Resumé

How do you catch the fish you are after in a text-analysis, and what are the effects
of your world-view upon your abilities to fish? Although the analytic positivists can
indeed help us to answer these questions, their world-view has nevertheless
functioned as a constraint upon the development of political theory by ignoring
(more or less) the methods of the understanding traditions. Because, if you fail to
understand the complex of problems in a given text-system, you are also in danger
of analysing and criticizing nothing but your own misunderstandings — of kidding
no-one but yourself. In the case of David Easton's concept of political persistence,
such analytic misunderstandings have helped to produce a one-dimensional picture
of the systems model, which, however, can be dissolved by uniting world-view,
metascience, and science of respectively the analysts and the interpreters within the
framework of the critical traditions.

Introduction

'Some theories are successful. They become part of the intellectual instruments of
description, explanation and prediction. Other theories may be popular for a short time,
receiving attention and then just fading away ... .

David Easton's theory belongs to the category of theories that come into vogue and then just vanish . . . The systems approach seemed to promise what no other theories had ever achieved, a complete and integrated understanding of reality. However, the results failed to fulfil the promise.' (Lane 1978, 161)

The statement above describes rather well the development of the positivists' view upon David Easton's approach. From success to utter failure. But does Easton's concept of political persistence reflect a bad positivistic model? And if not, what are then the differences between Easton's approach and the various positivistic models of equilibrium? Might not Easton's text-system eventually hide some potential for a critical, democratic social science?

These questions imply that there may be other ways of defining success and failure than the positivistic one - that the questions of success and failure are always determined in the light of a specific world-view, which tells you what to perceive of the world and how to act in the world (Bang 1981).

In this case-study of David Easton and of his positivistic critics (exemplified by Jan-Erik Lane's criticism) I shall try to reveal the above correlation by demonstrating how the positivistic world-view makes Lane misunderstand the concept of persistence. Although my analysis of the effects of the positivistic world-view will thus be one of concrete negation, it does not represent a simple dismissal of the method and way of life. I just want to show that the 'right' of positivism does not include the whole of the existing field of traditions within the social sciences, which again implies that one has to bring out the limitations of positivism to specify in what situations, in what ways, and to what ends this tradition might be properly employed. And as I find Lane's Easton-analysis to be one of the most systematic positivistic criticisms of the systems approach, it also seems suitable for such an explication.

On the Position of Positivism

Lane declares that at least the following problems can be dealt with in a metastudy of a cognitive theory: (PO) the problems of the theory; (PI) the genesis of the theory; (P2) the logical and semantic structure oj: the theory, and (P3) the validity of the theory.

An adequate answer to (PO), Lane says, is a logical pre-requisite for dealing with (PI), (P2), and (P3), whereas (PI) is logically independent of (P2) and (P3), just as (P2) can be answered without (P3) being answered. A study based on (PI), he continues, is a study in the history of science, the psychology of science, and the sociology of science. A study based on (P2) is a metatheoreticai study, whereas a study with (P3) as the starting point concerns the truth of the theory (Lane 1973, 1).

Today, however, the metascience of the social sciences might be better described in terms of at least five different perspectives: 1) an axiothetic systems perspective viewing sciences as a goal-setting social system with aims which are recurrently asserted and which are capable of change and redirection, 2) a critical perspective, which poses questions regarding the meaning of science for man and society and seeks to transvaluate by rational forecasts and futurological analysis the impact of science on projected societal conditions; 3) a synchronicperspective (comprising (P2)

above) which seeks to understand the logical, semantic, informationtheoretical, and epistemological aspects of science; 4) a diachronic perspective which seeks to explain the genesis, growth, and development of scientific knowledge, and 5) a sociological perspective (in the widest sense) provided by the different disciplines, which focuses on producers, interessees, and users of knowledge in terms of a 'science' (Dunn & Fozouni 1975, 37).

All of these perspectives complement and mediate each other, and one may therefore ask why Lane perceives (P2) as the only 'valid' form of metatheory. But to answer this question we have got to know from what world-picture hypotheses and way of life Lane's metatheoretical position springs (Lundquist 1981, ch. 2).

The Problem of the Positivistic World-View

Although Lane does not express it explicitly, one may soon experience that there — according to Lane — exists only one legitimate world-view, namely the world-view of positivism - prescribing goals of knowledge which are final, certain, unified, and value-free — formulated in an extensional language and in axiomatic terms characteristic of formal logic and mathematical notation:

'The representation of a theory as a formal system - axiomatization, formalization and the exact stating of semantic reference of the terms of the theory - is a means to answer other questions the degree of the range (universality) and the scope (precision) of the theory, its theoretical falsifiability, its level of abstraction, alternative corrections of the theory on falsification, etc' (Lane 1973, 2)

'The difficulties presented by David Easton's theory stem . . . from . . . general considerations of its internal structure. There are simply too many conventional flaws . . . David Easton's theory is not bad because it is a systems theory. Easton's theory is bad because it is a bad theory'. (Lane 1978, 162)

As there exists only one legitimate world-view, Lane seems to say, there also exists only one valid form of metatheory. And if you employ this metatheory on Easton's model, you can clearly see that you are dealing with a bad, positivistic theory.

But what informational value does this statement possess, if Easton's approach does not represent a positivistic theory? How can you expect to grasp the problem of Easton's approach without an examination of his daily-life language — of the meanings he attaches to his model? Of course these questions may also appear as trivial seen from Lane's point of view, because from the quotations above you can follow how Lane tries to fix

the limits of language (and of thought and of the world) by disclosing its uniform logical structure: The universal structure of language is located in the logical form hidden beneath the surface of normal communication, and this form in turn mirrors the structure of the world.

You can almost sense the Kantian ring as well as Lane's conception of linguistics and of meaning: Logic covers everything that can be said a priori; the logical syntax of language fixes the boundaries within which meaningful factual propositions are possible; and what cannot be said in such a language cannot be meaningfully said at all.

In this way Lane's metamodel appears as a second-level construct (Schutz 1973), as his first-level construct is the world-view model above, through which he has already prestructured social reality prior to his metaanalysis. What cannot be said within the world-view of positivism must be rubbish. And if Easton does not accept the rules of positivism as binding for deciding what is true and what is false, then he is quite simply irrational. However, what Lane seems to forget is the fact that his; own agreement in the opinions of positivism also implies his agreement in a specific way of life, as in his learning of the language of positivism he also comes to engage in agreed common practices and to share agreed common criteria for their performance:

'So you are saying that human agreement decides what is true and what is false? - it is what human beings say that is true and false; and they agree in the language th«;y use. This is not agreement in opinion but in forms of life'. (Wittgenstein in McCarthy 1978, 165)

By treating Easton's actions and ideas as independent variables located within the world-view of positivism, Lane also demands Easton to accept the reduction of these actions and ideas to the standards and criteria of his own way of life. And as these standards and criteria themselves are incapable of justification within the logical syntax of Lane's language, how can he then justify his own conception of Easton as a 'bad' positivist, who responds in the 'wrong' way to the orders and the training of positivism?

'Following a rule is analogous to obeying an order. We are trained to do so; we react to an order in a particular way. But what if one person reacts in one way and another in another to the order and the training? which one is right?'. (Wittgenstein in McCarthy 1978, 165)

The internal connection of language with practice — with knowing how to
do certain things - thus implies that the application of a monological
metamodel as Lane's upon a foreign language very easily leads to misunderstandings,

as understanding of a language refers to skills that one
has acquired — to activities that one has learned to carry out in common
with others (Gadamer in Connerton 1976, 117—134).

