# Means and Ends

**Year:** 2001 (originally drafted in 1940)
**Author:** Lon L. Fuller
**Keywords:** #human_ends #eunomics #fuller #social_arrangements #institutions
**Permanent Notes:**
**Literary Notes:**

___

**Summary:**

___
## I. Introduction
(Fuller, 2001, p. 61-62)Eunomics, as Fuller defines it, is concerned not simply with any social order, but a good and workable social arrangement. More specifically, Fuller is interested in "an order that is just, fair, workable, effective, and respectful of human dignity".

> Ambitious as the title of this book is, it would be still more ambitious if it were truly to represent what is here attempted. For, of course, we are not interested merely in order—the order, say, of a concentration camp—but in an order that is just, fair, workable, effective, and respectful of human dignity. Indeed, I once suggested a term for describing the kind of study undertaken in this book. That term was eunomics, which I defined as “the science, theory, or study of good order and workable social arrangements.”

(Fuller, 2001, p. 62)Eunomics does not try to answer what is the highest human good or what is the ultimate aim of human life

> The first thing to be observed is that while a quest for the principles that underlie good social order animates everything said in this book, it nowhere attempts to answer questions like the following: What is the highest human good? What is the ultimate aim of human life?

(Fuller e Winston, 2001, p. 62)Fuller rejects the idea that in order to be able to evaluate some given particular order as good we would need to have a fully fledged account of ultimate values.
This point if further elaborated on the first chapter of The Morality of Law.

I think a possible objection against Fuller is that he can rely, at first, in moral intuitions about what is reasonably just or good, but that a solid account of good social orders has to ultimately rely on developed account of normative ethics.

> Nor will it be likely to please those who consider that until the question of ultimate values is resolved no meaningful judgment can be passed on particular ways of ordering men’s relations with one another. To them the book will probably seem to suffer from a fundamental flaw in logic: it attempts to chart its course by a compass that lacks a pole toward which it can point.

(Fuller, 2001, p. 62)Eunomics rejects both moral relativism and moral scepticism

> If the analysis here presented is such as to alienate those who demand absolutes, it is perhaps equally true that the book will have little persuasiveness for those who stand at the opposite pole of social thought. Its analysis is not likely to appeal to the cultural relativist who considers that all value judgments are impressed on the individual by his social environment. Nor will the appeal be any greater for the ethical skeptic who asserts that any judgment expressing a preference for one state of affairs over another must be emotionally grounded and hence not properly a judgment at all.

(Fuller, 2001, p. 63)Fuller highlights that eunomics begins by analyzing the set of means which seem to be present or available in human social orders and tries to determine or attribute the ends they serve (or might serve).

Only after determining the means and the ends they serve does eunomics attempt to answer whether they ought to be pursued.

Fuller characterizes eunomics' approach as one where first we construct an apparently ethically neutral framework to determine which means serve which ends and only later assess whether those ends ought really to be pursued.

This is a sense in which eunomics could be considered "technological natural law" according to Fuller.
> he objection to which I refer is that the analysis is upside down; it seems to begin with means and concludes with the ends they serve. It starts by inquiring, What are the ways open to human beings to arrange their mutual relations so as to achieve their individual and collective ends, whatever those ends may be? This question is pursued at some length, and a good many distinctions are taken, before any serious attention is devoted to the problem whether the ends sought are such as ought to be achieved. Ethical judgments are postponed until a framework has been constructed for them by an analysis that appears to be ethically neutral. In this sense the book seems to exemplify a sort of “technological natural law,” to use an epithet that has been applied to previous tentatives in the same general direction.

(Fuller, 2001, p. 63)One possible objection against this approach is that we cannot theorize adequately about what the means of social arrangements are means _for_ if we don't start with the ends first.

> It may be said against this procedure that every social arrangement or institutional practice is a means to some end. We cannot pass on its effectiveness, nor determine what form it should take, until we know precisely what that end is. It is all very well to eschew absolutes, but clarity about ends, even if they be only ends “of the middle range,” is essential in any serious analysis of the forms of social order. And this clarity about the precise end sought must exist at the outset, for without it there exists no subject matter either for disinterested analysis or for moral judgment.

(Fuller, 2001, p. 63)In order to begin to answer this objection, Fuller makes reference to a passage of Mill where he argues that in science and mathematics we usually get to the most fundamental truths at the end of our investigations, not at the beginning.

For instance, we usually get to considerably elaborate calculations first and only then to accounts of the nature of numbers.

