Vol. VIII, Sept. 1997

Episteme ep.i-steeme

\ epa'i '5te

me \ n.

[Gk. epistemej: knowledge; speciE: intellectual1g certain knowledge

Edi.tor,in~Chief

Darin McGinnis

Editorial Board Matthew Chandler Patrick Murphy Michael Plowgian Leah Row land Michael Shipka Amy Spears Matthew Wilson Veronica \iVilson

Faculty Advisor Barbara Fultner

STATEM.ENT OF PURPOSE: Episteme aims to recognize and encourage excellence in under~ graduate philosophy by providing both stu, dents and faculty with some of the best examples of work currently being done in undergraduate philosophy programs across North America. Episteme is designed to offer undergraduates their first opportunity to pub~ lish philosophical work It is our hope that Episteme will help to stimulate philosophical dialogue and inquiry among students and faculty at colleges and universities.

Episteme is published annually by a staff of undergraduate philosophy majors and millors at Denison University. Please direct all inquir, ies to "The Editors! Episteme," Dept. of Phi­ losophy, Denison University, Granville, Ohio 43023.

Episteme will consider papers written by

• Episteme is published under the auspices of the Denison University De' partment of Phllosophy.

undergraduate students in any area of philoso­ phy. Submissions should not exceed 4,000 words. Papers are evaluated according to the following criteria: quality of research, depth of phLlosophical inquiry, creativity, original in' sight and clarity. Please provide three double spaced copies of each submission and a one page biography including college or university name, major, class year, address and phone number, as well as a 3.5" disk in Microsoft Word format. Deadline for submissions [or Volume IX is February 1. 1998.

EPISTEME

A Journal of Undergraduate Philosophy

Volume VIII

Septemher 1997

Contents THOMISTIC METAPHYSICS AND THE SYNTHETIC

A

PRIORI ...........................

1

Brent D. Timmons; St. Francis Xavier University GENERAL RULES AND THE NORMATIVITY OF CAUSAL INFERENCES IN THE FIRST BOOK OF HUNIE'S TREATISE...............................................................

14

Joshua Derman; Harvard CoUege EXPLANATORY PERVASION AND THE UNITY OF SCIENCE .............................. 25

Andrew Miller; TiVartburg College PERSONS, MATERIALISM AND CONSCIOUSNESS ............................................. 39

Brendan Neufeld; Unil!C1'9ily of S<1IJ/(alcbcwall TilE

VAWE

OF NATURAL KINDS FROM A KRIPKEAN PERSPECTIVE:

A

CRITIQUE Of T!:IHC KATZ'S, "ORGANISM, COMMUNITY AND THE 'SUlISTITUTION PROBLEM'" ........................................................................... 54

Jessica H. Mitchell; Newcomb Colle..ife, Tulane University FREUD ON WRITING: SOME HISTORICIST PERSPECTIVES .............................. 68

Sanja PerovicjMcNlaster University

The editors express sincere appreciation to the Denison University Research Foundation, the Denison Office of Admissions, the Denison Honors Program, Pal Davis and Faculty Advisor Barbara Fultner for their assistance in making lhe publicalion of lhis journal possible.

We also extend special gratitude to tl'lc Philosophy Department faculty: Barbara Fultner, David Goldblatt, Tony Lisska, Jonalhan Maskit, Mark Moller, Ronald E. Santoni and Steven Vogel (or their constant enthusiasm. support and creative input.

THOMISTIC METAPHYSICS AND THE SYNTHETIC

APruoru Brent D. Timmons St. Francis Xavier University The metaphysical system of Thomas Aquinas is a power­ ful instrument, the correct application of which provides the solutions to many ancient and modern problems in philoso­ phy. The difficulty is, however, that the system is just as subtle as it is powerful: a misapplication of its principles in solving one of these problems will inevitably lead to further complications. In this paper I examine such a case.

