SORITES, ISSN 1135-1349

Issue #10. May 1999. Pp. 19-38.

Critical Comments on Laudan's Theory of Scientific Aims

Copyright © by SORITES and Armando Cíntora

Critical Comments on Laudan's Theory of Scientific Aims

Armando Cíntora


Larry Laudan has proposed in Science and Values a meta-methodology of science which attempts to avoid historical relativism and a relativism of scientific methods by providing a rational justification for the factual, methodological, and axiological aspects of scientific change.Foot note 3_1 He argues that if relativism is to be avoided cognitive aims, theories and methods, should be capable of rational adjudicationFoot note 3_2. Laudan argues that previous philosophers such as Popper, Carnap, Hempel and Reichenbach «opened themselves up to the relativist challenge», either because these philosophers considered the methods of science a matter of convention, or because like Reichenbach they thought that the aims of science are selected by `volitional decisions', or because they thought -allegedly like Popper -- that the only thing one could rationally ask of a set of cognitive aims is for this set to be internally consistent.Foot note 3_3

Laudan tries to provide a rational account of the development of science through a reticulated model in which justification is multi-directional, and in which scientific theories, methods and aims change during the history of science.Foot note 3_4 Temporarily accepted methods justify the theories of the day, and are justified by temporarily accepted aims. But these methods, in their turn, can also be changed by factual theories, while empirical theories and methodological rules also constrain the set of rationally possible cognitive aims. Hence there is a mutual and typically non-simultaneous adjustment and justification among factual theories, methods and ends. And none of these three levels constitutes an ultimate or even a favored or more solid ground.

Rationality is for Laudan about searching, and having good reasons, for believing one is following the most effective means for the attainment of certain ends that one has chosen. It follows, given this view of rationality, that the methodological rules of science are elliptical means-ends injunctions, `hypothetical imperatives', of the form: if you value, or desire `A', then you should do `X'. And since experience informs us which are the best means for our independently chosen ends, then methodological rules are fallible, corrigible and improvable via past or present experience. Since Laudan himself recognizes his methodological rules as hypothetical imperatives, if he is to avoid relativism, then he must tell us how to rationally select the desiderata in these conditionals' antecedents, the cognitive aims `A.' This because if the aims of science, the A's, could not themselves be rationally selected, if any cognitive aim were as legitimate as any other, then these aims could legitimate any conceivable methodological rule, and ultimately these aims could legitimate any substantial theory, thus opening the gates to a radical cognitive relativism.

A `scientific' creationist, for example, could propose as the central aim of science that of finding explanatory theories consistent with a literal reading of the Torah. And if this cognitive aim were to be scientifically legitimate scientists would have as central endeavours the search for, and elimination of, inconsistencies between scientific theories and the Biblical text, and scientists would search for an accurate translation and reading of the Torah. Creationism's central aims and methods would disqualify contemporary geology, paleontology and evolution theory while legitimizing the Genesis account.

Laudan himself admits that his reticulated view needs to be supplemented by a theory of legitimate aims -an «axiology» as he himself calls it. And in Science and Values he has given some hints on how to develop such an axiology. I will explore Laudan's suggestions on how to decide rationally between competing scientific aims, and whether these suggestions can avoid relativism.

Laudan's Theory of Aims

Laudan hinted in Science and Values -- and in other more recent worksFoot note 3_5 -- the view that our scientific aims can sometimes be rationally appraised by asking that they satisfy three constraints:

1) A pragmatic constraint of empirical realizability, or non-utopianism, this requisite is thought to follow from a means-ends perspective of rationality,

To adopt a goal with the feature that we can conceive of no actions that would be apt to promote it, or a goal whose realization we could not recognize even if we had achieved it, is surely a mark of unreasonableness and irrationality. (Laudan, 1981, p. 51) (Emphasis added)

Laudan believes that if one is means/ends rational then one cannot have `utopian' aims, because utopian aims are of no help in selecting means. Laudan is hence allegedly only making a conditional recommendation against utopian aims (if you will be rational, then avoid utopian aims), though he is possibly really making an implicit categorical recommendiation against utopian aims. This because in this last quote there is an implicitFoot note 3_6 recommendation to be means/ends rational, and therefore there is also an implicit categorical recommendation to avoid utopian goals.

A goal, for Laudan, can be `utopian' in three ways:

First, a goal is demonstrably utopian when

it cannot possibly be achieved, given our understanding of logic or the laws of nature... (Laudan, 1981, p. 52) (Emphasis added)

It would be utopian, for example, to aim in an infinite or immense cosmos, for certainty about empirical universal statements. And one way to find out whether some goals are achievable is to search the historical record to see if our goals have been, and therefore can be, achieved irrespective of whether they were consciously sought or were merely unintended consequences of some actions.

Second, a goal might be semantically utopian:

Many scientists espouse values or goals that, under critical challenge, they cannot characterize in a succinct and cogent way. They may be imprecise, ambiguous, or both. Such familiarly cited cognitive goals as simplicity and elegance often have this weakness, because most advocates of these goals can offer no coherent definition or characterization of them. (Laudan, 1981, p. 52)

Another example, might be verstehen, and epistemic coherence.

