I
observe, think and describe linguistically. Beyond the fact that there is another
member of the family who cannot be with them, what I see certainly cannot
include everything that the White family is. In other words, my perception of
reality is constructed and expressed as a relationship, but is not reducible to
language: language cannot express all the reality of the relationships involved.
Relations have the characteristic of always referring to other relations. Thus, the
meaning of a meaning is also a relation. This is also the case because relations can
only come into existence and be known through the mediation of symbols which
are ‘collective images’.
With this, however, the question of what that relationship which we call the
family really consists of and what we can attribute to the group of X persons whom
we see, has not yet been answered. In the first instance it consists in the fact that
the terms symbolically related by the observation (= relata) are or represent
‘something’ to each other. This ‘something’ is not fixed once and for all. But it is
necessary if one wants what is perceived to be deemed to be a relation. If it is not
‘that’ family, it would be ‘another’ family, but that does not prevent us from asking
them who they are. Is this ‘something’ only a subjective interpretation or is there
an objective entity that is established through an intersubjective agreement?
Therefore, the relationship that we call a family is not only a product of
perceptions, sentiments, and empathy, but is a fact which is both symbolic (‘a
reference to’, i.e. re-fero) and structural (‘a bond between’, i.e. re-ligo). As such,
it does not depend on the subject even though it can be actualized (‘live’) only
through subjects. It is in this activity dependence that the relation assumes its
particular sense. Nevertheless, the individualized processes by which we perceive,
sense and mentally imagine, even where there is creativity, cannot begin without
the aid of what we share with others.
The definition of the family, thus breaking out of the hermeneutic circle, is
rooted in the requirements of the relation itself. It amounts to understanding the
extent to which a relation is taken by its subjects as being a sui generis relation
and not as something else. From a sociological point of view, the family is not
ultimately what the interacting subjects or an inherited culture define as being the
family. Nor is it an a priori presupposition. It is that which corresponds to the
requirements of a particular type of interaction, or relational system, which must
render possible the full reciprocity between genders and between generations
within the life-world. The family is a relation sui generis, which must be recognized as such in a given context.
In this way, indeterminacy and infinite regress are avoided. Circularity is
interrupted by the relationship itself (ontological statement), when it is considered
to be what it is (epistemological statement), i.e. as a relationship, and not as
something else. The ontological fact can exist separately from the epistemological
one, but, if the latter is added, then the relation acquires a specific reflexivity (as I
will illustrate with reference to the ‘relational reflexivity’ implied in relational
reason). To be in a family relationship is not the same as to be a patient in relation
to a doctor, or an employee in relation to the employer, or a friend in relation to a
fellow, or a customer in relation to a shopkeeper, and so forth.
The resolution of the two above-mentioned problems, placing social reality at
the level of reality sui generis and defining the object in relational terms, today
involves an epistemological turning point to which there corresponds a paradigm
and a consequent pragmatic. Here I will limit myself for reasons of space to a
few propositions which are meant only to delineate a more general conceptual
framework representing a useful turn towards a relational sociology.
The most general presupposition of sociological thought can be expressed in the
slogan: in the beginning there is the relation. Social processes, with all their
various characteristics, proceed through, with and across relations. This can be
said about social reality and also about theory. In other words, the relational is the
‘start’ of social reality, both in theory and in practice. A relational perspective can
be developed all along the continuum of science stretching from the theoretical
dimension to empirical facts and vice versa, in a continual process of reflexivity,
including different methodological and intermediary phases, which are relatively
To carry the relation to the level of a primary, general presupposition in the
metaphysical environment of knowledge does not imply in any way assuming
the absolute contingency of the social world. Similarly, it does not imply any
accommodation to a type of ontology which denies the human being or social
subject. On the contrary, it means assuming that the relation has its own noncontingent ‘root’, whilst it unfolds in contingency. Obviously, such a ‘root’ is
found beyond a given society. It goes beyond concrete social phenomenology.