In an interesting paper, Henry Veatch explores the syn­ thetic a priori as it relates to the philosophy of Thomas Aquinas. In it he correctly argues that there is no easy reconciliation of the positions, yet Aquinas' principia per se nota fulfils Kant's demand for necessary truths about the world. The question 'how are synthetic a priori judgements possible?' becomes, for Thomists, the question 'how do the principia per se nota of the various sciences yield genuine information about the world?'. In answering the latter ques­ tion, I argue that Veatch fails to distinguish the methods proper to the different sciences. As a result of this failure, he turns Thomistic metaphysics into an empirical science, one which lacks the necessity and universality of true Thomistic metaphysics.

I: The Synthetic A Priori and Thomistic Epistemology Kant famously divides judgements into analytic and synthetic. An analytic proposition is one in which the predi­ cate is contained in the concept of the subject. A synthetic proposition is one in which the predicate is added to the concept of the subject. Thus analytic truths are true 'by Timmons is a 1996 graduate afSt. Francis Xavier University, where he graduated with a First Class Honours Degree. He is currently II first year Master's student in philosophy at the Universif:tj ofWaterloo.

EPISTEME - VOLUME

vm-S EPTEMBER 1997

BRENT

D. TIMMONS

definition', and are known to be true once their terms are understood. Synthetic propositions, however, need some足 thing other than the terms involved for verification. In the case of synthetic a posteriori judgements, sense experience verifies whether they are true or false. In the case of synthetic a priori judgements, however, Kant believed some 'third' thing must be added to the concepts to 'guarantee' its truth.

The first principles (the principia per se nota), or founda足 tions, of Thomistic science are self-evident propositions.

These principles are known in themselves (thus are per se nota) and are better known than the conclusions which follow from them. Aquinas states that 1/ a proposition is self-evident because the predicate is included in the essence of the sub足 ject" (Summa I,q.2,a.l). For example, 'Man is an animal' is self-evident because animalis included in the essence of man.

The truth of the proposition is known when one understands that animal is included in the essence of man.

Clearly, in the Kantian classification, the principia per se nota are analytic. This would not be a problem for Thornists, except for the implications of such a classification. As Veatch states, [a]nalytic propositions, it turns out, since they in足 volve no more than a mere analysis of what is already contained in our concepts, are held to be of no real factual import at all" (Veatch 243). Since, it is argued, analytic propositions are true merely because we happen to use language in a certain way they are informative, not of the world or facts, but only of the conventions of our language.

If a proposition claims to be about the world, it cam10t be analytic. If it cannot be analytic it must be synthetic, and thus appeal to something other than the nature of the proposition itself to determine its truth. Thomists claim, however, that the per se nota propositions are necessary and yet a t the same time give information about reality" (Copleston 28). On the Kantian classification, however, this is not possible. If a proposition is about the world or informative then it must be synthetic and appeal to something other than its intrinsic nature to guarantee its truth.

II

It

THOMISTIC METAPHYSICS AND THE SYNTHETIC A PRIORI

If the Kantian scheme is correct, then for Thomists the assertions of metaphysics would be purely verbat nothing more" (Veatch 256). A Thomist, therefore, simply cannot accept Kant's distinction1• The Thomist maintains, contra Kant, that it is possible for a proposition to be necessarily true in virtue of its intrinsic nature and still be genuinely informa­ tive (Le. not be purely verbal). It is not enough, however, merely to say Kant is wrong, for Kant showed how he thought synthetic a priori propositions, on his account, were possible. One must show how such 'analytic informative' propositions are possible. If not, there is no reason to reject Kant, and Thomistic science would be groundless. Thus for Thomists, the question 'how are synthetic a priori judge­ ments possible?' amounts to 'how are the principia per se nota informative about the world?'.

This is exactly what Veatch attempts to answer. He tries to show that the per se nota propositions can be true by the meanings of the terms involved (analytic in Kant's terms), give information about the world (synthetic in Kant's terms) and necessarily true (a priori). He does so by arguing "one can readily see that there is no reason at all why such truths should not be informative or should not be truths about the world" (Veatch 254). At the end of his response, however, he runs into serious problems for the principles of metaphysics, claiming "is there not a sense in which even these principles fall short of strict universality and necessity?" (Veatch 260).