To understand is to explicate to a point at which the problem in question appears worthy of consideration from a common point of view of humanity. The meaning of a given text-system is in principle incomplete, open for interpretations from future perspectives. The movement of history and the changing situation of the interpreter bring out new aspects and cast former elements in a new light. The meaning of a text goes beyond the author, and understanding is therefore not merely reproductive, but also productive (Riceur 1979 A+B).

The interest in understanding is consequently an interest in dialogue with others, with the past, with alien cultures and so forth about the common concerns of human life. And such an orientation is not that of the neutral observer but that of the partner in dialogue. If we fail to understand a given text-system, we can of course appeal to error, logical fallacies, self-delusion and so on to explain something that we are unable to understand. But since we have no monopoly on truth and goodness, we must maintain an openness to the ideas of others and be prepared to learn from them - which in turn implies that any text-system must be understood and explained in relation to the existing field of traditions of which positivism stands as a part to the whole.

On the Relation between Knowledge and Human Interest

We shall now follow the consequences arising from an unreflected application of a problematic world-view, and the obvious point of departure is Lane's conclusion — that the systems model represents a bad explanatory, structural-functionalistic theory - as this conclusion shows us that Lane begins his metaanalysis from the conclusion that Easton's approach is an explanatory, structural-functionalistic theory:

'A systems analysis of politics has to solve some fundamental theoretical problems:

a) the identification of the unit: what is a political system?

b) the definition of the homeostasis: what does persistence imply?

c) the statement of the variable of the system which maintains the homeostasis: what is
the relation between demand, support, decision and action on the one hånd and
persistence on the other?

Easton fails to solve these vital theoretical questions . . . Easton's theory is bad because
it fails to solve some problems that are at the very core of what a systems analysis
implies.' (Lane 1978, 178)

I wonder if Lane could name just one social theory which can stand up to these criteria, but apart from this I shall argue that his conclusion relies on his own misunderstanding. I can find no passages in Easton's texts where he defines homeostasis as the central problem of political science, and for instance looks upon politics only as a question of how the blood can maintain relative chemical stability in the face of loss of oxygen (negative entropy). On the contrary, Easton in all his texts explicitly tries to outdistance himself from this type of systems theory, as he feels that the problems above are impossible to answer within a framework based upon an analogy with biology:

' ... the framework elaborated here has not been able to lean on any ready-made
model; and no eclectic borrowing from other varying kinds of systems approaches would
do.' (Easton 1965A, xii)

'Although in the outcome, systems analysis - as adapted for purposes of social research - remains within the same general conceptual terrain in which it has grown up, we shall find by the end of our examination of it that it has gone off in substantially different directions. Biological and natural scientists would no longer feel at home in it, although it might well stir faint and nostalgic memories of a conceptual homeland that they once knew.' (Easton 1965A, 2, italics mine)

However, Lane would surely remark, effusions like these cannot be taken for their face value, as Easton constantly speaks about functions and thereby indicates that he is operating from the concept of homeostasis. To postulate otherwise is flatly erroneous, as a functional explanation is at bottom a species of causal explanation, as its distinctness is a matter of formulation and point of view and not one of logic. Therefore one cannot perceive Easton's model as a challenge to the world-view of positivism, but only as a particular strategy on the scientific level which Easton applies rather badly.

But from the fact that Easton speaks about functions does not follow that he also sees the word 'function' as connected only to the world-view of positivism. On the contrary, Easton sees the inadequacies of the existing functional approach as in principle insuperable so long as the word 'function' is subordinated to a positivisitic world-view seeing theoria as contemplation of the cosmos and aiming to describe the universe theoretically in its lawlike order, just as it is:

'Although political scientists are taught to criticize fifth-century Greek thought lor its
dangerous and indeed fatal search for the conditions of stability, it is a tragedy of
contemporary research that it too stands committed to the investigation of similar

conditions . . . (It) . . . has even graver consequences than the similar pre-occupation of the Greeks. The critical inclinations of the latter stand in marked contrast to the strong pre-disposition ... to view the going political system as though, with all its avowed imperfections, it were the best of all possible practical worlds . . .

Today political research seldom transcends the frame of reference of its own age. However painful it may be to admit, political research leaves the impression that the study of the sources and the direction .of basic change is not of great consequence or urgency.' (Easton 19538, 43, italics mine)

By making the world appear as a universe of facts whose lawlike connection can be grasped descriptively, positivism suppresses the transcendental framework that is the precondition of the meaning of the validity of such theoretical propositions. But as soon as this objectivist illusion is dispelled and theoretical statements are understood in relation to prior frames of reference in the life-world, their connection with interests that guide knowledge becomes apparent, and the objectivist idea is therefore nothing but an artificial abstraction necessary only as a means for simplifying changing reality:

'The aspect of the event selected for description as the facts about it, is determined by the
prior interest of the observer; the selection is made in the light of a frame of reference that
fixes the order and relevance of the facts .... A fact is a particular ordering of reality in
terms of a theoretical interest.' (Easton 19538, 53, italics mine)

In this way it becomes possible to state that the really crucial problems of social research are concerned with the patterns of change, as no social structure is stationary but in continuous — if at times imperceptible — change.

On the Problems of Positivistic Functionalism

Although the crucial problems of social research are concerned with the patterns of change, one does not need to shy away from a functional interpretation of society, as such an interpretation can help us to explain how we get from one moment to another in history, Easton says. Both Marxism and phenomenology postulate — at least implicitly — some kinds of function. And neither of these approaches focuses exclusively - as positivistic functionalism - upon the self-maintenance of systems but primarily upon the conflict, diversity, and change of systems. We simply need a new functional model in which stability is treated as a special example of change and not as a generically different one, as we never find a social situation in which the patterns of interaction are absolutely unchanging:

'Even to begin to develop theoretical inquiry, it is necessary to go far beyond the relating of varying structures to functions ... If stability is to have any sensible meaning, it must represent a condition in which the rate of change is slow enough to create no special problems due to change . . . Similarly so-called change draws attention to another special case in which the rate is high enough to create special consequences of which it is necessary to take note, both analytically and empirically.' (Easton 1965A, 106/107)

If only we remember that the problems and solutions of theory are always also the problems and solutions of the society to which it belongs, Easton tells us, what is then to prevent us from defining a society minimally as the broadest grouping of human beings who live together and collectively undertake to satisfy all the minimum prerequisites of group life? (Easton, 19538, 135).

According to Lane, however, this functional interpretation of an ever
changing society is quite unsatisfactory:

"The criterion refers to "the minimum prerequisites of group life"; however, it is not
stated what these are. Only after these have been established empirically, can a concept
of society be introduced.' (Lane 1978, 172)

In order to present an adequate functional explanation, Lane tells us, it must be possible to give a reliable empirical delimitation of the boundaries of the system in question; to identify and specify precisely the state in which the system tends to maintain itself; to determine empirically the functional requirements of the system, and to designate the alternative processes through which these requirements can be met. If we were concerned with a biological system, then Easton surely would agree, but he denies that these conditions can be met in a social system.