> ill observes that in science and mathematics the paradoxical situation obtains that clarity about fundamental concepts generally comes at the end of an analysis, not at the beginning. In mathematics, for example, elaborate calculations were possible for a long time before men even raised such questions as the nature of number itself. Mill continues: But though in science the particular truths precede the general theory, the contrary might be expected to be the case with a practical art, such as morals or legislation. All action is for the sake of some end, and rules of action, it seems natural to suppose, must take their whole character and color from the end to which they are subservient. When we engage in a pursuit, a clear and precise conception of what we are pursuing would seem to be the first thing we need, instead of the last we are to look forward to.2

## 2. The Analogy of Architecture
(FULLER, 2001, p. 64) In order to reply to the objection above, Fuller draws an analogy with architecture. According to him, architecture could be considered a practical art which exists for the satisfaction of certain human ends, such as utility and beauty.

The means to those ends are materials such as cement, lumber, steel, etc., as well as the skills necessary to assemble them accordingly.
> Let me paraphrase and extend Mill’s argument as it might be applied to architecture. Architecture is nor an abstract science but a practical art. it exists for the satisfaction of certain human ends, which may be described as utility and beauty. The means to those ends are materials such as cement, lumber, and steel to which must be added the technical skill necessary to assemble them. All of these means are subservient to the ends of utility and beauty. 1n jury particular structure they take their character and color from the particular lkinds of utility and beauty sought in designing that struCture.

(FULLER, 2001, p. 64) According to this objection, a good account of architecture has to start with a satisfactory account of beauty and utility and only then elaborate on the means that are relevant to architecture.
> It therefore fol- lows that the study of architecture must begin with ends, with a definition of utility and beauty, for it is only when these ends have been clarified that it is possible to deal intelligently with means, or even to know what means are relevant to the obiects of architecture.

(FULLER, 2001, p. 64) Fuller starts by saying that this argument does not seem to be convincing but that an analysis of _why_ it does not seem convincing is interesting to his aims.

Before that, however, he also adds that there is a reply to this argument that seems obviously false, but which is commonly accepted when we are talking about _social architecture_.

The argument goes as follow: (1) we have to start with the ends of architecture and only then go to the means of it; (2) we cannot define beauty and utility adequately; (3) therefore, architecture does not exist.
> It will be worthwhile to analyze the reasons why this argument is uncon- vincing. At the outset it may be observed that one particular response to it would scarcely be taken seriously. This is a response that would accept the primacy of architectural ends over means and proceed then to argue that, since utility and beauty cannot be defined, architecture itself does not exist. Para- phrased into terms of social architecture, the argument has seemed persuasive to many. As applied to physical architecture it scarcely rises to the level of a bad joke.

(FULLER, 2001, p. 64) A more serious reply to this objection would be to say that discussing ends in abstraction is less productive than discussing the means which are available, because we must first know what is possible and only then discuss what is desirable. This is true both pragmatically as well as a matter of "ought implies can".
> A more serious response to the suggested primacy of architectural ends over means would be to point Out the futility of discussing ends in abstraction from available means. We must know what is possible before discussing what is desirable. A building suspended in mid-air might have a certain esthetic ap- peal, but since there is no means for constructing such a building we may dismiss it from consideration, or at best turn our minds to the problem of creating the most effective illusion of such a building by the means at hand.

(FULLER, 2001, p. 65)Fuller also reminds us that _some_ limitation of means is essential to creativity. He quotes Stravinsky on the importance of limiting oneself for the sake of creativity on this point

> It is an obvious economy of thought, therefore, to survey available means before addressing ourselves to ends. But the point goes deeper. Some limitation of means is essential to liberate the creative spirit. | know of no better way to make this point than to quote at some length from Igor Stravinsky’s Poetics of Music: The creator’s function is to sift the elements he receives from [imagination], for human activity must impose limits on itself. ... As for myself, I experience a sort of terror when, at the moment of setting to work and finding myself before the infinitude of possibilities that present themselves, I have the feeling that everything is permissible to me. If everything is permissible to me, the best and the worst; if nothing offers me any resistance, then any effort is inconceivable, and | cannot use anything as a basis, and consequently every undertaking becomes futile. ... What delivers me from the anguish into which an unrestricted freedom plunges me is the fact that I am always able to turn immediately to the concrete things that are here in question. I have no use for a theoretic freedom. Let me have something finite, definite—matter that can lend itself to my operation only insofar as it is commensurate with my possibilities. And such matter presents itself to me together with its limitations. I must in turn impose mine upon it.... [I]n art as in everything else, one can build only upon a resisting foundation; whatever constantly gives way to pressure, constantly renders movement impossible.

(FULLER, 2001, p. 65-66)Having means as our starting point does not imply that we should have no image of ends at the beginning, however.