Third, a goal might be epistemically utopian:

It sometimes happens that an agent can give a perfectly clear definition of his goal state and that the goal is not demonstrably utopian, but that nonetheless its advocates cannot specify (and seem to be working with no implicit form of) a criterion for determining when the value is present or satisfied and when it is not. (Laudan, ibidem.) (Emphasis added)

Laudan thinks that truth, understood as correspondence, is an example of epistemic utopianism,Foot note 3_7 therefore Laudan believes truth is an irrational goal. And this, even though, the search for explicative truth has been highly valued by many scientistsFoot note 3_8, and even though for truth we do have fallible criteria for determining when truth is absent, and even though, we also have fallible criteria -- such as intersubjective consensus -- for deciding when truth maybe present (or at least, we have fallible criteria for rational belief.)

Notice that there is an ambiguity in Laudan's last quote, since it is not clear what to understand by a `criterion'. Is a criterion something everyone in a scientific community is to agree on? Must the criterion be infallible? Is it enough to have a criterion for determining when one approximates the goal, even if lacking a criterion for determining the attainment of the goal? Since Laudan thinks that truth is epistemically utopian it seems that for Laudan a criterion must be infallible or at least consensual.

2) Laudan also asks scientific goals to be jointly consistent.

3) Laduan finally proposes as another constraint on scientific goals that these goals should be consistent with the `Tradition', that is with the canonical achievements of a successful scientific discipline.

Laudan's constraints of non utopianism and mutual consistency for scientific aims let in too much, that is, even if these constraints were to be sought and satisfied, one could still end with faulty scientific aims such as:

Look for theories in agreement with a literal reading of the Torah! Or, gather data at random! Or, seek false theories!

Laudan therefore further narrows the spectrum of possible cognitive aims by requiring that any proposal for new scientific aims must also be consistent with the scientific `Tradition'. We are told that new aims or standards, if acceptable, must be able to capture, to redescribe, the canonical achievements of a successful scientific discipline. And the success of any scientific `Tradition' is judged by some implicit pre-philosophical» pragmatic canons.

... any proposals about the aims of science must allow for the retention as scientific of much of the exemplary work currently and properly regarded as such. (Laudan, 1996, p. 158.) (Emphasis added)

We are being told that any proposals for scientific aims should retain as scientific «much of the exemplary work» of a scientific discipline. But how much is enough to retain? What of the `exemplary work' must be retained, and what may be omitted? And why what Laudan and many of us think of as `exemplary work' (say, the work of Newton, Maxwell, Einstein) is really exemplary? Foot note 3_9Laudan believes there are «pre-philosophical» pragmatic canons of scientific successFoot note 3_10. These canons are cognitive goals such as prediction and control, and these canons judge what is scientifically proper, they judge what is scientifically successful. Still, if Laudan is to avoid relativism he should justify these standards of success.Foot note 3_11

Are Laudan's Recommended Constraints for Cognitive Aims Adequate?

I will illustrate many of these criticisms with examples from non-cognitive ends.Foot note 3_12 Because we are often more acquainted with these other goals, and thus they provide a useful and clarifying analogy. There are analogies between cognitive aims such as the avoidance of ad hoc hypotheses, the search for verisimilar scientific theories, or to aim at simple or elegant scientific theories; and non-cognitive aims like the search for Buddhist Nirvana, the yearning for God, the `pursuit of happiness'Foot note 3_13, or goals such as wisdom, or love, in that all of these goals would be, according to Laudan, semantically and or epistemically utopian. While an aim such as perfect social justice is analogous to a cognitive aim such as complete truth in some field, or to full objectivity concerning some subject, in that all these aims cannot be achieved (`given our understanding of logic or the laws of nature') and so these goals would be, for Laudan, demonstrably utopian.

If it were to be argued that examples involving non-cognitive aims are misconceived because Laudan's theory is intended only for cognitive aims, then one would expect these critics to argue why analogies can't be drawn between these two types of aims. In other words, why would it be rational for a Laudanite to have utopian non-cognitive aims? The ball is in these putative critics court. It is hoped that the examples involving non-cognitive aims will have Laudan admit what he denies as rational in the case of cognitive aims. I argue that,

I. Laudan's prescription for non demonstrably utopian aims is ambiguous

An ambiguity becomes apparent when the first and second quotes in the last section are compared, while in the first of these quotes a utopian goal was characterized as one that could not be promoted by any actions, in the second quote a utopian aim was characterized as one that is impossible to achieve. A goal such as social justice or the whole truth about some discipline would not be utopian, according to the first characterization, since we rationally believe that we can come nearer them, that we can promote them. But by Laudan's second characterization social justice would be utopian, since given our understanding of human fraility it is strictu sensu unachievable.