In a biological system, Easton says, the above conditions are rather easily met, as such a system can be physically demarcated from its environment, and the state in which it maintains itself can be characterized in terms of a series of processes, with empirically specifiable tolerances, necessary for life. But the same cannot be said for social systems:

'What political systems as a type of social system possess uniquely, when compared to both biological and mechanical systems, is the capacity to transform themselves, their goals, practices, and the very structure of their internal organization.' (Easton 1965A, 99, italics mine)

Unlike the reproduction of organic life, Easton says, the reproduction of social life is not fixed by values that can be measured descriptively. This means that even if social systems could be precisely demarcated in some complicated way, it seems impossible to determine the goal states of such

systems in an empirically reliable manner. As there exist varying pathways for attaining alternative ends, physical survival can only be a necessary and not a sufficient condition for the maintenance of a given society. Therefore, positivistic functionalism becomes imprecise when applied to society, where the clearly defined problem of death and a corresponding criterion of survival are lacking, because societies never reproduce 'naked' life, but always a culturally defined life:

' . . . the adoption of equilibrium analysis . . . obscures the presence of systems goals
that cannot be described as a state of equlibrium. It also virtually conceals the existence
of varying pathways for attaining these alternative ends.' (Easton 19658, 21, italics mine)

The presence of alternatives in social systems, Easton continues, implies that the standards of historical life and survival are dependent on the interpretations that obtain in social systems. In the course of history not only the elements but also the boundaries and goal states of a given social system undergo change, and it consequently proves impossible to achieve consensus on either the numbers or types of requisite functions necessary for the maintenance of social systems or to come up with a logical way of finalizing any list. It is impossible to demonstrate logically that a given function is either necessary or sufficient for system maintenance (or for change for that matter) (Easton 1972, 130).

If we should forget the presence of alternatives in a given social system, we furthermore come to blur the identity of this system, as a given modification may then be regarded either as a learning process and regeneration of the original system or as a process of dissolution and transformation into a new system:

'What may be "inadequate" socialization for maintaining existing political structures may be highly "appropriate" for bringing into being new structures based upon new ideals and new kinds of political accomodations among the members of the system.' (Easton 1969, 42).

Therefore, positivistic functionalism must end up by offering us a normative answer to an objective problem. And if functionalism is to serve as a framework for empirical analysis, Easton says, we must also transcend the positivistic world-view and face the paradox that social systems seem able to maintain their identities even though not only their elements but also their boundaries and goal states undergo substantial and significant alterations in the course of history.

On Society and Its Subsystems

As defined by Easton, a society refers to all interactions through which the members of a group seek to cope with the problems of living together. One of these classes of interactions is that of providing goods and seivices for members of the society. This set of interactions is typically called the economic system. Another problem involves the provision of norms and goals through which members organize their activities. This set of interactions we may describe as the cultural system. Yet another problem concerns the way in which the members of society provide for the making and implementing of authoritative decisions. This set of interactions we may describe as the political system. No society, Easton says, could usefully be identified without taking into account at least these three subsystems, as they all appear as necessary —but not sufficient) conditions for societal existence (Easton 19738).

According to Lane, this way of describing a social subsystem - such as
the political one - represents a circular definition:

'The phrase "political system" is defined by means of the word "society", which in turn is defined through the term "political system" (political allocation) . . . How can one society be distinguished from another, if a criterion of a political system is not applied? the phrase "political system" is necessary for the definition of "society", not vice versa.' (Lane 1978, 173)

But as Easton points out above, you cannot define a society without including politics as an important element, nor can you define politics exclusively in terms of the social. The political system, Easton says, stands to society as a part does to the whole. And since the whole is different from its parts in either a quantitative or a qualitative sense, we may expect a definition of society to include the political system and yet to be different from it. (Easton 19538, 135-138).

Political activity is indeed vital in a society. But so are economic
activity, culture, social structure, and the like.

A society has numerous aspects, each of which contributes its share in a totally interactive process. We may, as Marx, talk about society as a 'whole' to make it clear that the interdependence of an element is only ever the form of its dependence. We may also suppose that the interplay of differences in a social system is regulated by the unity of a determination in the last instance. But we cannot insist on some kind of primacy, out of hand, for one or another social aspect:

'We may grant that empirically, at given times and places, one or another aspect of
society - the political, economic, cultural, psychological or structural - may indeed attain

selective emphasis for special reasons. But it is difficult to imagine a kind of society in
which each of the aspects mentioned could not be said to be "formative" in some
significant sense.' (Easton 19738, 294)

By the help of this multicausal - but not uncritical - interpretation of society, one can comprehend a social system as a complex, organized, uneven whole. It is complex in the sense that it is composed of many parts, the relationship of which are difficult to sort out. It is organized in the sense that these parts are not haphazardly related but stand in some determinate order, so that from the relationships characteristic consequences result that are not otherwise attainable. And it is uneven in the sense that one or another part — at given times and places — regulates the interactive process of the parts by a determination in the last instance.

Thus conceived, it is impossible to reduce a social system to either of its subsystems, because a complex, organized, uneven whole displays properties that are both different from the properties of its constituent parts and from a mere aggregation of those parts. And it is also impossible to reduce the different subsystem of a system to one another, as the parts of each subsystem also stand in some determinate order so that from their relationships characteristic consequences result that are not otherwise attainable (for example, with regard to society, the provision of goods and services, the provision of norms and goals and the provison of authoritative allocations of values).

Our objective as social scientists is in this way to begin with the whole, search out its characteristic properties, break it up into its component parts or subsystems, decompose these parts in turn into smaller subsystems at a second order of analysis, and so on in a descending order until we come to the lowest 'particle' of behavior.

Theoretically, if we had the detailed knowledge necessary, we might some day be able to deduce from individual behavior all of the collective consequences. But in the existing, uneven life-practices we have to take into account that a social system acts as a set of limits — which can of course be changed - but which cannot be ignored either by the actor or by the social scientist (Easton 1972).

On the Analytic and Definitional Nature of Knowledge

By treating Easton as a partner in dialogue, by focusing on his daily-life language and not only on the uniform logical structure of language, I have revealed that although society is virtually a primitive term for Easton, it is neither arbitrary nor without an empirical base, as Lane seems to believe.

We have also followed how Easton's definition of a political system — as a set of interactions through which values are authoritatively allocated for a society — is explicitly connected to his conception of a social system. A political system, Easton says, is more than just an aggregate of members. It provides a context in which actors may organize themselves for influencing political outputs as well as the overall political structures within which political conflict and cooperation occur.

Nevertheless, Lane does not only see Easton's definition as a circular
one but also as an instrumental one:

'As Easton himself points out . . . DFI is neither true nor false, but a stipulative definition. Easton considers that DFI expresses what is similar in and common to all political systems and that it clarifies what is meant by "from a political point of view", "the political aspect". DFI is an example of explication.

Easton points out that the acceptance of DFI depends in the first instance on whether
the definiens words are unambiguous.' (Lane 1978, 169)

One can almost sense how Lane's definition of Easton as a positivistic functionalist makes him perceive Easton as an advocate for a radical functionalist perspective where truth is not a matter of correspondence or of rational consensus but of stabilizing certainty under the pressure for decision, of guaranteeing a de facto consensus (McCarthy 1978, 230). But once again I shall stress the importance of keeping one's eye on the author's own understanding of the phenomena being investigated. Because, for Easton himself, (DFI) and the phrases 'from a political point of view' and 'the political aspect' express something extremely fundamental - closely related to his conception of change and of 'the whole': the analytic and definitional nature of knowledge.

What Easton seeks to signalize by his use of the term 'the whole' is that for every group of biological persons there exists a mass of undifferentiated interactions among which the social scientist finds those that concern him. These interactions are the 'apperceptive mass' with which the social scientist begins his analysis:

'In referring to society, we are conceiving of all behavior undifferentiated as to type, what we might call the apperceptive mass of observations present to our senses. As a concept, society calls attention to the gross mass of conceptually unorganized social interactions that we might perceive if we were able to take in the whole of society, literally, in one glance.' (Easton 1965A, 38)

The knowledge of all social sciences, Easton says, is by nature analytic and
definitional. Social scientists cannot use 'totalistic' phenomena, or what

we can also call the apperceptive mass of direct unorganized perception, as their object of reference. It would correspond to the blooming buzzing world that confronts the child, before it has any experience or concepts with which to organize its perceptions of the world. Knowledge does not proceed from the concrete to the abstract but from the abstract to the concrete, and this whole process takes place in thought, while the apperceptive mass, which gives rise to the process, exists outside of thought. Observation requires that we select out of the total reality (the concrete apperceptive mass directly known to the sense, if this were genuinely a 'knowable' event), those sets of interactions that are of interest to us according to our varying criteria. No one, Easton states, looks at the whole world at once or even at any total event or action. The nature of perception permits us only to view the object-world selectively, to see aspects of it. Although the knower seems to select freely from the apperceptive mass, the knower is so to speak disciplined by nature, as the ultimate key to the world is the adequacy of explanation and understanding offered by our theories. The knower has to work upon the world in order to add any new knowledge to it.