We usually should have some vague conception of the ends. In the case of architecture, for instance, a certain amount of stubborn pursuit of the ends of beauty and utility could sometimes lead to the creation of means that didn't previously exist.

> But because it is a mistake to assign an unconditional primacy to ends over means in thinking about creative human effort, it does not follow that the mistake can be corrected by a turn of a hundred and eighty degrees. Because our architectural paraphrase of Mill’s argument is false as stated, it does not follow that it can be corrected by reversing every assertion in it. Some vague conception of architectural ends at the outset is essential to define the range of means worthy of consideration for architectural purposes. In the development of architecture it is safe to assume that an obstinate quest for new forms of beauty and utility has sometimes led to the discovery of means capable of realizing them, so that firmly held ends stubbornly pursued can sometimes create means that were previously nonexistent.

(FULLER, 2001, p. 66)Fuller claims that in order to develop eunomics, we need a more exact account of the role of the means-ends relation in human thinking and whether it has a peculiar function in the design of social institutions

> What is needed is obviously a more exact analysis of the role played in human thinking by the means-ends relation. We shall also need to explore carefully the peculiar function performed by this category of thought in the design of social institutions.

### 3. Means and Ends in Social Architecture
(FULLER, 2001, p. 66-67)Fuller quotes a passage of Aldous Huxley where he claims that there general agreement about human ends, but not about the means which best serve them
> With regard to the goal, I repeat, there is and for long has been a very general agreement. Not so with regard to the roads which lead to that goal. Here unanimity and certainty give place to utter confusion, to the clash of contradictory opinions, dogmatically held and acted upon with the violence of fanaticism.‘

(FULLER, 2001, p. 67)Fuller quotes a passage from Isaiah Berlin where he talks about the deep disagreement about human ends and how where this disagreement ends, the only questions left are those of the appropriate means, which are not political, but rather technical questions.

> If men never disagreed about the ends of life, if our ancestors had remained undisturbed in the Garden of Eden, the studies to which the Chichele Chair of Social and Political Theory is dedicated could scarcely have been conceived. For these studies spring from, and thrive on, discord. Someone may question this on the ground that even in a society of saintly an-archists, where no conflicts about ultimate purpose can take place, political problems, for example constitutional or legislative issues, might still rise. But this objection rests on a mistake. Where ends are agreed, the only questions left are those of means, and these are not political but technical, that is to say, capable of being settled by experts or machines like arguments between engineers or doctors.°

(FULLER, 2001, p. 68)Fuller focuses primarily on institutions, although his remarks apply to other kinds of human orders with adjustments

> In what follows I shall speak primarily of institutions, in which I mean to include legal, political, and economic institutions, as well as those which do not fall readily under any of these rubrics. Most of what | shall have to say will apply, with some modification, to social arrangements which would not ordinarily be described as institutions, such as relatively transient orderings of human relations.

(FULLER, 2001, p. 68)The first common assumption about social orders that Fuller wants to reject is that ends that human institutions serve are severable, in the sense that we can account for them separately and that explaining it can be done without taking into account the other aims which are served by human institutions

> he first common mode of thought I want to examine is one which supposes that the ends served by social institutions are severable, that they stand as distinct entities, each capable of a separate appraisal. If we say that an institution takes its whole character and color from the end to which it is subservient, we assume its end to be something that can be severed from those of other institutions, so that a clear conception of its end is a necessary and sufficient condition for an understanding of any institution. Even if we recognize (as we certainly should) that most institutions serve more than one end, we may still assume (as I think we should not) that the several ends served remain essentially discrete, having been assembled merely for reasons of economy

(FULLER, 2001, p. 68)Against this assumption, Fuller argues that human aims or ends usually arrange themselves in "circles of interactions", instead of being a "neat row of desired 'end states'".

> Against this way of thinking we must set the plain fact that human aims and impulses do not arrange themselves in a neat row of desired “end states.” Instead they move in circles of interaction. We eat to live and we live to eat. We love that we may be loved, and we want to be loved that we may love freely. The pattern of our private desires reflects itself in the pattern of our social institutions.

(FULLER, 2001, p. 68)We cannot save this assumption by seeing institutions simply as a kind of social order which serves several ends in a direct way (Fuller uses the metaphor of a "conduit directing human energies toward some single destination").
Instead, we should see institutions as something which reshapes our several ends in different ways and degrees.

> We should not conceive of an institution as a kind of conduit directing human energies toward some single destination. Nor can the figure be rescued by imagining a multipurpose pipeline discharging its diverse contents through different outlets. Instead we have to see an institution as an active thing, projecting itself into a field of interacting forces, reshaping those forces in diverse ways and in varying degrees.