This ambiguity about the nature of utopian goals may be the result of a confusion in what Laudan understands by means/ends rationality, in one place he says that a methodological rule is rational if it promotes some desired cognitive end(s), and in the next page he says that a rule is rational if following it is more likely than its alternatives to produce the desired end(s) (cf., Laudan 1987, pp. 24-6.) There seems then to be a confusion between promoting and producing some desired result. It seems that Laudan has conflated two different types of goals as `demonstrably utopian':

i) Valuable goals known to be both impossible to attain and to approach.

ii) Valuable goals known to be impossible to attain, but yet known to be approachable or promotable.

I will concede to Laudan that valuable goals both impossible to attain and to approach (a rather uncommon kind of goal) may be irrational.Foot note 3_14 While I will argue -- contra Laudan -- that valuable goals known to be impossible to attain, but still approachable (I will call this type of goals ideal goals) can be rational. To call ideal goals irrational is like asserting that if it is impossible to fully attain some valuable goal, then we should forsake this goal. This would be akin to a tantrum: `either I know I can fully get what I desire, or I don't care about it.'

On the other hand, Laudan's advice against demonstrably utopian goals may be cogent in a situation in which one has only one possibility: complete failure, without the possibility of partial failures, without intermediate options. In such a hypothetical situation if the valuable goal sought is known to be unreachable, and if this goal is also known to be unapproachable, it might then be rational to resign ourselves and look for another goal. Thus Laudan's advice against complete truth in some scientific field would be cogent, for example, only if false scientific theories couldn't have degrees of falsity or verisimilitude. In the case of many ideals, however, we don't have such a radical situation, even if ideals are unachievable they can still be approximated. There are often intermediate states between not achieving the utopian goals at all, and fully achieving these goals.

Valuable ideals can be rational objectives if we understand means/ends rationality as the attitude of someone that searches for the warranted optimum means for the attainment or approximation of his desired aims. Means/ends rationality then only requires that our means be at least conducive to your aims, it does not require that the rational means actually deliver the aims. Means/ends rationality excludes impossible, but promotable aims as rational only if it is understood narrowly, as Laudan sometimes seems to do, only if means/ends rationality is understood as requiring that if rational we should look for strategies that take us to our goals.

Laudan's lack of discrimination between the previous two types of demonstrably utopian goals turns his injunction against demonstrably utopian aims into an `imprecise' and `vague' recommendation. Hence Laudan's injunction against demonstrably utopian goals is itself `semantically utopian', and therefore Laudan's theory is self-referentially inconsistent.

II. Ideals cannot be dispensed with, because we don't know how far from an ideal is appropriate to aim.

Laudan may argue that while he excludes impossible goals as rational, he is not excluding as rational some achievable goal close to the the unattainable one. He may argue that many admired idealists supposedly striving after an impossible aim were really striving for more modest achievable goals. These idealists were really striving for goals close to, or analogous, to the impossible one. But this let out doesn't work: we try to reach impossible, but promotable valuable goals, because there is no cogent way of specifying in advance how close or how far from the unreachable goal is good enough. So one aims for the ideal itself, even if we are condemned -- as Sisyphus -- to always fall short of it. That is, we don't know how to weaken the ideal without it losing its appeal.

One may, for instance, struggle à la Socrates for self knowldege, even though the process of self discovery may never end. Yet we persist, because, we don't know how to weaken the ideal goal, without it losing its appeal or value. If not how much self knowledge would be good enough?

Other examples, are the search for self coherence, or the quest for a loving attitude. We don't know how much of these goals would be sufficient, we don't know how much would be appropiate, so we aim for the ideals themselves. One aims at the ideal because there is no acceptable weakening of the ideal, therefore it is rational to aim at valuable ideals.

III. Laudan's recommendation against ideal aims is in fact a prescription for intellectual and moral complacency, for mediocrity, and for conservatism.

Laudan's recommendation against impossible but approachable valuable aims (that is against ideal aims) discourages us from aspiring after excellence, cognitive or otherwise. Laudan's recommendation is contrary to a traditional virtue: courage, a virtue necessary to lead a good life. Laudan's advice susbstitutes courage by conformism and stoic resignation. Laudan's ambiguous theory would imply that a Soviet dissident Foot note 3_15who struggled for political freedom in the 50's was irrational, since this dissident knew that his goal was practically impossible to attain. And this was precisely the opinion of Soviet psychiatrists, who considered these dissidents as insane. These dissidents were thought to be insane, because they would not adapt or conform, because they were maladapted, as was shown by their stubborn and hopeless contest, they were maladapted as was shown by the enormous personal costs they were ready to incur for the sake of their impossible dream: `bourgeois freedom'. Still these dissidents persisted in trying to promote, to approximate, the impossible goal. For Laudan a conformist or resigned slave would be rational, but a frustrated idealist who would not conform would not be rational.

Laudan seems to have confused success, expediency, with the struggle to do the right or correct thing. For Laudan, success understood as the attainment of attainable goals is the ultimate goal. Success is Laudan's idol. But success cannot be the ultimate standard, it cannot be the ultimate value, because we can always ask: is the success sought (i. e., the attainment of the attainable goal) right? Is the success sought just? Is the success sought worthwhile? Is the success sought desirable or valuable? For example, if the aim sought is knowledge, we often think of it it as undesirable, if to achieve it, human or animal suffering is required. This is shown by the restrictions on human medical experimentation, and by the ongoing debate on animal experimentation.