In the following, I argue that Veatch's response is mis­ taken beca use he fails to distinguish the different methods of the sciences. He doesn't recognize that principles can be about the world in different ways. By answering'how can per se nota propositions be about the world?', I show that the correct response avoids the problems Veatch runs into. /I

II: The Division of Thomistic Speculative Sciences

Aquinas stressed the necessity of distinguishing the dif­ ferent methods of the different sciences: For this reason they are in error who try to proceed in the same way in these three 1/

4

BRENT D. TIMMONS

parts of speculative science" (Aquinas Q.6,a.2). The three speculative sciences are natural science, mathematics and metaphysics. Since Veatch focuses on natural science and metaphysics they will be my primary focus. The sciences differ in object and method and so, naturally, they will differ in the nature of their judgments. In order to see how and why they differ, I must examine their objects and methods.

As Aquinas states, lithe speculative sciences are differen­ tiated according to their degree of separation from matter and motion" (Aquinas Q.5,a.l). The objects of natural science cannot exist nor be considered without matter. The natural scientist studies "mutable and material things existing out­ side the soul through natures... that are immobile and consid­ eredwithout particular matter" (Aquinas Q.5,a.3). The meta­ physician studies things which mayor may not exist without matter and can be considered without matter.

Aquinas' account of the nature and method of the sci­ ences requires two distinct operations of the intellect. The first is the "understanding of indivisible by which it knows what a thing is" (Aquinas Q.5,a.3). This operation considers a thing's nature without taking the existence or non-existence of the object into account. It can also abstract truthfully what is not separate in reality, since it does not concern itself with actual existence.

The second operation of the intellect "joins and divides, by forming affirmative and negative statements" (Aquinas Q.5,a.3). This operation deals exclusively with a thing's existence. Since truth is an adequation of mind and thing, this operation must compose and divide according to what is composed and divided in reality. Upon grasping what 'grass' and'green' are, the mind composes the two in 'Grass is green'. Contrastingly, the mind divides man and stone, in 'Man is not a stone'. This operation deals with a thing's existence, judging that it exists in one way or does not exist in another. As Aquinas states, lithe operation by which it composes and divides, it distinguishes one thing from an­ other by understanding that the one does not exist in the other" (Aquinas Q.5,a.3).

THOMISTIC METAPHYSICS AND THE SYNTHETIC A PRIORI

Both of these operations are important for the present issue. The per se nota propositions are judgements made by the second operation of the intellect. As such, they are judgements about the being or existence of the objects in question. The'analytic nature' of the propositions rests on the first operation of the intellect. 'Man is an animal' is a judgement aboutthe way men exist. The judgement is based on the mind grasping the nature of man and animal in the first operation and forming the judgement in the second. This is a preliminary response to the problem of how judgements can be necessarily true by virtue of their terms alone, and also be informative (Le. not merely verbal). It is preliminary, however, because the different sciences' grasp' and 'judge' in different ways, a fact Veatch overlooks. To understand how each science proceeds, we must understand the different methods of the sciences.

Natural science and metaphysics distinguish their ob足 jects in different ways. Natural science distinguishes its objects by simple apprehension or "through the same opera足 tion which is the abstraction of a universal from a particular" (Aquinas Q.5,a.3). It grasps the essence or nature of an object, not by abstracting the form from matter (as in mathematics) butby abstracting the universal from the partkular individu足 als. This is abstractio totius. For example, the mind abstracts , man' from Plato and Socrates,leaving aside their designated matter and accidents.

Metaphysics, however, does not distinguish its objects by simple apprehension, but "through the operation of the intellect joining and dividing which is properly called sepa足 ration; and this belongs to divine science or metaphysics" (Aquinas Q.5,a.3). Since metaphysics is concerned with being as being, negative judgements are its primary method for distinguishing. The judgements of the second operation ofthe intellect deal primarily with the act of exis ting ofthings.