(FULLER, 2001, p. 68-69)According to Fuller, we cannot simply ask whether some given institution serves a good end adequately.

Rather social institutions make human life different than it would have been without it, which includes the set of aims we seek.

To put it differently, institutions for Fuller seem to interact with other social arrangements and make a difference in our moral landscape.
As such, we should rather ask whether a given institution creates a "pattern of living" that is "satisfying and worthy of a man's capacities"

> A social institution makes of human life itself something that it would not otherwise have been. We cannot therefore ask of it simply, Is its end good and does it serve that end well? Instead we have to ask a question at once more vague and more complicated—something like this: Does this institution, in a context of other institutions, create a pattern of living that is satisfying and worthy of man’s capacities?

(FULLER, 2001, p. 69)The second assumption Fuller wants to reject is that the _first_ task of social philosophy is to arrange human ends in a hierarchical order.
This assumption follows from the other idea just rejected, namely, that ends are severable.

I think one could object that even though it is true that it is not the _first_ task of social philosophy to arrange human ends, it still does not follow that we cannot attempt to assess which ends are more "satisfying and worthy of a man's capacities" than others.

Of course such a hierarchical account would have to consider how different ends modify how we interact with different means and ends, i.e. how they make a difference in our lives and in our moral outlook as a whole, but this does not make the task of constructing such a hierarchy of ends impossible or uninteresting.

> A second mode of thinking that seems to me both common and mistaken is that which assumes that the first task of social philosophy is to arrange human ends in a hierarchical order. This conception follows rather naturally from the fallacy just discussed. If particular social means are subservient to distinct and diverse social ends, so that in each instance the end sought must first be precisely defined before it is possible to select an apt means for achieving it, then it follows that the first task of institutional design is to draw up a schedule of ends in the order of their urgency. This hierarchical schedule then becomes a master plan into which the details of social architecture can be fitted.

(FULLER, 2001, p. 69)The construction of a hierarchical list of human ends, Fuller argues, meets several difficulties, such as:
- does consensus about End A being more important than End B entail that A is in fact more important than B?
- does the higher importance of A over B imply that every decision in favor of A should be preferred over decisions which promote B? Or should the costs of this trade-off be properly considered or perhaps balanced?
- can we know how much of A we really want before we can acknowledge how much a given loss in B would affect us?

> hough such a table of priorities is often supposed to be indispensable, it is in fact never really drawn up, though verbal facsimiles are sometimes offered. There are reasons why this default does and must occur. Let us suppose a consensus that End A is generally “more important” than End B. Can we on the basis of this agreement place End A ahead of End B on our schedule of urgency? Before we can do this we must know at least what such a listing would imply. Would it mean, for example, that if the two ends come into conflict any satisfaction of A, however small, is to be preferred to any satisfaction of B, however complete? Can we really know how much we want A and B respectively until we know how much each will cost in terms of a sacrifice of other ends?

(FULLER, 2001, p. 69)Fuller thinks that the different ways in which institutions are configured imply different changes in the ends which we seek and which are served by social institutions.

For Fuller, it follows from this fact that the project of defining priorities prior to the availability of means is doomed to fail. As he adds, "a social end takes its 'character and color' from the means by which it is realized".

> If we try to solve this problem we shall have to abandon our project of defining priorities and explore the problem of available means. In this exploration we shall discover two things: First, no abstractly conceived end ever remains the same after it has been given flesh and blood through some specific form of social implementation. Reversing Mill, we may truthfully say that a social end takes its “character and color” from the means by which it is realized.

(FULLER, 2001, p. 69)Aside from "means-costs" relations ("how much will the choice for means X, which prioritizes End A, cost for the attainment of End B?") we also have to take into account the so-called "means-surplus" relation, that is, the fact that adopting a given means Y, which primarily serves and A, can also generate the satisfaction of B as a by-product.

> Second, we must appraise not only means-cost (How much will End B be sacrificed to End A if we realize the latter through Means X?) but also what may be called means-surplus. Can we discover a means of realizing End A that will yield, as a kind of by-product, some satisfaction of End B?

(FULLER, 2001, p. 69)Fuller argues that these complexities should lead us to the conclusion that the project of constructing a hierarchical list of ends is not the proper place to _start_ in dealing with problems of social architecture.

I would add that the key expression here is "proper place to start". It is not the case that constructing a hierarchy of ends is impossible, but that we cannot carry out this project prior to an adequate assessment of available and actual means.

> hen these complexities are taken into account it becomes plain that the so-often recommended hierarchical ordering of ends is not the proper place to start in dealing with problems of social architecture.