Also pyrrhic victories, and unjust victories (in the case of these last as shown by the ongoing debate on just war) are often thought undesirable. And we may even value a defeat, an example is provided by the battle of Kosovo that Serbs -- and their Hungarian and Albanian allies -- lost in 1389. Still this defeat has been hallowed by Serbs for centuries «in several great heroic ballads»Foot note 3_16 possibly because it is believed that some ideal value was sought or defended, say, liberty, or honor. Analogously we sometimes also value failed past theories (failures as judged by Laudan's pragmatic canons of scientific success) because these theories suggested new perspectives or problems, possible examples of such theories are those of Aristarchus and of ancient Atomism.

IV. Idealists aiming for valuable and strictly impossible goals have been praised by legions, and these idealists have been admired precisely because of their idealism. Laudan's disqualification of ideal aims is counter-intuitive, since it contradicts these widespread value intuitions.

Laudan says,

We customarily regard as bizarre, if not pathological, those who earnestly set out to do what we have very strong reasons for believing to be impossible. (Laudan, 1981, p. 51) (Emphasis added)

Perhaps we customarily judge thus, when considering common goals, but one is not governed by customary judgments, when assessing extraordinary cases. Thus, the epithets of `bizarre', `pathological', or `unreasonable', are frequently withheld if the impossible but promotable aim sought is considered to be extremely valuable. In such a case the subject (or generations of subjects) who struggle, or who is thought to struggle, after ideal aims won't be called mad or bizarre, but will instead be considered an idealist, a hero, a martyr, a courageous man, or a saint.

The revered individual has through history often been the tragic idealist who aims at impossible, but promotable goals, even if this idealist has to take arms against a sea of troubles, and even if during his lifetime he cannot prevail. A well known example of idealist conduct is provided by the standard reading of Socrates' conduct after his trial. Socrates chose to stay in Athens even after the death penalty had been pronounced against him. Socrates didn't flee (which he could have done) because he allegedly thought that the correct thing to do, was to be self-coherent, to be true to himself, to be true to his sense of justice, and to obey his city's lawsFoot note 3_17. Now, full personal and intellectual integrity is an impossible aim because of human frailty, and because its full attainment would require of full self knowledge, its full attainment would require of no self-deception, of no inner hypocrisy. Still Socrates had it as an aim, and he was ready to sacrifice his life for this aim. Would we call Socrates irrational by aiming at this end?Foot note 3_18

Laudan may likewise say that all those Christians that have aspired to be like Christ, and have aimed at a perfect Christian life are irrational, qua religious persona. An example of such a Christian would be St. Francis, Laudan may disqualify Francis as irrationalFoot note 3_19 because to strive towards perfection is irrational. It is irrational because we cannot expect human perfection. Still the Church enjoins its adherents to seek Christian perfection, for example it advices its faithful to struggle for the ideal of Christian marriage.

Someone may argue that all the previous examples of `idealists' are wrong, because all the individuals mentioned were not genuine idealists. He could argue that all of these individuals were not really striving after utopian aims, but were rather trying to satisfy their vanity, or were looking for power, or for some other non-utopian goal. But, even if this were the case, these individuals have been admired because they have been believed to have been idealists. In other words the argument here only needs to assume that idealist behaviour has been widely held to be admirableFoot note 3_20. This common esteem for idealist behaviour appears to contradict Laudan's epithet of «irrational» or «pathological» for idealist conduct, and this even if we were to grant that idealist conduct has never been genuinely exemplified by anyone.

A philosophy that disqualifies as irrational widely admired or revered goals, as well as their admirers, is under suspicion of having too exacting standards. Laudan's proscription of ideals as irrational contradicts what we know about common human valuations and behaviour. It contradicts what we know about the behaviour of the admired idealists, as well as what we know about the behaviour of the admirers of these idealists. Then Laudan's advice against utopian aims is itself under suspicion of being `demonstrably utopian', because it contradicts our understanding of some laws of nature, in this case, those laws relating to the behaviour and valuations of a significant segment of humanity. If so Laudan's meta-methodology is under suspicion of being precisely what it condemns, and then Laudan's anti-utopianism is itself suspect of being self referentially contradictory.

If not, consider the following set of three theses:

i) With Laudan sustain that idealist behaviour is irrational, ii) Notice that in our culture `irrational' is a term with derogatory implications of foolishness or madness, and iii) Consider the empirical fact that there have been idealists aiming at valuable goals throughout history,Foot note 3_21 and that many of these idealists have been widely admired qua idealists. This set appears to be incoherent, since from (i) and (ii) one concludes that idealists are foolish, or crazy, and this conclusion clearly clashes with (iii). One could try to escape this incoherence through one of the following options:

a) Conclude that the term `irrational' whatever our de facto social use says is not a term of disapproval or abuse. But to conclude this, one would have to ignore an empirical fact.

b) Assert that idealists searching valuable ideals -- whatever their numerous admirers have said -- are not admirable qua idealists. But it is counterintuitive to say, for example, that Socrates search for intellectual and personal integrity was «bizarre» or «pathological.»

c) Conclude that the search for very valuable, strictly unattainable, but promotable goals is not irrational. Since a world without such utopian goals would be for many an impoverished world, and if such utopian goals were irrational, then full rationality wouldn't be desirable for these many.