In contrast, the act of simple apprehension deals with a thing's essence. Metaphysics, therefore, will have to be an inherently and uniquely existential science. The objects of metaphysics are those things which may, but need not, exist

5

BRENT D. TIMMONS

6

in matter. By considering such objects in negative judge­ ments, the metaphysician at once denies their dependencyon matter and affirms their act of being. In short, he considers being as being.

Metaphysics and natural science also proceed by differ­ ent methods. Natural science proceeds by an especially rational method. 'This procedure is 'rational' in two ways.

First it remains closest to sensation. Since its objects can neither be considered nor exist without matter, it is inevitably linked to the material world given in sensation. Secondly, it proceeds discursively from one thing to another. In it dem­ onstration takes place through extrinsic causes, [and] some­ thing is proved of one thing through another thing entirely external to it", and therefore" the method of reas on is particu­ larly observed in natural science, and on this account natural science among all others is most ill conformity with the human intellect" (Aquinas Q.6,a.l).

Metaphysics, however, proceeds in a very different man­ nero The meta physician proceeds intellectually. As Aquinas states, " divine science proceeds intellectually not as though it makes no use of reason ... but because its reasoning most closely approaches intellectual considerations and its conclu­ sions are closestto its principles" (Aquinas Q.6,a.1.,r.1). Since the objects of metaphysics are intellectual (i.e. surpass the imagination and sensation), the method must also be intellec­ tuaL It proceeds by apprehending truth simply, more syn­ thetic than the analytic, discursive rational method. In this sense it more closely resembles the intellect of separate substances, and thus it is a struggle for man, who is called a rational animal precisely because the rational method Is proper to him.

Natural science and metaphysics differ in the way in which they distinguish their objects, the method by which they proceed, and they also differ in their end or terminus. 2­ As Aquinas states all our knowledge begins in the senses ... but knowledge does not always termillate in the same way" (Aquinas Q.6,a.2). Natural science deals with objects which cannot be conceived or exist without matter, and the rational method it employs is tied closely to sense data. The tenninus, /I

1/

THOMISTIC METAPHYSICS AND THE SYNTHETIC A PRIORI

therefore, "of knowledge in natural science must be in the senses, with the result that we judge of natural beings as the senses manifest them" (Aquinas Q.6,a.2). Our judgements in natural science must correspond to, and must be modified with new sense data, our sense experience.

The objects and method of metaphysics, however, are intellectual. As Aquinas states: "There are other beings however, that transcend both that which falls under the senses and that which falls under the imagination", and "when we know things of this kind through judgement, our knowledge must terminate neither in the imagination nor in the senses" (Aquinas Q.6,a.2). The judgements of metaphys­ ics terminate in the intellect and are not subject to empirical verification like the judgements of natural science.

In this section I have examined the different methods, ways of distinguishing objects, and termini. These differ­ ences have important consequences for the way in which the principles of the different sciences are'about the world'. In the next section, I will explore some of these consequences, and in the process solve the problems encountered by Veatch.

III: The Principia Per Se Nota as 'Synthetic'

As Veatch argued, and I agreed with above, the question for Thomists of the possibility of synthetic a priori proposi­ tions is really the question of how the principia per se nota can be about the world or genuinely informative. Veatch an­ swers this problem quite convincingly. His solution, how­ ever, is overly simplistic. It is because of this simplicity he encounters problems in dealing with the principles of meta­ physics.

The argument against the informative nature of'analytic' propositions was that, since their truth was determined by the definitions of the terms involved, they were merely truths abou t the way in which we use language. Veatch argues that self-evident propositions are informative because they are talking about the world. For example, 'Man is an animal' is not a judgement about the way we happen to use the word 'man', Rather it is a judgement about those objects which we

BRENT

D.

TIMMONS

happen to call 'men', specifically they are the sorts of things we call 'animals'. The confusion, Veatch argues, rests on a confusion between what is meant and the meaning. Per se nota propositions are judgements about things.