In this way Easton indicates that the pre-condition of the whole process of knowledge of a real object is the existence of this real object outside of thought. (Althusser 1976, chapter 3). This real object is the apperceptive mass, which remains — before as after the knowing process - outside the intellect and independent of it. But on the other hand Easton also shows us that the process of knowledge takes place in thought, and not in the real object, as thought operates on the transitional set of interactions, which designate the real object in order to produce a concept of it, and in the next place a correlation of it, which may return to the real object as the truth about itself.

We must consequently discriminate between the process through which the object of knowledge — the different sets of interactions — is transformed into concepts, and the process through which the concepts are transformed into truths about the real object_— the apperceptive mass.

When it comes to the transformation of the object of knowledge into concepts, one can say that this process remains before as after inside the intellect and dependent on it, as knowledge — through the knowing process — is related to the circumstances under which it is produced. What you perceive of the real object as the object of knowledge depends upon how you define it, and how you define it depends upon the technical and practical developmental level of your own life-world. The perceiver selects or abstracts from the real object those sets of interactions relevant

to his interest. What is perceived is organized through what the perceiver brings to the knowing process. What the perceiver brings to the knowing process may in turn be conditioned by historical and social circumstances, and the end product of this process — our concepts — is consequently something other than the real object, although they are indeed related to this object through history. Because, as our definitions of the world change with the factual development so do our concepts, and it therefore becomes a question of the utmost importance not to confuse the real object with our concepts: (Easton 19738)

'Concepts are neither true nor false; they are only more or less useful.' (Easton 1965A,
33)

The reason why Easton draws a dividing-line between the real object and the concept of knowledge is precisely his afore-mentioned wish to unite the social sciences around the problems of an ever-changing society in order to prevent them from becoming a dogma in the bad sense of the word — a dead enterprise seeing the world as dead 'obvious' facts. By drawing this dividing line Easton thus invites us to think in quite a different manner:

'Knowledge can in this way be both related to the times and yet be part of a larger process through which each generation seeks to add to the accumulation of our understanding, with the ultimate test of validity of being explanation and prediction.⁷ (Easton 19738, 287, italics mine)

As the definition of life is not pre-given in a social system, we also need a mediation of historicism and the nomothetic point of view in order to demonstrate that the continued attempts to define cultural life are a necessary component of the very life process of socially related individuals. Because, although the process of knowledge is related to time, knowledge itself does not need to be of relativistic nature. From the thesis that you cannot establish eternally valid frameworks it does not follow that you cannot obtain valid generalizations of the real object, as the usefulness of our concepts lies precisely in their ability to change something in reality by adding something known to it.

Since knowledge of reality belongs in advance to reality, since it is knowledge of nothing but reality, our concepts only add something to reality on the puzzling condition of adding nothing to it. Through the process of knowledge each age adds to reality its own knowledge of that reality, but in each age reality is only putting in the pocket what in advance belongs to itself.

This does not mean that the distinction between the real object (the apperceptive mass) and the object of knowledge (the different sets of interactions) only presents us with a nullity, because in order to be annulled it must be constantly affirmed. The infinite cycle of all knowledge — adding something to reality only to give it back — can only be a living cycle, so long as it is able to reproduce and reorganize itself. The social sciences do not reflect disciplines wandering aimlessly from one conceptualization to another, as they respond to changing historical conditions, but disciplines whose history reveals an insistent search for increasingly reliable knowledge about social reality for the benefit of mankind.

Truth about social life, Easton says, does not lie in the discovery of some one and only system cohering out there in the phenomenal world; nor does it call for us arbitrarily to order that world of complex relationships in terms of our concepts. Meaning does not lie in the phenomena alone (the correspondence view) nor does it derive exclusively from the utility of the way our socially conditioned minds may order these phenomena (the instrumentalist view). Meaning arises from the dialectical process through which the knower perceives and understands relationships among phenomena and the existential limits these phenomena impose on the process itself. These limits, Easton says, lie in the need to test, through experience, the utility of a particular ordering of phenomena by its contribution to our explanatory and predictive powers (Easton 19738, 287-288).

Social science is neither a question of aimless relativism nor a question of dead, frozen, ossified knowledge. Knowledge of the concrete does not come at the beginning of analysis but in the end, and it only becomes possible on the basis of useful concepts, and not on basis of the immediate evidence of the concrete:

' . . . I have not been willing to deny the impact that values, social class, ideology and the like have on our research. Willingness to make allowance for distortion and error due to social factors is not to be mistaken, however, as a "concession" to historicism, except insofar as this epistemological view has helped us to appreciate what could be easily neglected. Nor is recognition that knowledge is part of a social process necessarily a denial of the attainability of objectively valid science.' (Easton 19738, 290)

' . . . the ultimate key, of course, is the adequacy of explanation and understanding
offered by the theory.' (Easton 19658, 473)

On the Concept of Political Persistence

In the remaining part of this article I shall try to show how Easton's critique of functionalistic positivism and his definitions of change, society, system, and knowledge express themselves in his systems approach. And at the same time I shall try to explicate how Lane's failure to grasp Easton's intentions manifests itself in his criticism of 'persistence', although his own metamodel — to a certain extent — is able to follow Easton's actions.

'How is it that in the face of continuing blows, from within or without, even the minimal
organization, cooperation, commitment of resources and energies, and obedience to
authority are possible?' (Easton 1965A, 78)

'How do any and all political systems manage to persist in a world of both stability and
change?' (Easton 19658, 17)

According to Lane, Easton here specifies two basic problems:
(PI) What are the necessary conditions for the persistence of a political
system?

(P2) What are the sufficient conditions for the persistence of a political
system?

These two problems, Lane continues, are in turn related to two basic
types of sentences:

If a political system persists then it satisfies the conditions n1?n1? n2,
nn.

(S2) If a political system satisfies the condition Nx, or N2, or NN, then it
persists (Lane 1978, 162).

The important thing to note in Lane's presentation of the problems is
that he forgets to see them in relation to the fundamental hypothesis from
which they spring:

(HI) If a society shall exist, then it must be able to provide the processes through which some kinds of political decisions can be made. The presence of a political system of some sort is a necessary condition for societal existence (but not a sufficient one, of course).

By settling on systems persistence Easton seeks to raise the question of how societies are ever able to provide for the making and implementing of authoritative decisions in order to discriminate explicitly between the conditions for persistence and the conditions required for the maintenance (low-rate change) or change (fundamental change) of one or another type of political system:

' . . . the need to distinguish sharply the maintenance of a particular kind of system . . . from the perpetuation of the basic functions of political life . . . compels us to adopt a different concept . . . systems analysis seeks a theory of persistence, not of selfmaintenance or equilibrium.' (Easton 1965A, 88)

I must agree with Lane that Easton sometimes confuses the question of social identity with the question of the identity of a specific type of political system - especially in (1965A). But I shall argue that these errors of Easton's become unimportant if we see them in relation to Easton's statements on the world-view level and metascientific level. They are only blemishes and not malformations.