(FULLER, 2001, p. 69)Fuller considers the objection according to which at least _one_ end could be ordered prior to the consideration of other available means, namely, life.

This would be so because no other end could be meaningful without human life.

> There is one superficially persuasive defense of such an ordering that requires at least passing mention. It can be said that without human life, no other human end can be meaningful. It therefore follows that the preservation of human life must have the highest priority in the design of any social edifice.

(FULLER, 2001, p. 70)Against this objection, Fuller replies that (i) it does not help us when we have a case of "life against life"; (ii) we in fact tend to value life less than the attainment of other ends in some cases, such as in war or the construction of a transcontinental highway (where actuarial science can predict with some accuracy the cost in human life to take place for the attainment of a given entreprise).

> An easy answer is to point out that such a table cannot solve the problem presented where life is inexorably pitted against life, as in the ancient example of the two shipwrecked sailors holding to a single plank capable of supporting only one of them.* The true objection, however, runs deeper. We do in fact undertake enterprises, not necessarily those of war, knowing full well their cost in human life. Actuarial science could predict with considerable accuracy, for example, the lives likely to be lost in constructing a transcontinental highway.

(FULLER, 2001, p. 70)The third assumption with which Fuller takes stock is the so-called "infinite pliability of social arrangements"

> The next common assumption I believe to be mistaken is a belief in what may be called the infinite pliability of social arrangements. Some such belief must in fact lie back of the assumption J have just been discussing, the assumption, namely, that the first task of social philosophy is to establish a hierarchy of social ends.

(FULLER, 2001, p. 70)The infinite pliability of social arrangements is the notion that social institutions can be shaped to any end and through any available means.
Fuller's point against this assumption is that it is always assumed that the implementation of possible social institutions to any end is never adequately demonstrated and always taken as just a "matter of technique", although the features by which we could identify someone as a good technician of human social orders are never identified.
I am not sure whether this is a good objection. Elaborate later.

> he assumption that social institutions can always be shaped to any desired end reveals itself in the common notion that implementation is a mere matter of “technique.” Curiously, though the technicians capable of devising the apt means for social ends are never identified, it seems to be assumed that their competence is unlimited. There are signs by which we know a good carpenter, one of them being his knowledge of the limitations of the materials with which he works. But the technician in social implementation seems to remain at once anonymous and omnicompetent.

(FULLER, 2001, p. 70)One limitation that Fuller cites against the assumption of infinite pliability of social arrangements is the "requirement of simplicity"

> I shall have occasion later to discuss the limitations intrinsic to the process of designing institutions and procedures to implement social ends. One such limitation may be mentioned in passing, and that is the requirement of simplicity.

(FULLER, 2001, p. 71)As Fuller highlights, social institutions are not just a matter of technique, because in many cases social tasks have also to be "manageable". What is suggested here, therefore, is that although some social arrangements could be technically possible, they can be so complex and complicated to the point of being pragmatically unfeasible or perhaps doomed to fail.

> It takes something more than a rub of the technician’s lamp to bring into existence a social procedure apt for the solution of any given problem. And we must not overlook the possibility that a particular social task may turn out to be, in Michael Polanyi’s words, unmanageable.’

(FULLER, 2001, p. 71)The fourth common assumption that Fuller challenges is the idea that only means have limitations, whereas ends would be preferences to be chosen

> The fourth common notion that I should like to challenge is one which assumes that elements of formal structure are to be found only in social means, not in ends. Means we design, ends we merely choose. Drawing up a hierarchy of ends is therefore not an exercise in social architecture, but a ranking of preferences. The element of formal design is supplied by our anonymous friend, the expert in means, the social technician—the specialist in implementation.

(FULLER, 2001, p. 71)Fuller thinks this understanding of morality in terms of values to be the object of our preferences is tributary to Nietzsche and that it is not a good account of morality.
The flaw in this account, it seems, lies in its failure to acknowledge the "element of formal structure" of ethics. Fuller's point is not free com obscurity, but I think it has to do with the relationship between ends and the social implementability of them.

> his mode of thought finds expression in, and at the same time is encouraged by, the now common usage which assigns to the word value the sense of a moral objective—a usage that appears to be chiefly owing to the influence of Nietzsche.'!° What used to be ethics is now often called value theory. Where we once judged a man by the cast of his character, we now inquire about his “value preferences.” There is nothing about the word value that suggests any element of formal structure.

(FULLER, 2001, p. 71-72)Fuller highlights how this account of ends is profoundly mistaken. If we want social goals to be meaningful, Fuller claims, we have to conceive a life "worthy of emulation" in "structural terms", by which he means in terms of patterns of relations which takes into account the forms and formal structures in which some ends are served.