The relational methodological paradigm
Society, in this approach, becomes understood according to a paradigm which is
neither that of the whole and its parts, nor that of system/environment, nor that of
autopoiesis, but that of the social net. Society is understood as a net or web of
relations. Society is a relation of relations, which unfold themselves according to
their relational symbolic codes, a term that can be also interpreted as a ‘logic’,
though not of an instrumental rational kind alone.
It is important to understand the relation between the concept of a social network and that of a social system. The first is broader than the second, not vice
versa. Systems are a sort of ‘condensation’ of networks in the sense that networks
conduct, are conductors of, more reality than is possible for communicational
networks, with nodes, density, functionality, connectiveness and other ‘systemic’
Here the current division between structuralist analyses of networks and
cultural-communicational analyses comes to be rethought, precisely in relational
terms. What have until now been paradigms that were ambivalent, dualistic,
oppositional, complementarity etc. need to be reformulated according to a relational logic of social webs.
The practical implications of such a relational sociology can be conceived
and structured in accord with network analysis and result in approaches that can
broadly be called ‘networking interventions’.
The basic idea is that the operation of a sociology, which is required as a support
for practical or clinical action – for example in social policy and social services –
must clarify that:
subjects and objects do not exist in an isolated state, but as complex relational
webs in which subjects and objects are defined relationally, auto- and allopoietically. This point does not mean that they are defined in a relativistic
manner, as if every matter could be understood or modified at one’s pleasure.
The problem of relativism is resolved by specifying the relations among the
different systems of reference, or by specifying the variables characteristic of
nontrivial states of the system, which are used for analysis;
when one intervenes with respect to a phenomenon, it is necessary to work
on the relational web in which what is observed is maintained, that is, to
consider other relevant subjects and surrounding objects plus the ‘effects of
the network’, which the actions can involve;
such interventions must be conducted with an awareness that a relatedness
exists between the observer and the observed, between the actor and the acted
upon, which entails the double hermeneutic.
Obviously, there are multifarious levels at which the above can be taken into
consideration, become known and sometimes measured, and finally be realized in
practice. What is crucially important is that options for intervention that are a priori
reductionistic, or of a reductionistic character, should not be legitimated.
Summary: a program of work
What we want to know are social ‘facts’ insofar as they are real. However, we
cannot know them except in and through relations:
the relation is the key to entering reality and to leaving it;
the relation does not eliminate the terms which it connects; instead, it
reclaims, explores and expresses them;
the relation is not a pure abstraction, that is pure form or pure communication,
but is a concrete reality;
it goes without saying that only in extremis can such a concrete relational entity
be thought of as dichotomized (i.e. ambivalent, dual) or fuzzy. Normally, it
has a network structure: it connects, bonds and creates interdependencies, but
tensions, conflicts, and also contradictions stem from it too;
norms and rules are an absolutely necessary and inevitable mode through
which to regulate (i.e. under non-extreme conditions) the contingency of
situations which, in the social world, are not determined a priori.
In brief, the relation, not the duality or the ambivalence or anything else, is ‘the
game of games’. It is curious to observe how sociology attempts to respond to
problems without having awareness of how much it affects the phenomena that it
seeks to understand. Such a reflexive and self-observational capacity requires the
adoption of a non-reductionistic, i.e. relational stance.
The relational paradigm
Its implications for the
understanding and organization
The underlying issue: when the ‘social’ is no longer ‘human’
Today, the feeling is that society is becoming increasingly fragmentary, uncertain,
fluctuating, liquid, estranged and manipulated. In other words, it is decreasingly
‘human’, if not inhuman,1 in its everyday social relationships. Social problems are
increasing, whilst solutions are receding. In the Western world, especially in
Europe, there is a feeling of deep unease and increasingly vocal expressions of
dissatisfaction. This issue has been discussed by most major contemporary
sociologists, such as Niklas Luhmann, Ulrich Beck or Zigmunt Bauman.