Secondly, the objective nature of per se nota propositions serve to refute the claim that analytic propositions are mere verbalisms. The proposition 'Man is an animal' is self­ evident (analytic), "[b Jut clearly, the self-evidence of this principle does not have to turn on our understanding the meaning of man or animal" (Veatch 258). 'Man is an animal' is self-evident because of the intrinsic nature of the proposi­ tion, independently of our understanding it, or calling a man a man, or an animal an animal. The proposition is only self­ evident because the nature of man includes animal. Thus the proposition must be informative. As Veatch concludes: if 'dependence on meanings and concepts' is under­

stood as dependence on the things so meant and the objects so conceived, then the truth of a self-evident proposition will of course depend on the facts in the case and on the way the world is" (Veatch 259).

Concluding from these two arguments, Veatch asks: Who says then, that the self-evident truths of meta­ physics cannot be about the world and cannot even claim to be factual statements at all? No; the self­ evidence of such principles means just that they are evident through the facts themselves that these prin­ ciples are about and not through any other facts or anything else whatever. (Veatch 260) After thus concluding, Veatch runs into more problelTIs. If the principles of metaphysics are informative in this way, are about the facts, then how can they maintain their necessity and universality? He develops this problem in two ways.

First, it seems as though "these principles [of metaphys­ ics] are able to have factual import and to apply to the real world only if there is such a world or created universe for

THOMISTIC METAPHYSICS AND THE SYNTHETIC A PRIORI

them to apply to" 01ea tch 261). Furthermore, the existence of the world is contingent upon the will of God. These prin足 ciples must positthatthere are such objects as they are talking about, and such a positing is notnecessary nor universaL The example he uses is 'If anything is an accident, then it can only be the accident of a substance'. If such a proposition can have factual import only if there is a world to apply to then it must also add 'There are such accidents in the world'. The second proposition, Veatch argues, is neither necessary nor univer足 saL The second problem Veatch faces is that if the principles of metaphysics require i facts' to apply to, and the facts of our world are contingent, then will it not likewise be contingent and open to question whether these particular self-evident propositions or some others are the ones that apply to the facts of our world?" (Veatch 261). Veatch argues that if this world is contingent then it may be ordered according to different principles then those of Thomistic metaphysics. If so then how is it possible to determine which principles are correct? One can conceive of other possible worlds, and one can conceive that our world is ordered differently than we had thought. Given the conceivable alternatives, "how are we to know which order is the actual one?" for: II

it would seem that it could not be by experience that

we know that the order of our world is one of sub足 stance-accident, of cause-effect, and so on, since expe足 rience can never guarantee the requisite universality and necessity of such principles. (Veatch 262) And this problem returns once again to the beginning, and to how propositions about the world can be necessarily true, or 'How are synthetic a priori judgements possible?'.

Veatch faces these problems because he did not answer the original question correctly. More directly, he failed to distinguish the methods of the different sciences. Drawing from what I said above, I shall now show where Veatch falls into error, and that the true Thomistic position avoids such problems in its solution.

BRENT

D. TIMMONS

Veatch has no problems with the way in which the principles of natural science are informative. In fact, his example, 'Man is an animal' is a principle of natural science.

In natural science, the intellect abstracts the essence or nature of its object, in this case man. The judgement based on this nature, 'Man is an animal' is self-evident to us because we have grasped the nature of man. This judgement is based on the nature, but strictly speaking it is about the world because the terminus of the science is sensation. We know they are true because they are self-evident, but they are only true because the judgements conform to what sensation delivers.

It is clear that natural science and its principles will be informative or 'about the world' because its method and terminus never strays from sensation.

Here Veatch is correct in claiming the principia per se no l'a are informative, even though his explanation is simplistic.

All principia per se nota, however, are not alike. What Veatch does is force metaphysics to conform to the methods of natural science. In short, he makes metaphysics empirical by demanding its principles be 'about tlle world' in the same way the as the principles of natural science. Metaphysics has a different method and terminus than natural science and therefore its principles apply to, the world in a different manner; if one does not recognize this, he will be faced with the same problems as the 'empirical metaphysics' of Veatch.