> ll of this line of thought is, I believe, profoundly mistaken. Any social goal, to be meaningful, must be conceived in structural terms, not simply as something that happens to people when their social ordering is rightly directed. In individual morality, | think it safe to say that the life worthy of emulation is one that forms a coherent pattern, not one that is manipulated to bring about a series of desirable states of mind and body. So I believe it is with society.

(FULLER, 2001, p. 72)The fifth common assumption that Fuller wants to challenge is that means are not in themselves valuable and that we would be better off if we could achieve the ends they serve without bearing the "costs" related with bringing about those means.

> A fifth and final mode of thought that is here rejected has already been suggested in passing. This is the notion that social means—institutions, procedures, rules—are necessary evils and that the world would be better off if their costs could be avoided, that is, if social ends were attainable directly, without introducing any intervening rigidities of social structure.

(FULLER, 2001, p. 72)Fuller thinks that this notion does not take into account the notion of means-surplus (or means-bonus)

> It will be sufficient to suggest, as a counterbalance to the notion of means-cost, the notion of means-surplus or means-bonus.

(FULLER, 2001, p. 72)Fuller argues that our "best institutions" are "pregnant with these beneficial side-effects". That is, they aren't simply conduits to desirable-end-states but "part of the pattern of our lives".

I think the point of Fuller is clearly stated by the examples of families and friendship groups. It does not seem the case that we just compose them to achieve certain ends, but rather that the achievement of these relationships _is_ the end.

> Our best institutions are, I believe, pregnant with these beneficial side-effects. It is chiefly for this reason that I have so vigorously objected to the view that institutions are mere inert conduits directing human energies, with much frictional waste en route, toward certain desirable end-states. Our institutions are a part of the pattern of our lives. The task of perfecting them furnishes an outlet for the most vigorous of moral impulses.

(FULLER, 2001, p. 73)As an example of this fifth assumption, Fuller mentions how, despite a steady expansion of the functions of government, freedom is taken to be an end of human social life which is at the top of an assumed hierarchy of human ends because of its intrinsic value and not because of any structural relation with other ends and means.

> Despite a steady expansion of the functions of government, large and small, formal and informal, during the last hundred years, freedom as an objective of social policy is still conceived largely in Mill’s terms, as a mere absence of constraint. Its relations with other social goals, and in particular with that of security, are scarcely mentioned. In a tacitly assumed hierarchy of ends, freedom is accorded a high position—perhaps the highest position—this ranking being assigned to it, not because of any assumed structural relation with other ends, but because of an attributed intrinsic value.

(FULLER, 2001, p. 73)Fuller cites a passage of James Fitzjames Stephen in which he replies to Mill. Contra Mill, Stephen would argue that freedom should not be studied _in abstracto_ or as the ultimate and intrinsic goal of social arrangements.

Stephen traces an analogy between the rules and structure of social arrangements as the set of human energies directed toward certain goals and a pipeline which directs water to places it can be most effective according to a relevant set of interests.

Stephen argues that _if_ we want to know what should be, for instance, the size of the hole of water pipes and their direction, _then_ we should be interested in the nature of water, the nature of pipes and the objects for which water is desired, but that we would learn rather little by studying the nature of holes, as their nature is "simply to let the water pass".
If we would elaborate further on this analogy, I think we could say that Stephen would claim, contra Mill, that studying the nature of liberty abstractly would be counterproductive or even misguided. Liberty (or freedom, for Fuller) and the correct amount thereof cannot be theorized apart from the other ends which are valuable in human society and apart from the means which best realize them.

> ll of this is well expressed by James Fitzjames Stephen in his famous but little-read reply to Mill. Stephen is invoking his favorite analogy by which the rules of social living are thought of as bringing together and directing the flow of human energies, as pipes direct water to places where it can be most effective. The phenomenon which requires and will repay study is the direction and nature of the various forces, individual and collective, which in their combination or collision with each other and with the outer world make up human life. If we want to know what ought to be the size and position of a hole in a water pipe, we must consider the nature of water, the nature of pipes, and the objects for which the water is wanted; but we shall learn very little by studying the nature of holes. Their shape is simply the shape of whatever bounds them. Their nature is merely to let the water pass, and it seems to me that enthusiasm about them is altogether thrown away.'*

(FULLER, 2001, p. 74)Fuller mentions how Michael Oakeshott also points out something similar when he says that the freedom a given society enjoys is dependent upon the form arrangements for its realization and that, more than that, freedom _just is_ the availability of those procedures.