Empirical research confirms this unease to be largely the case because it is
reflected by objective indicators, both as far as institutions are concerned (reputed
to be incapable of responding to people’s problems) and also for inter-subjective
relations (perceived as being increasingly volatile and devoid of meaning). In
stressing this issue, it is not denied that there are also signs of great vitality and
regeneration. Nevertheless, such ‘happy areas’ are usually under siege and increasingly difficult to develop and to cultivate. What is happening?
Communitarians deplore the growing dysfunctionality of the community.
Liberals denounce the demise of the individual as the centre of free and responsible
By ‘human’ I mean that ‘which is distinctive of the human being’ in his/her way of existing, living
and acting; and by ‘social’ I mean that ‘which takes place between individuals or results from their
interaction’. Human and social are therefore considered as essentially dynamic ‘orders of reality’.
It is appropriate to add that the concepts of ‘human’ and ‘non-human’ are used in a descriptive and
not in a normative, sense. In other words, they are not moral evaluations or judgements, but refer
to realities which are respectively specific or not specific to humanity. The non-human is therefore
understood simply as that which does not belong to (is not species-specific to) humankind: for
example, the animal and vegetable worlds or those of machines and artefacts are non-human. (This
is the meaning of a sentence such as, ‘Purely instinctual sexual intercourse is not human but animallike’.) When we refer to something which – whilst deriving from human agents – denies or distorts
the human, we use the term in-human (for example, when referring to an act of violence or
aggression we say: ‘it is in-human to treat people this way’). From the normative point of view
therefore, the human/non-human distinction means that only the human (and in-human) fields are
matters of ethics, whilst the non-human remains a purely descriptive concept of a reality, which
does not allow of ethical judgement in itself, that is, beyond humankind’s treatment of it – which
is an ethical judgement about human behaviour and not about its victims.
The relational paradigm 21
decision-making. Socialists draw attention to the loss of solidarity and robust
social cohesion. To solve these problems, is it sufficient to call on people to be a
little more humane (to show unity, tolerance, a willingness to be peaceable etc.) in
their personal relations? Many believe so and stop there. Others demand structural
reforms with the aim of inducing people to behave pro-socially. These are like
dogs chasing their own tails, because the more they work to implement their social
reforms the more noticeable is the stridence of individualism and privatization,
with all their negative consequences, namely, the further loss of solidarity and
significant social bonds.
It is in this manner that the majority of modern sociologists describe the human
condition in the West. There is little hope for positive outcomes (in the sense of
these becoming more ‘human’), but only uncertain lurches towards a ‘post-human’
This underlying problem of widespread unease can be encapsulated by saying
that ‘social life’ gradually seems to be losing those qualities that until recently
were attributed to it, the very characteristics that we usually associate with
humanity, with social relations that are also the expression of a person and of their
subjectivity (feelings, will, reason, and including intentionality and rationality).
The warmth of family relations, among couples or between parents and children,
diminishes every day. Similarly, professional work (assuming it can be found) is
no longer a source of significant human relations. Even school is no longer a place
where children can make friends. But, some will say, has it not always be like this?
Is it not the case that in every era some have always criticized each epoch for its
decadence and dehumanization?