The question now is, how do the principles of metaphys足 ics apply to the world? The judgements and principles of metaphysics are about the world in a rather peculiar way. To begin with metaphysics distinguishes its objects by separa足 tion. As we have seen this makes the subject inherently existential. In the beginning, therefore, metaphysics has a great deal of factual import. The procedure or method of metaphysics, however, removes it from the realm of sensa足 tion, and presumably form Veatch's factual world. This culminates in the terminus of metaphysics, which is the intellect. It seems as though its judgements are too transcen足 dent to be about the world. On this very subject,however, the eminent Thomist Jacques Maritain remarks:

METAPHYSICS AND THE SYNTHETIC A PRIORI

The judgement wherein knowledge is perfected opens upon the pure intelligible. For it is not because (as in the case of the philosophy of nature) it depends essentially on sensible experience, but rather because of its transcendence, that metaphysics descends (as mathematics does not) to the world of sensible exist足 ence. (Maritain 56) It is because of the transcendence of metaphysical objects and

judgements that they apply to the sensible world.

The subject matter of metaphysics is being as being.

Anything which actually or possibly exists does so because it possesses being. Since the judgements of metaphysics apply to ens inquantum ens, they apply to all beings. Since the judgements are about all being, it applies to the specific type of being which is called sensible being. Furthermore it applies to God as that which is saidofthe effect applies to the cause. This is how the judgements of metaphysics apply to our world.

From this we can see the problems Veatch encountered need not have arisen. Since the judgements apply to all being (ens commune), they necessarily apply to sensible being, even though they are not about sensible being. We need not posit that 'there are such things as accidents in the world' for 'If anything is an accident, then it can only be the accident of a substance' to be universally and necessarily true. For it is a claim about being in itself, the perfect instance of which is God, and God is ultimately necessary. Furthermore the claim is universal insofar as it applies to all possible being and beings. Veatch, therefore, is mistaken when he questions how we know the principles apply to other possible worlds or even if our world is so ordered. The judgements of metaphysics apply to all possible being and beings.

As to the question 'how are synthetic a priori judgements possible?' we have seen that for Thomists this amounts to 'how do the principia per Be nota of the various sciences yield genuine information about the world?'. I have argued that Veatch answered this question incorrectly and in doing so turned metaphysics into an empirical science. I have also given the correct Thomistic response to the question, a re足

11

12

BRENT

D. TIMMONS

sponse which (based on the distinction of the objects and methods of the various sciences) solves the problem while preserving the integrity, necessity and value of metaphysical knowledge.

THOMISTIC METAPHYSICS AND THE SYNTHETIC A PRIORI

WORKS CITED

Copleston, F.e. Aquinas, Penguin Books, 1959.

Maritain, Jacques. The Degrees afKnowledge, Trans. G. Phelan.

Charles Scriber's Sons: 1959.

Thomas Aquinas, St. Summa Theologiae, in Basic Writings, Random House, 1945. ~--

Commentary on the De Trinitate ofBoethius, Questions Five and Six in The Division and Methods ofthe Sciences, 4th Edition. Trans. Armand Maurer. The Pontifical Institute of Mediaeval Studies,1986.

Veatch, Henry B. liSt. Thomas and the Question, 'How are Synthetic Judgements A Priori Possible?'/I , TheModem Schoolman, XLII, March,1965. p.239-263.

NOTBS

1. IIln short, on the issue of the possibility of the synthetic a priori, there just is not any way of avoiding a choice as between Aquinas and Kant. One either has to fish ... with Aquinas, or be content to cut bait with Kant" (Veatch 253).

2. Maurer defines terminus as where the intellect finds in the data grasped by the faculty in question the evidence on which it bases the truth of its judgement".(p.74,n.1) II

GENERAL RULES AND THE NORMATIVITY OF CAUSAL

INFERENCES IN THE FIRST BOOK OF HUME'5 TREATISE

Joshua Derman

Harvard College

IndianaJones, famous scholar and archaeologist, is walk­ ing along a booby-trapped corridor in an ancient temple.