Fuller thinks that Oakeshott overstates this point a bit, although he does not elabore much why.

> Stephen is by no means alone in perceiving the fallacies that underlie so much of the literature of freedom. Michael Oakeshott, in particular, has emphasized the extent to which freedom, conceived as an object of social policy, is dependent upon formal arrangements for its realization. This aspect of the problem he expresses well—perhaps I should say he overstates effectivelywhen he writes: “The freedom which we enjoy is nothing more than arrangements, procedures of a certain kind: the freedom of an Englishman is not something exemplified in the procedure of Habeas Corpus, it is, at that point, the availability of that procedure.” '5

(FULLER, 2001, p. 74-75)Fuller makes the brief but interesting point point that those assumptions which he rejected appear to encourage and legitimate what he calls the "salesmanship of value preferences", a view which makes it difficult to distinguish between serious efforts of analysis and propaganda.

> One of the strongest objections to the assumptions that have been rejected in this chapter lies in the fact that they encourage and appear to legitimate what may be called the salesmanship of value preferences. The general acceptance of these assumptions creates an atmosphere in which it becomes difficult to make any distinction between serious efforts of analysis and what are essentially exercises in propaganda.

## 4. The Problem of Equality
(FULLER, 2001, p. 75)Fuller mentions Jeremy Bentham as one philosopher who tried to trace a hierarchy of social ends where pleasure occupied the highest place and security and equality were subsidiary objectives.

> In the development of his utilitarian philosophy Jeremy Bentham made some tentatives in the direction of a hierarchical ordering of social ends, in which perforce pleasure had to occupy the highest place. In particular he discussed the proper ranking of security and equality as subsidiary objectives.'

(FULLER, 2001, p. 75)Bentham makes an interesting point that, according to the law of marginal utility, we should promote equality of income as a strategy to achieve happiness in terms of utility, since the utility lost by the rich would be much less than the utility gained by the poor. However, we still should not favor a policy of equality of income because the need for further and repeated interventions in order to maintain such equality would destroy security, which would be necessary for happiness.

> From this law of the diminishing utility of money it would seem to follow that Bentham’s highest good, namely, happiness, would be promoted by legislative measures deliberately directed toward equality of income. What the rich would lose by such a measure would be more than offset by what the poor would gain, measuring losses and gains not in pounds but in units of human happiness. But Bentham rejected the notion that total happiness would be increased by equality on the ground that in order to achieve and maintain equality repeated interventions by the state would be necessary, and such interventions would be destructive of security, itself an ingredient of happiness.’?

(FULLER, 2001, p. 76)Ross criticizes Bentham for not seeing how this new arrangement of social life that prioritized equality of income could be brought about with a new mode of division which presupposed other conceptions of security

> Ross in comment on Bentham’s argument writes: The remarkable thing about this chain of reasoning is that Bentham does not mean by “insecurity” the disturbance and disappointment produced by a revolution in the rules which govern the distribution of ownership, a disturbance which takes place once and for all. Rather he plainly has in mind that equalization can be effected only through a series of repeated interventions in the rights of property, whereby something that is mine is transferred to you. Accordingly he assumes that an equalization in the distribution of wealth requires permanently recurring redivisions and continuing insecurity—in other words, a condition without order or stability. Apparently it never even occurred to Bentham that when a new mode of distribution was once introduced that which under the former order would have become mine would no longer become mine, and that the new mode of division could obviously be maintained with the same security as the old.?°

(FULLER, 2001, p. 76)Fuller points out how Ross' comment on Bentham exemplifies the assumption of infinite pliability of social arrangements: this would be the case because, as commented above, Ross criticizes Bentham for not seeing how this new arrangement of social life that prioritized equality of income could be brought about with a new mode of division which presupposed other conceptions of security. However, as Fuller emphasizes, he does not suggest any particular set of institutions which could achieve these objectives.
I think Fuller's point is a bit uncharitable. We could interpret Ross' point as advancing the claim that _some_ set of social rules and institutions could be constructed in a way to further _some_ conception of security (according to this new set of rules) which would assure the attainment of equality of income or, to put it in another way, that we could attain the goal of equality of income once we realize that human conduct could be regulated by other set of social rules than the one which regulates it currently.

Of course, if we are to make the claim that such a social arrangement is possible and, more than that, viable, we have the burden of constructing a case (i) that it is at least possible and viable and (ii) ideally advance a normative account of those social institutions which would assure the goal of equality of income, but this is not to say that Ross' is holding such a strong assumption as Fuller thinks.