The answer is yes and no. It is, indeed, a common theme, as in Plato’s censures
on the decadence of Athens. However, today matters are different because they
are no longer simply instances of dehumanization, but of an irruption of the
inhuman into the social, one that progressively displaces what is still human. The
radical change lies in the emergence of a ‘social’ which, because separated from
the ‘human’, disallows ethical judgment (or, as Luhmann would say, all that is
socially ‘possible’ becomes acceptable and accepted practice from an ethical point
In brief, the epochal change that we are witnessing represents an emerging
society characterized by the fact that the ‘social’ is no longer seen, heard, or acted
upon as something immediately human, and yet it acquires a personal relevance,
unknown to our predecessors. For example, many young people perceive work as
a provisional relation, a precarious and unstable one, devoid of ‘humanity’
(Sennett 1998). The family is configured as a construct, according to choice, and
with this it loses the characteristics of a deep inter-subjective relation based on the
encounter of a man and a woman and on reciprocal exchanges between generations. Family and work become human only under certain conditions and at
specific moments. Many parts and aspects of social life are no longer immediately
human because they are automated or assigned to autonomous organizational
mechanisms, becoming human only under certain conditions and in certain
contexts. These ‘conditions’, ‘moments’ and ‘contexts’ express the time and space
22 The relational paradigm
to which the social is confined. It is time and space in which new distinctions
emerge, with their own symbolism, when and where it has to be determined if that
work, that family, that way of being and of being social is human or not.
In conceptual terms, we are facing a new, macroscopic historical process
of differentiation between the ‘human’ and the social. That which is ‘human’
becomes detached from the social (or, if preferred, the social is separating out from
the ‘human’). This implies considerable problems because the social sciences do
not have an appropriate frame of reference from which to understand and manage
this differentiation in order to deal with it.
The reason for this inability (which explains why social science diagnoses only
uncertainty, risk, fragmentation and instability, that is, problems that cannot be
resolved, but which can only generate new problems) derives from the fact that the
social sciences observe, reason, evaluate and intervene with reference either to
individuals (actors) or to social structures (systems). Because individuals are weak
and becoming increasingly so, and as social structures become autonomous or start
to decompose, it is understandable that we are unable to see a solution to the
problem of connecting agency (seen as the human side of ‘the social’) and social
structures (seen as the non-human side of ‘the social’). Turning the clock back to
times when the ‘social’ was immediately ‘human’ is not possible. On the other
hand, the continuing and growing separation between the ‘human’ and the social
implies huge risks. What is to be done? This panorama describes the postmodern
condition and the impotence of postmodernist thought that characterizes it.
The relational paradigm for the social sciences – as the version of critical
realism which is advanced in this book – is put forward precisely in order to clarify
the process of historical change that we are witnessing, with the aim of showing
how it may be confronted and challenged in order to build a society suitable for
To fully comprehend the meaning of the relational paradigm it is necessary to
delve deeper into the historical problem from which it originates, namely, the
progressive distancing over time between the ‘human’ and the social. For modern
sociology, this statement involves at least two fundamental issues: (a) first, it
signifies the expulsion of the ‘human’ person as the ‘subject’ or ‘bearer’ of the
‘factors’ which explain social and cultural dynamics (this expulsion was first
theorized by structural sociology and later by various forms of poststructuralism
and variations upon postmodernism); (b) second, and more broadly, it implies that
social actors find it increasingly difficult to confer a human quality upon the social
(difficulties first manifested in the original theory of agency)2 insofar as their new-
As is well known, the debate surrounding the influence of structure and agency on human thought
and behaviour is one of the central issues in sociology and other social sciences. I am referring here
to the theory of agency as developed in the last decades of the twentieth century before the issue
of reflexivity was introduced. In this context ‘agency’ has been widely understood as the capacity
of individual humans to act independently and to make their own free choices, in this way giving
their own meanings to social relations. ‘Structure’ has been used to refer to those factors such as
social class, religion, gender, ethnicity, customs etc. which seem to limit or influence the
The relational paradigm 23
found freedoms of action and spontaneity have increasing difficulties in giving
any real subjective significance to social relations. The first aspect has perhaps
been more explicit than the second, but obviously the two are closely interlinked.