When he happens to step on a triangular blue flagstone, a poisoned dart fires from the ceiling, barely missing him. He continues another few paces, and steps on a second triangu­ lar blue flagstone; another dart fires, missing his head by only inches. This pattern happens five or six times, until Indiana Jones forms the causal belief, "if I step on a triangular blue flagstone, a dart will fire atme." He avoids the next flagstone, grabs the golden idot and return safely to America. Later that week, he happens to be walking down the corridar of the small college where he teaches. As he turns the corner, he realizes he is about to plant his foot on a small, triangular blue flagstone. Immediately, the thought occurs to him, "Dh oh, darts!" After a second thought, he steps down on the flag­ stone, survives, and h,eads off to his class.

Hume's notorious account of causal inferences, in which "all reasonings are nothing but the effect of custom," claims that we have no reason for making the causal inferences that we do. In this theory, Indiana Jones forms the inference "if I step on a triangular blue flagstone, a dart will fire at me" solely by force of custom: after observing the constant con­ junction of flagstone-stepping and dart-firing, custom leads his imagination to associate the impression of the former with the idea of the latter. Once he has developed a sufficient custom, Indiana Jones' imagination automatically jumps to the idea of dart-firing whenever it has the impression of flagstone-stepping.

Derman is asophomore at Harvard College. He is concentrating in philosophy and economics. His interests outside philosophy include physics, history and poetry.

EPISTEME • VOLUME VIII· SEPTEMBER 1997

RULES AND THE NORMATIVITY OF CAUSAL INFERENCES

This example raises a serious problem for Hume' s theory of causal inferences. We do not simply endorse every causal inference that occurs to us. When walking down the college corridor, Indiana Jones is reasoning on something other than the custom he has acquired in the Temple of Doom - he chooses not to endorse the inference" if I step on this flag足 stone, a dart will fire atme./1 Can Hume still claim that all our reasonings are nothing more than the effect of custom? If all causal inferences are unreasonable, why does Indiana Jones not endorse the belief he formed in the Temple of Doom?

In Section XIII of Part III of the Treatise, Hume argues for a naturalistic theory that allows us to make normative judg足 ments about causal inferences. This theory relies on an account of general rules, causal beliefs that generalize over many different experiences of inference-making. While Hume intends for general rules to distinguish between 11 good" and "bad" causal inferences, his argument often seems confusing and contradictory: general rules figure as both rash" and "unphilosophical probabilities", as well as guides by which to distinguish our causal judgments from the activity of the imagination. If we approach general rules from an external! objective stance, in which we judge the goodness of causal inferences by some standard external to the agent, we face great difficulty in resolving this contradiction. It is only if we adopt an internal! subjective stance, in which we judge the goodness of ca usal inferences by the beliefs an agent already has, that we can understand Hume's intended role for gen足 eral rules. In this interpretation, general rules can both. subvert and reinforce the customs which form our beliefs 足 through an equilibrium process in which we seek consis tency among our beliefs/ customs, we can distinguish between those inferences which we ought to hold and those we ought not.

The potential problem with his naturalistic account of causality is not lost on Hume. In Section XIII, he describes the possible implications of his naturalistic account of belief, and the role that custom plays in forming belief: /I

DERMAN

It may, therefore, be concluded, that our judgment

and imagination can never be contrary, and that custom cannot operate on the latter faculty after such a manner, as to render it opposite to the former. This difficulty we can remove after no other manner, than by supposing the influence of general rules. (Hume 149) Though Hume never explicitly provides a definitional dis­ tinction between judgment and the imagination, we might ascribe the following view to him: the imagination is the faculty that forms beliefs, and the judgment is the faculty that gives its approval to them. But even if we are to make the distinction along these normative lines, we have to acknowl­ edge that we cannot judge causality by a different standard than imagination, since they are both "judgments" and the conclusions of the "imagination" are formed by custom.