Still this argument is somewhat weak. We only know that the set of thesis (i)-(iii) is incoherent, but logic does not tell us which of these thesis to give up. In the following two sections, I will give some further arguments for taking option (c). These arguments taken in isolation are not conclusive, but the sum of all of them may have some weight.

V. The fact that ideals are humanly impossible to attain, and that one can only approach ideals, provides paradoxically a powerful psychological reason for striving after valuable ideals; striving after valuable ideals can also create an enduring emotion of self-respect.

Open ended valuable goals can be more fullfiling, because they permit us to move forward, because there is no end to our endeavour after them. This because the journey can often be more fulfilling than reaching the destination.

The idealist aims for ideals because he wants to keep on improving his acomplishments, because he believes in the perfectibility of life on earth, the ideals help him in avoiding selfcomplacency. The ideals provide aspirational goals, regulative ideas, which guide the idealist's imagination, which guide his hopes and energies, even if he cannot expect to ever fully achieve his ideals. In the case of the search after non-utopian goals one often experiences a letdown, if one achieves them, what else is there? Foot note 3_22It is continued hoping and continued striving that propel a person through life, this psychological fact, supplies one reason for aiming at ideals.

Furthermore a life's struggle after ideals can cause -- at least in certain temperaments -- lasting emotions of self respect or self-esteem, and these emotions are necessary for a good lifeFoot note 3_23. Therefore it may be rational -- at least for these temperaments -- to strive for ideals and their concurrent emotions. Consider, for instance, the case of an idealist such as that of the 5th century Syrian anchorite St Simeon Stylites-the Elder, who lived on top of a tall column for decades (permanently at the mercy of the elements, almost never descending to the ground, and then very briefly) looking for salvation, searching spiritual enlightnement. From Laudan' perspective his fakirsh conduct appears as irrational, but if one takes into account Simeon's situation, i. e., the background beliefs and valuations of St Simeon and his society, one then discovers that his ascetic plan of life was considered praiseworthyFoot note 3_24 and thus it helped provide Simeon with enduring self, and social esteem.Foot note 3_25 And these emotions of esteem could arise only if both Simeon and his contemporaries believed -- perhaps wrongly -- that Simeon was really aiming at some valuable transcendental goals, and not just, for example, at status, fame or prestige.

What his contemporaries probably admired in Simeon was his heroic effort to do what was considered right, that is, they probably admired his heroic effort after the ideal of self coherence. Simeon's contemporaries probably admired his struggle to be true to his own values and principles (values and principles which were also those of most of his Byzantine contemporaries), that is, they probably admired his enkrateia.

The search of ideals can likewise provide whole communities with generalized emotions of self-respect. This fact has been known and exploited, for example, by army leaders who take care to motivate future combatants by arguing to them that the war they are to engage in is a just war, a war that aims at ideals, such as democracy, justice, freedom, honor, glory, etc. An army that believes that it is fighting for ideals is a motivated army, and therefore such a collective belief increases the likelihood of its heroic behaviour. In the case of scientific communities one may speculate that those scientific communities that aim (or believe to aim) at ideals such as truth gain in self respect, and therefore such communities also gain in motivation.

In Laudan's tripartite reticulated model of substantial theories, methodological rules and goals, emotions have been left out, possibly because we ignore so much about the nature of emotions and about their possible rationality. But as the previous example suggests, a complete theory of human action, and in particular of scientific behaviour, would need to take emotions into account. The rationality of aims needs to take into consideration their coherence with other goals (cognitive, moral, practical), and it needs to take into consideration the coherence of aims with substantial theories, as well as their coherence with methods, but it also should take into account the coherence of aims and emotions.

VI. Laudan's prescription against `semantically' and `epistemically' utopian aims is inadequate, because it often happens that one doesn't know, at least consciously, what one is aiming at, and still one can approach obscure goals by the `via negativa'.

One can aim at a goal as a sleepwalker, many have tried to reach fuzzy ideals even if they had to strive half in the dark. For instance, when one longs for somebody, it often happens that one doesn't really know what it is that one desires. It is easy to confuse a longing for love, beauty, knowledge, or companionship with sexual desire.Thus, a personal relationship could start as a result of the search for fullfilment of a supposed erotic desire, just to discover that this desire is only an aspect of what we are really looking for. One discovers that the original longing was for something more than sex. What precisely that more is, it is something we cannot clearly express, it is a je ne sais quoi. It could be a desire to know and to love that person, or it could be a desire for beauty, or for inmortality, transcendence, or for self knowledge.Foot note 3_26

Rimbaud describes such a search in his dreamlike poem «Le Bateau ivre» where he describes the journey of a seer in a tipsy boat, and where the seer is on a search for some unnamed ideal that he seems to only glimpse. Luis Buñuel has also portrayed such a situation in his Cet obscur objet du désir.