The ‘social’ is no longer perceived, either by lay people or by professional
sociologists, as the place where human beings live. Given the dominant currents of
thought in sociology, namely, ‘individualism’ or holism, that which is ‘human’
has to be looked for elsewhere – in those parts of the social canvas where meetings
take place between subjective agents; to life experiences, and to the interior lives
of people’s minds (their feelings, motivations, rational or irrational psychology);
or in representations or collective fantasies, including virtual ones (works of
fiction, the nostalgia for a pre-global world, the aesthetics of orgiastic and tribal
groups, or the new dream of surpassing human limits through post-human technology). It is no new phenomenon that the scientific field theorizes the necessity of
maintaining a distinction between that which is human and that which is social,
assigning the social element to an ‘operative system’, whose dynamics are external
to the actors, even if shaped by their wills and intentions. More generally, it is not
new that what is social is identified with systems in essence devoid of ‘subjectivity’, that is, neutral to the will and intentionally of actors. Today, however,
this separation is assuming radically new dimensions. The ‘human’ is increasingly
reduced to a matter of personality characteristics and impulses, capable of
introducing disturbances or merely constituting ‘background noise’ to the system
of behaviour and rules that ‘make’ our society.
It is often said, the market has its rules, politics its games, the mass media its
logic, law its procedures, scientific research its paradigms and so forth. The human
subject ‘floats’ within the environment of the social system. The human person is
identified with the needs, desires and dreams – good or bad – of a ‘subject’
perceived and represented as external to and undetermined by the organized social
relations, identified with the social system.
Common sense generally reacts to the above process with uncertainty. Principles
of caution, responsibility for consequences and of damage limitation are invoked.
individual’s opportunities. In a more restrictive sense, the theory of agency was born as a peculiar
belief in the cause-and-effect world. This approach requires that the agent be regarded as a
‘substance’. The agent is always already a ‘self”, having all the necessary morals and beliefs within
it. All acts are caused by the agent, who is, as it were pre-formed (i.e. having no causal history).
What this means is that he/she is not affected by anything in space–time; this is a one-way street,
with the agent affecting space–time but not being affected by it. The requirement of the agent being
a substance makes him or her a predicate and not predicated of anything else. This theory
encountered many difficulties precisely because, in the new societal context (postmodern and
‘globalized’), empirical research has shown that agents feel more and more unable to act as
independent actors and must take into account new interdependencies with the surrounding social
context, i.e. become more ‘reflexive’ and adopt new forms of reflexivity. This is what the theory
of M.S. Archer (2000b, 2003) has clarified, thus revising the former (original) theory of agency.
To me, this ‘reflexive turn’ means that we need a new relational theory of agency and structure in
order to analyse the processes through which the meanings and practices of social relations are
generated in a globalized society.
24 The relational paradigm
These are not sufficient because in daily life individuals experience the division
between the human and the social as a lack of meaning, as an absence of ends, as
disorder, and as dramas of reality. For example, it is said that no human intervention
can avoid the economic system generating certain levels of unemployment, that
entire populations starve to death, that political games may lead to war, that mass
media destroys, damages or denudes interpersonal communication. All of this takes
place because the social has ‘its mechanisms’. Only in certain cases does the
distancing between the social and the human instigate positive, or even enriching,
relations in the sense that they promise new benefits (such as the argument that if we
succeed in automating work, developing assistance, and digitalizing communication, will we have more space for other activities).
Scientific reflection underlines the ambivalent character of this phenomenon.
On the one hand, it is noted that the distancing between the human and the social
is manifested in a popular feeling of loss of personal identity, going all the way to
de-humanization, of relations, of life styles and more generally of social ties,
particularly in the public sphere (though are matters not the same in the private
sphere?). On the other hand, it is remarked that such processes also represent an
opening up of new possibilities for creativity, freedom, motivational drive for
individuals and humanity in general. In the scientific sphere, contrary to popular
thinking, these positive aspects are over-valued: it is often stated that asking for
‘too much’ of the ‘human’ element in the social sphere is equivalent to seeking
regression or stagnation. For example, to those complaining that work has become
simply an exchange relationship, that politics are now only a struggle between
interests, that the mass media increasingly manipulate our lives, that scientific
research lacks an ethical core, it is replied that change would amount to reversal,
to renouncing progress and therefore a return to the middle ages.