In fact, Easton's discrimination above is far from surprising, if we see it in relation to his conception of knowledge and of a social system. The analytic and definitional nature of knowledge, Easton tells us, implies that social systems in opposition to natural systems never reproduce 'naked' life, but always a culturally defined life. The social evolution, so to speak, runs through the constructive learning of socialized individuals and not only through a goal of biological survival. Through the process of evolution, organisms have indeed developed a range of behaviors that make survival a preferred state, or the goal in quasi-teleological terms. But even though higher order organisms can substitute structures and processes in order to achieve this goal, they cannot as social systems set and change their goals. Social systems are not bound to any single preferred state such as survival. Physical survival is only a necessary and not a sufficient condition for the maintenance of a given social system (Easton 1953A; 19558, 1973 A).

Positivistic functionalism claims that the control values of a system are 'given' in the same way as the cultural values that determine social norms, but in reality the goal state of a social system cannot be ascertained in the same way as for the parametrically determined equilibrium state of an organism. The empirical values that, can be ascertained for a given system cannot be related to an optimal value, as control values are not 'given', but can at best be 'found' by way of a political formation of will.

The formation of such a will would however demand that the cultural values themselves would be drawn into the discussion, would be examined pragmatically and cleansed of their ideological components (Easton 1950). But such a cleansing would presuppose an analysis of the moral worth of the values in a given social system, and this is not the task of an explanatory science but the task of hermeneutical reflection (Easton 1951; 1955A). By claiming that the control values are given, positivistic functionalism also comes to legitimize authority by producing the objective

appearance of justifying norms that are precisely not capable of being
justified within their own framework:

'If the survival of society were guaranteed under any eventualities, then the social sciences could afford to tolerate research that was indifferent to the framework within which it was cast. But . . . the urgency of social problems demands a reconsideration of the link between the conceptual framework in each social science and the utility of the results of that science for the attainment, preservation, and extension of goals upon which men have agreed.' (Easton 1950, 476)

The concept of persistence, therefore, does neither represent an attributable goal of any system nor a goal to be fostered on any system. It is only a theoretical tool and explicitly not a value (Easton 19738, 282). The members of a given society might well prefer to destroy the capacity of their society to act politically rather than accept their specific political system type. What persistence only wants to stress is the argument that if the members of a given society succeed in destroying its overall capacity to provide for the existence of some kind of political system, then their society also ceases to exist.

Persistence versus Self-Maintenance

If we return to Lane's definition of persistence, we shall be able to reveal
the fundamental difference between 'persistence' and 'self-maintenance'.

Propositions of the type (SI) and (S2), Lane says, may explain v/hy a political system persists and does not persist respectively, but not why it is of a certain kind. But by introducing more conditions into (SI) and (S2), he continues, it will - for example with the aid of a proposition of the type (SI) — be possible to explain, why a political system failed to persist:

(1) For all political systems it holds that, if a political system persists,
then it has support.

(2) For all political systems it holds that, if a political system lacks
support, then it does not persist.

(3) The French political system under l'ancien regime lacked support.

Conclusion: The French political system under l'ancien regime did not
persist! (Lane 1978, 163).

However, according to Easton, sentence three above does not follow from sentence one and two, as sentence three does not refer to the question of persistence but the question of the maintenance (or change) of a specific political regime. In Easton's model the concepts maintenance

and change refer to three important aspects of a political system: the authorities, the regime and the political community. Separately or in combination they refer to the type of system - as for example l'ancien regime — whereas persistence is reserved exclusively for referring to a society and its capacity to provide the processes through which some kind of political decisions can be made, regardless of the type of authorities, regime, or political community involved.

A French persistence failure under l'ancien regime would have implied that the French society had ceased to exist and either had been divided into two societies or absorbed into another — as was the case with Vietnam and Korea on the one hand and with Lithuania, Estonia and Latvia on the other. In each of these instances the society in question experienced a persistence failure, as the members of the society — bearing these names - were unable to sustain processes through which binding decisions could be made for the members of each of these societies. Of course, the lack of support for a specific political regime - such as l'ancien regime - may lead to a persistence failure - if the regime in question can be neither maintained nor changed. Because, and this is the second fundamental hypothesis of Easton, if a political system of any kind fails to provide for some set of authorities, a regime, and a political community then it cannot convert competing demands into outputs (Easton 19768).

One can say that Lane's example above is a clearcut example of Easton's criticism of positivistic functionalism. As not only the elements but also the boundaries and goal states of political systems undergo change in the course of history, positivistic functionalism very easily comes to blur their identity, as it cannot discriminate between the problems of a specific political system type and the problems of political systems of any kind. In this way positivistic functionalism makes it impossible to determine whether a given modification - in Lane's case of l'ancien regime - represents a learning process and regeneration of the original system or a process of dissolution and transformation into a new system. But by subordinating the problems of the self-maintenance and change of political systems types to the societal problems of persistence such a determination becomes at least possible - also in the French case:

'In the case of France . . . there has been a succession of such changes ... As a group conducting its political affairs in common the metropolitan French have maintained their community relatively intact . . . But the French political order (regime) has displayed greater instability than that of most other modern democracies.' (Easton 19658, 190)

' . . . some kind of political system has managed to prevail in France in the face of fundamental transformations in the regime over the ages, oscillating as it has among monarchial, democratic, and authoritarian forms. Indeed, persistence of any kind of political system may well have been contingent upon its readiness to respond to environmental and internal changes by regulating itself through such basic political alterations. But for this readiness to accept periodic and radical transformations in the regime, the society we call France might have found itself unable to continue to make and implement authoritative decisions for its members. Not that any particular historical change was necessarily inescapable; many alternative solutions might have existed. But the persistence of a political system of some sort may have hinged upon the ability to adopt important modifications of the status quo\ (Easton 1973C, 34, italics mim;)

What Easton tries to explicate here is precisely that the persistence of processes for authoritatively allocating values in society may be contingent either on the maintenance or on the change of the type of political system — in this case the type of authorities, regime or political community in France.

Persistence and Social Order

From the conclusion 'the French political system under l'ancien regime did not persist', Lane says, one can deduce that Easton confuses (PI = what are the necessary conditions for the persistence of a political system) with another problem (P3): what are the necessary conditions for social order, e.g. that the society persists biologically and socially and that it is not characterized by anarchy or anomie. If these conditions are satisfied, Lane continues, there is a frame for variations of different kinds of political systems. If a political system exists, Lane says, additional conditions must be satisfied. If a political system ceases to exist, there are often other alternative systems which can replace it:

' . . . these problems should not be confused. Political systems may perish, even though the necessary conditions for political life are satisfied, because they may be replaced by other systems. When political systems perish, there is always the option between no social order and the replacement of the old system with a new one.' (Lane 1978, 164)

However, this is exactly what Easton himself tries to emphasize. Political systems types may perish, he says, even though the necessary conditions for political life are satisfied, because they may be replaced by other systems types. The members of a given society always have the option between no political decision processes at all (and consequently no society), and the replacement of the old system type with a new one. The problems generated in a society as it continues to provide for some means for allocating values authoritatively must not be confused and comingled

with the conditions required for the maintenance (or change) of one or
another type of political system.

Lane claims that Easton in (PI) confuses (1) those necessary conditions which, if they are not present, exclude social order and (2) those necessary conditions which, if they are not present, exclude a political system. But if the phrase 'a political system' refers to a political system of any kind then these two problems are two aspects of the same problem. Because if a political system of any kind is breaking down, Easton says, then the social order is also breaking down (social order = a minimal degree of order in the conflicts arising from the different subsystems. Anarchy, understood as absence of government, is impossible according to Easton, but anomie must be interpreted rather liberally).