Such ends, due to their obscurity, are likely to be semantically and epistemically utopian, that is, these goals cannot be characterized in a `succinct and cogent way', and/or we don't have a `criterion' for determining when we have reached them. Hence, Laudan would disqualify aiming at them as irrational, but one can approach an ideal withouth having a clear idea of what it is, by struggling to eliminate what it is not, by a via negativa à la Popper, a via that is as fallible as any other strategy. Thus one hopes to promote obscure goals such as wisdom or verisimilitude by striving, in the first case, against cases of foolishness, or in the second case, by eliminating error. And one follows the via negativa only because one values, only because one desires the obscure positive ideals.

VII. Laudan does not justify as valuable his pragmatic canons of scientific success, and therefore relativism threatens.

Laudan told us that scientific aims ought to be consistent with the scientific Tradition. And Laudan's pre-philosophical pragmatic canons of scientific success distinguish the success of science -- the scientific Tradition -- from that of other disciplines, also with a tradition, such as for example philosophy or theology.

Laudan's pragmatic canons provide a de facto demarcation criterion between successful science and other cognitive endeavors, and this demarcation criterion has the character of an intuition, since Laudan told us that his pragmatic canons are «prephilosophical» notions:

Scientists' judgments as to the success of a scientific practice depend not on abstract epistemological and methodological matters but on palpably pragmatic ones (...) Thus, a medical practice is successful or not depending to the degree to which it gives its initiates the ability to predict and to alter the course of common diseases. An astronomical practice is successful to the extent that it enables one to anticipate future positions of planetary and celestial bodies.

... If my suggestion that there must be a prephilosophical notion of empirical success -- which is not itself beholden to controverted epistemic or methodological doctrines -- seems controversial, we might ask how it could be otherwise. (Laudan, 1996, pp. 148-9.) (Emphasis added)

This notwithstanding Laudan's rejection of intuitionism,

... we will have no need for our `pre-analytic intuitions' about concrete cases, of for value profiles of the `scientific elite', or for any other form of intuitionism about concrete cases. (...)The naturalistic metamethodologist, as I have described him, needs no pre-analytic intuitions about cases, ..., and no prior assumptions about which disciplines are `scientific' and which are not. (Laudan, 1996, pp. 137-8.) (Emphasis added)

Laudan seems to be saying,

if you are to be rational, and if you want to do successful science, then you should not ignore the pre-philosophical pragmatic canons of empirical success.

There is in this conditional an implicit prescription in favor of the pragmatic canons, since Laudan would not call someone who would ignore his pragmatic canons, while wanting to do science, fully rational, and rational is for Laudan a term of praise (cf., notes # 1 & 6.) The question now arises of how to justify Laudan's conditional norm.

If one rejects, as Laudan has done, justification in terms of intuition, convention or stipulationFoot note 3_27, then we may look for an empirical justification. And this is precisely what Laudan tries to do, he believes that as a matter of fact, or as a matter of historical description, the successful sciences satisfy his pragmatic canons, and that therefore the previous conditional follows. But why aren't theology, philosophy, musicology, scientology, creation science, or even magic, and demonology, taken as examples of bona fide scientific disciplines, as examples of successful sciences? Why aren't the canons of these other activities prescribed to whoever wants to do successful science?

It appears that empirical prediction and control have been taken as canons of scientific success, because allegedly they happen to be the implicit standards of disciplines considered as successful science. Laudan has selected some disciplines as examples of successful science, because they fullfill his preconceptions or intuitions (intuitions which are also ours) of successful science. And then of course, it is a fact that the disciplines so chosen exemplify his pre-philosophical canons of successful science. We are then left with pre-analytic canons which are merely declared as idiosyncratic of successful science. We are then left with canons that are dogmatically asserted as those of scientific success.

The situation is analogous to that of someone who would say: if you want to be just, do as St Francis! And if we ask why do as St Francis? Then we would be answered, because the just, as a matter of fact, behave as St Francis. The question then arises, which standards were used to select the just? and, why weren't Hitler, Prince Dracula, or Francisco Pizarro selected as one of the just?

The answer may be that some individuals were selected as just, because their conduct is consistent with widely held «pre-philosophical» preconceptions or intuitions of justice (though these `pre-philosophical' preconceptions of justice are not shared by all, for example not by Hitler.) And then, of course, it is a fact that the chosen individuals exemplify our pre-philosophical canons of justice. The problem is now to justify as correct the preconceptions or intuitions that helped to select the allegedly just individuals. If this petition of justification is not satisfied, then we could rightly conclude that it has merely been dogmatically asserted that St Francis conduct was justFoot note 3_28.

But then relativism threatens because if Laudan's pragmatic standards have to be taken for granted, if they have the logical character of dogma, then the logical possibility arises of a Babel of different dogmatic canons. The creationist, for example, could reject Laudan's canons and invoke other standards, standards which the creationist could rightly argue are not irrational but only different from Laudan's.