In any case, the relation between the human and the social are no longer
immediate as they were in the past. To a greater extent than in other periods, our
society produces social forms which, although originating from people, are
perceived, lived and represented as in-human or de-humanizing.
The qualitative switch is manifested as (i) a co-fusion between what is human
and what it non-human, as (ii) the paradoxical inversion of the two and as (iii) a
nullification of the limits between these two. What society could emerge from a
historical change of this importance? This is what the relational paradigm asks. It
questions in what sense, if there remains one, can sociology still ask ‘what is there
that is human within the social?’ Is the confrontation between humanism and antihumanism in sociology dead? Is the outcome what is now and increasingly called
In this chapter I will attempt to answer these questions by illustrating that the
comparison between humanism and anti-humanism remains as valid as ever and is
currently pushing sociology towards new frontiers.
First, I propose to illustrate why we need to concentrate our scientific attention
on the category of social relations. The thesis is that in order to understand the
differentiation between the ‘social’ and the ‘human’, we need to understand their
relations (that is, the social relationship between them). Certainly, everybody talks
The relational paradigm 25
of relations. However, relations are often taken for granted and mainly appear as
an empty concept or a purely formal one. In any case, it is one of inferior and
secondary importance, which must be attached to a ‘substance’ (or derived from it)
if it is to gain acceptability. All of this prevents us from analysing the processes
under discussion. This situation derives from modern sociology’s failure to
thematize relations between the human and the social.
That form of contemporary sociology known as postmodernism is examined
to show that it does not go any way towards resolving the issue of what remains of
the human within the social, a problem which we have inherited from modernity.
Instead, the human becomes a paradox in postmodernism. In later sections a
conceptual scheme is advanced to define what elements of the human within the
social can still flourish if this issue is approached in terms of relationality. The
intent is to clarify, and where possible overcome, the reductionisms, the contradictions and the dilemmas inherited from modern sociology. In short, we must
start by re-envisaging relations between the human individual and the social.
Finally, I outline a way of thinking about the configuration (and re-configuration)
of society according to the relational perspective. This means conceptualizing the
social order as a ‘subsidiary society’, one that originates from what is civil rather
than from the lib/lab (liberal/labour) compromise between the market and the state,
between personal liberties and systemic controls.
Why a paradigm based on ‘social relations’?
Although the relational paradigm has met with increasing interest over recent
years, it nevertheless remains largely unknown and generally misconstrued, if not
totally rejected a priori. The underlying reason for this seems to be fairly clear.
More or less implicitly, the observer (the social scientist) takes for granted that the
concept of relations qua talis is not of first importance, but must come ‘after’ the
terms that it connects. This means that social relations are viewed as a product of
individuals and of social structures, as something that comes after them. On the
contrary, the relational paradigm affirms: ‘In the beginning there is the relation!’
(Donati 1991: 25). How is this possible? Why, if we wish to understand what there
is or is not of the human within the social, must we start from the relation between
these two terms (the individual and the social structure, the social and the human)
and not from one or the other or from a mixture of the two? (These themselves
represent opposing forms of thought.3)
Starting from structures leads to downward conflation (seeing actions of individuals as determined
by structures), whilst starting from individual actions leads to upward conflation (thinking that
social structures are the product of individuals); remaining on the horizontal and regarding
interaction between individuals and structures as being mutually constitutive leads to central
conflation (to think that individuals make structures and structures make individuals without it
being possible to distinguish their different contributions, their temporal priority or the direction of
causality). On these three forms of conflation, see Archer (1995).
26 The relational paradigm
What are the difficulties in assuming this point of view and in looking for a
development of the social sciences in this direction? To repose the question: what
is there of the human in the social? To illustrate: what is there of the human within
an economy which calculates everything in terms of cost–benefit analysis, which
values products (goods) at their monetary exchange rate, which represents social
development in terms of a percentage increase in the GDP? Similarly, what is there
of the hu