Lane perceives a non-persisting political system in a society as a logical possibility, although the whole point of persistence is Easton's claim that this logical possibility is an impossibility - what we even can read from the Easton-quotation chosen by Lane himself to illustrate Easton's 'mistake':

'The two different problems are confused in one and the same sentence in Easton:

"Here we could not help but accept the interpretation that the political system had come under stress so severe that any and every possibility for the persistence of a system for that society had disappeared." In the first part of the sentence it is stated that a political system in a society is breaking down; in the second part of the sentence it is maintained that in 'the same society the social order is breaking down. If the latter state of affairs is the case, the former state of affairs is also the case, but the reverse does not always hold good.' (Lane 1978, 164)

This quotation seems to be a good indication of the way in which different definitions of social science lead to different organizations of the perceived world. Because, what Easton may say in the above quotation is pure and simple that a specific political system type has come under stress so severe that it can neither be maintained nor changed into another type which can fulfill the societal function of persistence - i.e. that any and every possibility for the persistence of a political system of some sort for that society has disappeared. And without a political system of some sort a society cannot exist.

But although Easton perceives Lane's logical possibility as an impossibility, he would nevertheless hold that you cannot exclude the logical possibility of finding social order in a society without a political system. As we have no logically valid way of determining the problem of death and the corresponding criterion of survival for a society, we have no logical way of demonstrating that a given function is either necessary or sufficient for the maintenance of a given society, Easton tells us. You must confuse

neither the object of knowledge (our selected set of interactions) nor our
concepts with the real object, as knowledge of the concrete does not come
at the beginning of analysis but in the end.

Even if we should succeed in specifying a set of variables around the concept of persistence to which most social scientists would agree, these variables would only comprise a natural system in the sense that they appear to cohere significantly. Without them, one could argue, it does not appear likely, on a priori grounds, that an adequate explanation of the major aspects of political phenomena could be obtained — that is:

' ... the interconnectedness of the variables seems clear and obvious, at least until dispelled by subsequent inquiry; in this way alone may they be considered "given" in or by nature. But this is just another way of saying that they form what I have been calling an interesting, as against a trivial, system.' (Easton 1965A, 33)

The phrase 'an interesting system' refers to a set of interactions about which we have formulated a set of variables which appear to cohere significantly. It does not refer to the real object. And if we should think so, we also are in danger of falling behind history, of repeating 'truths' which are only the names of things. 'Persistence' is only a concept, and concepts can never be valid but only more or less useful for finding valid correlations in the real world. As knowledge improves, each age is called upon to reassess its concepts for the purposes of scientific understanding, explanation and prediction. And this will surely happen for 'persistence' too. But from the present vantage point of history, Easton says, it seems as if any society in order to maintain itself must be able to assure a minimal degree of order in the conflicts resulting from the pursuit of its scarce values (Easton 1958, 171). And it also is a fact that every society has developed characteristic political institutions to mobilize the resources of the society for the purpose of regulating these differences. Consequently, it also seems meaningful to mediate the variety of political theory types by asking: How can a society sustain some kind of processes and structures for making and implementing binding decisions?

Lane perceives this question as a fallacy of misplaced abstraction, as he
finds that it confuses the meaning of essential variables respectively as a
criterion of identity and as a type concept:

'Easton's mistake leads to a situation in which the investigation into persistence of
political systems - democratic, authoritarian, traditional ones - somehow becomes an
inquiry into how societies manage to maintain social order.' (Lane 1978, 169)

However, seen from Easton's point of view, Lane in the above quotation becomes a victim of the same fallacy as he ascribes to Easton: confusing a criterion of identity (the persistence of political systems) with a type concept (the maintenance - or change - of the democratic, authoritarian, and traditional political regime type. And yet his own metamodel's revelation is in a way correct, as Easton's investigation into persistence indeed can be seen as an inquiry into how societies manage to maintain themselves - for example by exchanging a specific political regime type with a new one.

A necessary (but not a sufficient) condition for empirical entities to be
societies, Easton says, is that these units do certain things, i.e. perform
two basic political functions:

a) the allocation of values for a society, and

b) the acceptance of the allocations as binding by the members of the
society (Lane 1978, 165).

And wherever and whenever you find a and b you also find some set of authorities, a regime, and a political community, as the presence of such a set is a necessary (but not a sufficient) condition for the conversion of competing demands into outputs in a political system for a society. Finally, wherever and whenever you find such a set you also find a specific type of authorities, regime and political community, the maintenance or change of which is quite another problem than the problem of persistence.

A given political system type may totally disappear in order to secure
the persistence of a and b for a society through some set of authorities, a
regime, and a political community in the political system in that society.

On the Types of Facts Connected to Persistence

If we say, Lane tells us, that the criterion of identity of a political system is the function G (= a and b) and nothing else, then if G occurs in a society, a political system occurs. If a society Si exists a long time and has G, one and the same political system, the political system in s1?s_1? exists during the same period of time. If a society — e.g. British society - exists over a period of time, Lane continues, it is possible to determine if G occurs in that society at different times. If during a certain period, e.g. from the year 1700 to the present, G occurs in British society, it has had one and the same political system, designated by Easton as 'the British political system', Lane says, and he continues:

'In FP A Easton gets involved in contradictions owing to the functional interpretation. The starting-point is that there is a political system, "the British political system', which "... has maintained its basic identity as a system continuing through time", i.e. G has existed and exists. The basis of this statement is that "at this general level a political system of some sort has persisted through time with respect to the British society . . .", which is an entirely different matter. Easton confuses the functional interpretation with the structural one. The fact that there has been some kind of political system, i.e. a succession of different political systems over a period of time, is not the same as that there has been one and the same political system all the time.' (Lane 1978, 166)

However, we can also interpret the quotations above, as if it is Lane, who
confuses the functional interpretation, not only with the structural one,
but also with the phenomenological one.

We can say that Easton in the quotation above is trying to explicate that the fact that there has been some kind of political system in British society, is precisely the same as the fact that there has been one political system type all the time.

As a specific political system type the British political system has maintained its basic identity as a type continuing through time, although British society itself has changed its basic character and extent, and even though both the authorities and the regime have changed several times. This is because, at any moment during this period, we can find that the members — bearing the names of this specific political system type — have shown some minimal readiness to continue working together to solve their political problems — i.e. to maintain their political community (as the previous sentence represents Easton's definition of a political community (Easton 19658, 171-189).

When Lane speaks about a succession of different political systems over a period time, he speaks in fact only of a succession of different political regimes (a semi-popular dictatorship under Cromwell, a monarchial political order during the eighteenth and nineteenth centuries, and a popular democracy during the late nineteenth and the twentieth centuries) and thereby also of different authorities (as the old authorities normally disappear with the old regime). But only where all three objects (authorities, regime, political community) change simultaneously can we consider the former system type as totally disappeared. This has not been the case in the British example, because although the original British political community has expanded to include Scotland and Northern Ireland, and at one time embraced a now shrunken empire, there has always been a continuing core of political system members, calling themselves British, who have evinced, and still evince, an attachment to the British political community (Easton 1965A, 85—86).

In the course of history, not only the elements, but also the boundaries and goal states of the British political system type have undergone important changes, and yet we can perceive this change as a learning process and regeneration of the original system, if we - to a certain extent - rely on the interpretations of the members of this political system type who through the centuries have maintained a feeling and belief in their political community.