Laudan may argue that to ask for justification all the way down to the `bedrock' is unreasonable, that it is unreasonable because bedrock justifications cannot be provided. Laudan may argue that to aim at such ultimate justification is a `demonstrably utopian' aimFoot note 3_29, and therefore an irrational aim. Still, Laudan himself has told us that what gives comfort to relativism is the lack of justification of methodological rules and standards (cf., footnote 2, above.) And Laudan's pragmatic canons are de facto scientific aims or standards, though of a very general character since they apply to all scientific disciplines. For example, to abide by the canon of scientific predictivility is the same as to set prediction as a goal that must be fulfiled by all scientific theories. This becomes specially clear when one notices that these canons «serve as certifier or de-certifier for new proposals about the aims of science»Foot note 3_30, so these canons are the supreme scientific aims, the aims that judge any other scientific aims. And if we are to accept Laudan's directive on how to beat relativism, we must then try to justify these canons. And since this justification is, and it appears that it will be unavaliable, then one must conclude that relativism -- as characterized by Laudan -- is unbeatableFoot note 3_31. To beat this relativist threat Laudan would require of a criterion of rationality by which to judge his prephilosophical canons. And then Laudan should try at least to explicate -- if not to justify -- this prior criterion of rationality. But both explication and justification are missing.

Notice also that Laudan's pragmatic canons are de facto ahistorical and universal basic scientific aims, because these canons judge the success of any scientific Tradition, these canons judge the success of traditions as disimilar as those of medicine and astronomy. The fixed and universal character of these canons contradicts, however, Laudan's thesis that the aims of science have changed.

The view of science now emerging in some quarters (including my own) is Heraclitean through and through, insisting that science -- diachronically viewed -- changes its content, its methods, and its aims from time to time. (Laudan, 1996, p. 143) (Emphasis added)

VIII. Even if we grant to Laudan, without justification, that his pragmatic canons of scientific success are valuable scientific aims, he also needs to assume without justification that his canons are `primus inter pares' amongst valuable scientific goals.

Laudan prescribes that scientific goals -- amongst these one would expect to find his pragmatic canons of scientific success -- should be jointly consistent. Mutual goal consistency, however, is not a trivial matter, because our aims are not always completely independent, and acting to fulfill some aims may make it difficult or impossible to achieve others. Because of this situation a rational life does not consist of a series of successive actions, each one directed at satisfying one or another of our goals.

And it also follows that full individual human realization is an impossibility, because our different valuable aims have to be somehow negotiated or sacrificed so as to be made complementary, so as to be accomodated in a coherent whole.

There can be, for example, tensions between cognitive aims such as, explanatory power and conceptual simplicity, or between explanation and empirical adequacy, or between clarity and brevity, or between description and explanation, or between accuracy and explanatory scope, or between conceptual simplicity and systemic coherence. Foot note 3_32And there are also incompatibilities between many of these cognitive aims with other type of goals, such as social usefulness, psychological well being, and with moral values. This last case has been exploited by fiction writers with the character of the `mad scientist' or technologist such as Dr. Frankenstein. Examples of every day life contradictory aims, or of aims that are at least partially incompatible, are:

The tensions between social egalitarism and individual freedomFoot note 3_33.

The incompatibilities between preservation of life and quality of life, as illustrated by the axiological debates around abortion and euthanasia.

The inconsistencies between economic growth and standard of life, and a healthy ecosystem.

The inconsistencies between full employement and no-inflation in a market economy.

The tensions between individual freedoms and community values, for example, the case of individual private property vs communal property.

The tensions between freedom of speech, and the preservation of life and physical and moral integrity, as exemplified by the axiological debates about child and sado-masochistic pornography.

Tensions between goals can lead, when unsolved to a Buridan's ass's situation. Hence it is necessary to know how to prioritize, weight or reinterpret aims, so as to combine them in a new consistent synthesisFoot note 3_34. Axiological debates often merely have to do with diverse ways of weighting ends or values, and not with the selection of the set of valuable aims itself. For example, assume two XVIth century astronomers share the same cognitive values, and share the same value hierarchy, except that the first astronomer gives more weight to conceptual simplicity, while the second one gives a higher rank to inter-theoretical coherence. If so, our first astronomer would prefer the Copernican system, because of its conceptual simplicity, while the second scientist would side with the geocentric system, because of its coherence with Aristotelian physics and cosmology.Foot note 3_35 Or another example, a British Laborite, allegedly, gives more weight to social justice than a Tory, though both might share the same list of liberal values.

There are many possible value hierarchies all of them allowed by reason, because to weigh aims we need to order them in terms of relevance, centrality, importance, or pertinence. And these last criteria are themselves values, rather meta-values, meta-values that can be different for diverse communities, scientists, and times. If one tries to justify as valuable some of these meta-values, and if one excludes as Laudan would like to do justification by convention or intuition, it seems one will end with Sextus trilemma: or infinite regress, or an argumentative circle, or dogmatism.Foot note 3_36 And if the regress is to be avoided, and if one is looking for a non-circular justification, then we are only left with dogmatism. Therefore axiological inconsistencies will have to be dealt with different prejudices about what is important or relevant. Then the harmonization of aims is a question to be decided by biographical or historical accident, not by reasonFoot note 3_37. This means that even if different rational communities were to share the same values, they still could have different value hierarchies. And none of these value hierarchies can be shown to be rationally better than any other, except, from their own meta-perspective. One has to choose between hierarchies without the help of reason, because reason cannot determine which hierarchy is to be preferred.