> o we not have exemplified in Ross’s criticism what I have called the fallacious assumption of the infinite pliability of social arrangements? In judging that question it is important to realize that neither Bentham nor Ross are addressing themselves to the old saw (perhaps not wholly without teeth) that after an equal division of wealth the shrewd and industrious will always find some way (including a change in the rules) of coming out on top. What Bentham and Ross are discussing is the possibility of a legal and economic order by which newly produced wealth would, as it came into being, be channeled so as to achieve and maintain an equal distribution. Bentham, whose mind was always intent on problems of implementation, apparently could not conceive how this could be done, or rather he could conceive of no way of doing it that would not require repeated redistributions. Ross, on the other hand, though he does not suggest any particular set of institutions that could achieve the desired objective, apparently assumes that any social goal can be given suitable implementation. Along with this assumption, there goes its corollary, that it is possible to define a social objective in abstraction from the means of realizing it

(FULLER, 2001, p. 76)Fuller emphasizes how it is _difficult_ to conceive a social objective such as equality without previously having an account of what equal treatment is and how it can be administered.
I think we should not overstate Fuller's point and ideally interpret it literally: it is _difficult_, since any account of social goals has to take in consideration how those goals are affected by other goals and by how our current social arrangements partially affects our moral outlook, _but_ this is not to say that a morally robust concept of equality has to be necessarily grounded to the kinds of social institutions we have today (we could claim that the kind of concept of equality which we value and which matters actually requires the rejection and abolition of most of our current social arrangements and their reconstruction.

> It is difficult to conceive a social objective of which these assumptions are less true than they are of equality. Until we find some means by which equal treatment can be defined and administered, we do not know the meaning of equality itself.

(FULLER, 2001, p. 78)Fuller claims that this assumption, the "infinite pliability of social arrangements" to any social goals is commonly encountered in treatises of economics, sociology and political science.

He mentions a passage from "Boundaries and Limits to Economics" where it is affirmed that when we come to the question of the desirable distribution of wealth, we are leaving the field of economic science and that those are matters for the philosopher, the theological, the statesman, etc.
Fuller thinks we can imagine that the author of this passage took it for granted that those with the task of setting the appropriate social ends would consult the economist to assess the appropriate and the economics costs of each end.

> When we come to the. ..question of the desirable distribution of wealth and income between individuals, we leave the field of science altogether. De gustibus non est disputandum: there is no disputing (scientifically!) tastes; and the same goes for ethics. We must leave the definition of social ends to the philosopher, the theologian, the statesman, and to public opinion.:! It may be assumed that the author of this passage took it for granted that those setting social ends would consult the, economist to learn the “economic cost” of what they proposed. Thus, the setting of ends would at least require collaboration between the end-setter and the means-specialist.

(FULLER, 2001, p. 78)Even then, however, Fuller thinks we encounter two difficulties with this division of labour.
(i) it is unclear how those untrained in economics could understand an explanation of all the economics problems involved in realizing a given social goal

(ii) the economic cost and the economic efficiency of the available means are themselves inserted in the acceptability of certain means-ends relations.
I think point one is a bad objection, since this is just a question of adequate translation of fields of knowledge.

> There are, however, two difficulties with this proposed division of labor. In the first place it is difficult to see how the philosopher or theologian, untrained in economics, could understand an explanation of all the economic problems involved in realizing, let us say, the objective of equality. The second difficulty is more fundamental. The concepts of “economic cost” and “economic efficiency” are themselves predicated upon the acceptance of certain means-ends relations.

(FULLER, 2001, p. 78)Fuller makes a strong claim that economics takes at its core a kind of ordering which he calls the "regime of reciprocity".

By this, Fuller means that when an economist talks about a given economic cost, he is talking about a set-back in the achievement of a particular social order, which means that the economist too is engaged in the business of end-setting.
I am really skeptical of this claim. I think the kind of endeavor in which the economist is engaged is better understood as one where making a claim about economic costs means making a claim about the set-backs in the _satisfaction_ of various social goals which are pursued by people as individuals and as different sorts of groups.

I find it hard to accept that the economist is engaged in any "business of end-setting" in any strong sense of the word.

> Thus I would argue that economics takes as its province a certain kind of social ordering which may be called broadly a regime of reciprocity.22 An “economic cost” is, then, a set-back in the achievement of this particular form of social order. The economist means-expert, who warns of certain “economic costs,” is therefore necessarily himself engaged in the business of end-setting.

___
FULLER, Lon L. Means and Ends. *In*: WINSTON, Kenneth I. **The Principles of Social Order**: Sellected Essays of Lon L Fuller. Ed. rev. Oxford: Hart Publishing, p. 61-78, 2001.