I think that Lane interprets system type problems as regime type problems in order to treat such problems from the point of view of a system's capacity to maintain its boundaries and its continued existence by mastering the complexity of an inconstant environment. In the case of organic systems, this point of view may be correct. But as the British example shows us, this point of view is insufficient with respect to social systems, where action involves a choice among multiplicity of meaningfully preselected possibilities. It is therefore the symbolically structured 'world' - in which world complexity is already registered through meaningful selection - which, in the case of social systems, is the immediate source of the pressure for selection (Habermas/Luhmann 1971). We cannot in the social sciences pursue the task of theory construction in abstraction from the prior categorical formation of the object domain, and we are consequently — to a certain extent — forced to rely upon the interpretations that obtain in a given political system:

'To explain the determinants of any particular act, we need to know not only the various circumstances or objective conditions surrounding it, but also the kind of personality, the improperly called "subjective" conditions, that individuals bring to the activity. No two individuals will define a situation and react to it in exactly the same way. The same facts may be so interpreted by two different persons as to lead to totally divergent decisions and actions.' (Easton 19538, 201)

From a theoretical (and not a moral) point of view, Easton says, the so-called 'subjective' definitions of the actors are of course neither less analytic and objective nor more empirical and subjective than the overt behavior of the actors or the objective social facts (institutions and structures), as our selection of either of these three basic types of facts (in the same way as our selection of a given set of interactions) refer to the process through which the knower abstracts from the tangible world of reference those aspects or types of social interactions that are of interest to the knower for one or another reason. The thesis of the phenomenologists — that sociological concepts represent 'second-level' constructs as the 'first-level' constructs are those through which social actors have already

pre-structured social reality prior to its investigation (Schutz 1973) - only holds good, if they include themselves among the social actors, thereby admitting that also this thesis is made in the light of a frame of reference that fixes the order and relevance of the facts (Gewirth in O'Neill 1973, 111-125).

Thus, the concept of persistence addresses itself towards three basic types of facts: 1) the social definitions of the actors, 2) the social behavior, and 3) the social facts (institutions and structures). And — as the British example pointed out - it does so in order both to avoid the reduction of the social definitions to behavior and/or facts and the subordination of the analysis of specifically social structures to the analysis of perceptions of, interpretations of, and orientations to these social structures.

When Lane ends up by concluding that Easton confuses a structural theory and a functional theory (Lane 1978,167), it seems to be due to the fact that he cannot perceive the subjective definitions and thereby the phenomenological interpretation of a political system in Easton's model. Easton tries to bridge the phenomenological and structural interpretation of a system by the help of the functional interpretation of a system. And he does so in order to discriminate between the process of integration in a specific political system type, and the political aspect of the societal process of integration.

On the Types of Explanation involved in Persistence

When we speak of integration of a specific political system type, Easton says, it is not enough to consider such a type from the point of view of its capacity to maintain its boundaries and its continued existence by mastering the complexity of an inconstant environment as positivistic functionalism proposes. After all it is subjects, who are involved in the process of integration. It is only when the members of a specific type experience the structural changes of their authorities and regime as critical for the maintenance of their political community that we can speak of a true crisis in a specific political system type. Disturbances of authority and regime integration endanger the continued existence of a specific type only to the extent that community integration is at stake — that is, when the consensual foundations of normative structures are so much impaired that the authorities, the regime, and the political community change simultaneously.

We must, however, be very careful to discriminate between typeintegration
and the political aspect of the societal process of integration.

Because, on the one hand disturbances of each of the three objects in a specific type, which do not represent a type crisis, may lead to a political and social identity crisis. And on the other hand a true type crisis does not necessarily lead to a political and social identity crisis, as a specific type may totally disappear in order to secure the political identity for a society (a and b) through some set of authorities, a regime, and a political community.

Thus, the functional criterion of political identity for a society indicated by persistence both involves a structural interpretation of the political aspect of social integration concerning a., and a phenomenological interpretation concerning b.

If we for example return to the British example, we can on the one hand say that the British society has been able to maintain its basic political identity by securing the persistence of some set of authorities, a regime, and a political community for the conversion of competing demands into outputs (the structural interpretation concerning a.). And we can on the other hand say that the members of the British society have been able to maintain their basic political identity by securing the persistence of a minimal level of support for some set of authorities, a regime and a political community (the phenomenological interpretation concerning b.). The British society has in other words showed both material and definitional continuity over the centuries by securing the persistence of a set of interactions capable of meeting the fundamental political functions without which no society could exist.

Persistence consequently opens up for a two-dimensional analysis of the political aspect of social crises (respectively social integration). Principally it refers to the interpretations that members of a society use in identifying one another as belonging to the same political group and thereby to the self-identity they assert through this group-identity. From this point of view a rupture in tradition - which causes the phenomenological political system that guarantees identity to lose its social integrative power — serves as an indicator of the collapse of society: a society loses its identity as soon as later generations no longer recognize themselves as members of a political system of some sort in their society. But as a society does not plunge into a political crisis when, and only when, its members so define their political situation, we must impose a survivalistic goal upon society in order to grasp that aspect of political crises in a society which owes its objectivity to the fact that political crises also issue from unresolved political steering problems. From this point of view we thematize a society's political steering mechanisms and the extension of the scope of contingency:

a society loses its identity as soon as it fails to provide fox the
existence of some kind of political system for the conversion of competing
demands into outputs.

This implies that although the members are not generally conscious of the political steering problems of their society, these very same steering problems create secondary problems through the feedback processes of a political system that do affect the members' consciousness in a specific way - precisely in such a way as to create political support problems for that society. Lane states that there can be no demand-stress, and that the relationship between demand and stress in Easton's model is therefore based on incorrect premises. But by viewing the question of crisis in the above way it also becomes possible to state that demand stress is indeed a matter of output failure (steering crises) although stress occurring from respectively the volume and the variety of demands ultimately may be reducible to support stress (Lane 1978, 177-178).

In order to grasp the interconnection between the two basic problems of persistence it is, however, necessary to introduce one more discrimination to avoid that we look upon the two problems only from the point of view of the control center of a political system. Theories of party politics, interest groups, legislative behavior, political leadership, administrative organization, coalitions and the like all deal with those structures or practices through which the outputs are influenced, formulated, and implemented and which thereby determine the way in which the valued things of the society are allocated:

'There is a status quo bias built into allocative research when it is untempered with an appreciation of the systemic conditions under which the allocations are taking place. It leaves the impression that the pie must always remain the same and even that the allocative processes change for reasons that are explicable solely in terms of the system itself.' (Easton 19658, 475)

We cannot allow ourselves to totalize politics in the above way. Because, as political output crises can be seen as occurring when the authorities cannot — under given boundary conditions — adequately steer the economic system, such a crisis can also be seen as a displaced economic crises. And as administrative manipulation of cultural matter may have the unintended side effect of causing meanings and norms - previously fixed by tradition and belonging to the boundary conditions of the political system - to be publicly thematized, a political input crisis may also be seen as a displaced cultural crisis (McCarthy 1978, 369).

By settling on systems persistence, however, we are able to see the

question concerning the control center of the political system in relation to other social problems. And in this way we may be able to solve the vital practical question of how to establish a society with a rational interaction — freed from any repressive forms of control — between the daily leaders and the people. That is, a society where the outputs of the political system reflect the collective efforts to secure the wants and needs of the people and where the inputs of the political system serve as indicators upon whether or not the outputs have had this specific outcome - and, if not, how to proceed to succeed. As Easton said as early as in 1947:

' . . . without the people, the rulers are as free spirits wandering lonely, dejected and unemployed in an empty world. But without rulers dominating their existence, the people, on the contrary, find that very freedom that calls forth their most creative efforts. Elitism places blind faith in an appropriate governing class. The democratic ideal incorporates a tempered trust in the wisdom and creative genius of the people.' (Easton 1947, 418)

In order to live up to this ideal Easton has concentrated his considerable energies on developing the positive side of critique by meeting the social sciences in general and political science in particular on their own field with the hope of shaping them into a critical consciousness. And the reason why he chose this way - which has obviously not been quite as successful as he had hoped - you can find in his 'programme of politics' from 1949 (his first article):

'Two decades ago, Ruggiero . . . drew attention to the liberal habit of speaking in "academic abstractions". Since the end of the war, we are threatened with another liberal inundation of lofty principles with scant attention devoted to the crucial problems of how they are to be realized. When a doctrine thus loses touch with immediate needs, when it no longer serves as a rational guide for concrete action toward its own fulfilment, it is well on its way to the ideological scrapheap of history.' (Easton 1949, 37)

Conclusion

I have, in this case-study, examined the rather obvious correlation — that
what you perceive of the world, and how you act in the world, depends
upon how you define it.