The resulting pluralism of value hierarchies implies that there are many possible rational plans of life,Foot note 3_38 or many possible rational scientific conducts. The awareness of this axiological fact may be an antidote against the danger of fanaticism, a danger to which the search for ideals can lead.

But if a pluralism of value hierarchies is to be innocous, if it is not going to become a relativism where anything goes, it must give priority to some aims, so as to confine the universe of value hierarchies to those acceptable. For example, in the case of contemporary liberal democracies the pluralism of life styles allowed by these societies is far from being full relativism, since contemporary democratic liberalism is restricted by the priority given to values such as human rights, democracy and tolerance.

While if a pluralism of scientific value hierarchies is to be innocous, it would have to be restricted by postulating that some scientific goals should have priority in all acceptable scientific value herarchizations. For Laudan the goals primus inter pares are likely to be his pragmatic canons. Laudan needs scientists qua scientists to value his canons, but Laudan also needs scientists to give his canons priority over other cognitive desiderata. Because if these canons were to be given a low weight, and if one were to emphasize in their instead -- say -- audacious especulation plus theoretical beauty, then one may end doing something closer to contemporary French philosophy than to empirical science.

But how to justify Laudan's priorization of his canons? Laudan has not told us how to weigh, prioritize or re-interpret incompatible but attractive cognitive aims, thus his injunction for aim consistency -- even if inconsistency were clearly establishedFoot note 3_39 -- is incomplete. And Laudan's theory of values is incomplete, because it may not be completed by reason. If so, Laudan's priorization of his canons has to be taken for granted, it has a dogmatic character, in the sense that it can not be rationally justified as correct.


It has been argued that Laudan's theory of values is inadecuate:

a) Because Laudan's theory has problems of self referential inconsistency. Thus, Laudan's theory is `semantically utopian' since it does not distinguish impossible-unapproachable aims from merely impossible aims. And it is suspect of it itself being `demonstrably utopian' when it proscribes idealism as irrational.

b) Because it is rational to aim at valuable ideals since there is no cogent way of specifying in advance how close or how far from the valuable ideal is good enough. So ideals cannot be dispensed with.

c) Because it sacrifices ideals for the sake of expediency, in particular this perspective considers valuable ideals as irrational, and this conflicts with widespread positive intuitive valuations of valuable ideals.

d) And it was argued that to aim for desirable ideals could be shown to be rational, if due consideration is given to the emotion of self-esteem of those that aim at ideals.

e) Laudan's proscription of `semantically utopian' and `epistemically utopian' goals, is too restrictive, because one can pursue an aim obscure to the conscious mind. And still try to approach the goal by a via negativa.

e) Finally it was argued that Laudan's theory could not beat relativism. Because,

i) Laudan does not justifiy as valuable his `pragmatic canons', and these canons have to be accepted without a non-circular justification.

ii) Laudan neither justifies his priorization of his pragmatic canons, a priorization that therefore also has a dogmatic character.


Berlin, I. (1990) The Crooked Timber of Humanity, Edit. H. Hardy, (Princenton University Press) New York.

Elster, Jon (1989) Salomonic Judgements, (Cambridge University Press) England.

Kuhn, S. Th. (1977) The Essential Tension, (The University of Chicago) Chicago, Ill.

Laudan, L. (1981) Science and Values, (University of California Press) Berkeley, Cal.

Laudan, L. (1986) «Some Problems Facing Intuitionistic Meta-Methodologies,» Synthese 67, 115-29.

Laudan, L. (1987) «Progress or rationality? The Prospects for Normative Naturalism,» Am. Phil. Q. 24, # 1, pp. 19-31.

Laudan, L. (1989) «If It Ain't Broke, Don't Fix It» Brit. J. Phil. Sci. 40, 369-375. Great Britain.

Laudan, L. (1990) «Aim-Less Epistemology?» Stud. Hist. Phil. Sci., 21, # 2, 315-322.

Laudan, L. (1990b) «Normative Naturalism» Phil.Sci., 57, pp. 44-59.

Laudan, L. (1996) Beyond Positivism and Relativism. (Westview Press) USA.

Rawls, J. (1972) A Theory of Justice (Oxford University Press) UK, 1973.

Rescher, N. (1985) The Strife of Systems. An essay on the Grounds and Implications of Philosophical Diversity. (University of Pittsburgh) USA.

Worrall, J. (1988) «The Value of a Fixed Methodology,» Brit. J. Phil. Sci. 39, 263-275. Great Britain.

Worrall, J. (1989) «Fix it and be Damned: A Reply to Laudan,» Brit. J. Phil. Sci. 40, 376-388. Great Britain.

Worrall, J. (1996) «Two cheers for `naturalised philosophy of science',» unpublished lecture.

Armando Cíntora

Department of Philosophy

Universidad Autónoma Metropolitana in México City