Having defined Easton's approach as a homeostatic theory to bring it into accord with his own world-view, Lane simply must experience his metamodel's findings as an indication of Easton's irrationality at the scientific level. And even though Lane's metamodel can 'see' that something is 'wrong', Lane himself interprets its 'seeing' of the phenomenological aspect of persistence as a sign of the logical fallacies of persistence.

Although I have only treated some aspects of Lane's Easton-criticism, I shall nevertheless argue that Lane's failure to grasp Easton's twodimensional form of analysis manifests itself in the remaining part of his criticism. In his analysis of the key-word 'authority', for example (Lane 1978,169—172), Lane imposes a Weberian power-model on Easton, even though Easton's own power-model (Easton 19558; 1958) expresses that you have to separate authority from legitimacy and legitimacy from legality (as the concept of support would otherwise lose its whole importance). However, an analysis of these problems must await a further study. But in spite of the fact that we have only been able to follow some aspects of Lane's criticism and of Easton's world-view, metascience and science, I think this case-study can tell us something about the need for a mediation of traditions in order to develop a political science which can meei: the theoretical and practical needs of the existing societies. The unreflective turning down of models springing from world-picture hypotheses, ways of life, and metascientific perspectives foreign to those of oneself is much too common, and it blocks the construction of a unified framework in which the positive elements of the different traditions are somehow preserved in a critical, democratic context.

As I have tried to show, Easton's framework reflects an attempt to unify the different political approaches. I am not going to argue that Easton's attempt (or that of myself) is free of either theoretical or practical contradictions. Surely, my interpretation of persistence calls forth a row of unresolved epistemological, ontological, metascientific, and scientific questions. But this is as it should be, for the exciting fact about political science is that every time we have solved a problem, we also raise some new ones. I have only tried to demonstrate the importance of knowing what to discuss, if our criticism shall be relevant. When it comes to human beings, it is not enough to impose one or another motive upon them. We

must also know their intentions, and the structure within which they
express themselves.

With regard to text-analysis I think this implies:

1. That we regard any text system as a complex, organized, uneven whole, which hangs together internally by virtue of its own complex of problems, and in such a way that you cannot remove one element without changing the whole content of the system (Lundquist 1981, chapter 2).

2. That you cannot determine the content of a specific text-system on behalf of its relation to a foreign text-system, but only on behalf of its relation to the existing theoretical and practical field of traditions, and to the social problems, motives, intentions, and structures which constitute and reflect themselves in this field of traditions.

3. That the development of a specific text-system in this way comes to
depend upon a) the text-system itself, b) the field of traditions, and
c) the social problems, motives, intentions, and structures.

4. That the prime mover in a specific text-system therefore must be the author as a concrete, historical actor, as history reflects itself in the individual development of the actor through the complex bonds, which - the process of socialization as mediator - connect the actor to history (Althusser 1975; 1976).

If we interpret text-analysis in the above way, I furthermore find it
possible to analyse the fundamental problems of political theories in a
wider social context by asking:

a) How is social order possible?

b) How is change of the social order possible?

c) How is individual emancipation in a social order possible?

d) What are the possibilities of action for the individual person in a
complex, organized, uneven social system? (Hondrich 1975, 12).

The deeper, the more differentiated a world-view, the more it draws upon real experiences, the more diversified and varied can its compositional expression be, Lukåcs tells us in his analysis of art (1978, 214). And perhaps this thesis is not that far out when it comes to the question of establishing a critical, democratic social science.

REFERENCES

Althusser, L. 1975. Filosofi, Ideologi og Videnskab (Philosophy, Ideology and Science).
Copenhagen: Rhodos.

Althusser, L. 1976. Essays in Self-Criticism, London: NLB.

Bang, H. 1981. 'The Conflict between the Traditions', Statsvetenskaplig Tidskrift 1, 1-21.

Connerton, P. (ed.) 1976. Critical Sociology. Middlesex: Penguin.

Dunn, W. N. and Fozouni, N. 1975. Toward a Critical Administrative Theory, Sage Professional
Papers in Administrative and Policy Studies vol. 3, 26, p. 5-75.

Easton, D. 1947. The Theory of the Elite. Harvard: Doctoral Thesis.

Easton, D. 1949 'Walter Bagehot and Liberal Realism', American Political Science Review
43, 17-37.

Easton, D. 1950. 'Harold Lasswell: Policy Scientist for a Democratic Society', Journal of
Politics 12, 450-477.

Easton, D. 1951. 'The Decline of Modern Political Theory', Journal of Politics 13: p. 36-58.

Easton, D. 1953A. 'Limits of the Equilibrium Model in Social Reserach', Chicago Behavioral
Sciences Publications 1 96-104.

Easton, D. 19538. The Political System, New York: Knopf.

Easton, D. 1955A. 'Shifting Images of Social Science and Values', Antioch Review 15 3-18.

Easton, D 19558 'A Theoretical Approach to Authority', Technical Report, Office of Naval
Research, 17, Stanford University, Department of Economics.

Easton, D. 1958. 'The Perception of Authority and Political Change' in Friedrich, C. J.
(editor) Authority, Harvard University Press.

Easton, D. 1965A. A Framework for Political Analysis. Englewood Cliffs: Prentice Hall.

Easton, D. 19658. A Systems Analysis of Political Life. New York: Wiley.

Easton, D. and Dennis, J. 1969. Children in the Political System. New York: McGraw-Hill.

Easton, D. 1972. 'Some Limits of Exchange Theory in Polities'. Sociological Inquiry 42,
129-148.

Easton, D. 1973A. 'Reflexions on Criticism', Social Science Information 12, 29-47.

Easton, D. 19738. 'Systems Analysis and its Classical Critics', The Political Science
Reviewer 3 269-301.

Easton, D. and Dennis, J. 1973C, 'A Political Theory of Political Socialization', in Dennis, J.
Socialization to Politics. New York: Wiley.

Easton, D. 1976. 'Theoretical Approaches to Political Support', Canadian Journal of Political
Science 9, 129-146.

Habermas, J. and Luhmann, N. 1971. Theorie der Gesellshcaft oder Sozialtechnologie.
Frankfurt: Suhrkamp Verlag.

Habermas, J. 1976. Legitimation Crisis. London: H. E. B. Paperback.

Hondrich, K. O. 1975. Menschliche Bedurfnisse und Soziale Steurerung. Hamburg:
Rowohlt.

Lane, J.-E. 1973. The Structure of David Easton's Theory. University of Åbo.

Lane, J.-E. 1978. 'On Easton's Theory: There must be Limits to Confusion', Politiika 3,
161-179.

Lukåcks, G. 1978 Essays om Realisme (Essays on Realism). Copenhagen: Medusa.

Lundquist, L. 1981. The Party and the Masses. Forthcoming.

McCarthy, T. 1978. The Critical Theory ofJurgen Habermas. London: Hutchinson.

O'Neill, J. (ed.). 1973 Modes of Individualism and Collectivism. London: H. E. B. Paperback

Ricoeur, P. 1979. Sprogfilosofi (Philosophy of Language). Copenhagen: Vinten.

Ricoeur, P. 1979. Fortolkningsteori (Theory of Interpretation). Copenhagen: Vinten.

Schutz, A. 1973. Hverdagslivets Sociologi (The Sociology of Daily Life). Copenhagen:
Reitzel.