Hume's answer to this ostensible paradox lies in the roles that general rules" play in causal inferences.

Although Hume is vague about the content of general rules, he indicates several ways in which they are used.

Hume cites prejudice as a prime -example of an "unphilosophical species of probability ... derived from gen­ eral rules, which we rashly form to ourselves" (Hume 146).

Several pages later, he characterizes them as rules form'd on the nature of our understanding, and on our e'xperience of its operations in the judgments we form concerning objects Tl (Hume 149). Hume also provides a functional definition of general rules: in two passages, the first following the previ­ ous quotation and the second in Section XV, he claims that they allow us to distinguish between causes and accidental circumstances.

When Burne says that general rules are "form'd on the nature of our understanding, and our experience of its opera­ tions/' he implies that general rules are a way of categorizing our experience of inference-making. This idea of generality first appears in Section VIII, where Hume accounts for how we form ca usal beliefs upon only a single instance of conjunc­ tion. Although he does not refer to general rules in this /I

/I

GENERAL RULES AND THE NORMATIVITY OF CAUSAL INFERENCES

context, there is a strong similarity between his explanation in Section VIII and his later development of general rules.

Hume acknowledges that we often seem to arrive at causal beliefs without the aid of custom, by only observing a single instance of conjunction of two events. Take the example of a baker who has never made a souffle before, and accidentally leaves it in the oven overnight. In the morning, he returns to find it inedibly burned. Right away, the baker believes his souffle will be ruined if he leaves it overnight again. This poses a challenge for Hume's account of how we form causal beliefs: it is hard to say that custom is responsible for the baker s belief, since this is the first time he has been faced with this particular incident.

Bume replies to this objection by expanding our defini足 tion of custom. While we have had only one experience of souffle-burning and overcooking, "we have many millions [of experiences] to convince us of this principle; that like objects, plac'd in like circumstances, will always produce like effects" (Hume 105). Hume does not mean to claim that we have a rational justification for believing that the future will resemble the past. Rather, in keeping with his theory, he argues we come to form a broad habit of expecting the future to resemble the past. We could picture the imagination, in Hume's theory, as a short-order cook in a diner (the custom足 ers representing impressions, and the entrees ideas). If Bob asks for a ham-and-swiss sandwich twenty times in a row, the cook will immediately go andmake the sandwich the next time she sees him, without waiting for him to ask. After meeting enough students like Bob who routinely ask for the same dish, the cook develops a peculiar manner of dealing with her customers: if Jill only asks once for a hamburger, the next time the cook spots Jill she will immediately run off to make a hamburger. Thus, she has developed through custom the principle that"customers will always want the same dish they've requested in the past," which is analogous to the imagination's principle of "like objects, plac' d in like circum足 stances, will produce like effects." l

DERMAN

Through this example, we can better understand what it might mean to have a general custom. In his account of how we form general rules, Hume relies on"the nature of custom not only to operate with its full force, when objects are presented, that are exactly with those to which we have been accustomed; but also to operate in an inferior degree, when we discover such as are similar" (Hume 147). This suggests that general rules are a kind of broad custom, akin to the one formed by the short-order cook, in which we habitually believe that certain patterns of experience will repeat them足 selves. Prejudice, therefore, is a general rule - custom leads us to generalize our experience under one broad banner, regardless of what particular experiences might indicate. For example, let us say we have observed that Bob, Fred, and Joe (who all happen to be Cretan) are compulsive liars. Instead ofjustformingthe particular rule that Bob, Fred, and Joe will continue to be liars, our imagination is often makes general rule that all Cretans are liars.

So far, general rules have allowed Hume to explain how we could make the inference, "If I leave a souffle in the oven, it will get ruined," even if we have only baked cakes 足 looking back on our experience, we form the general rule that all dough products behave similarly. Still, we have not yet gotten general rules to explain why we consider some infer足 ences good and others bad. To make rna tters more confusing, Hume's normative stance towards general rules often seems contradictory. He deems prejudice a "rash" form of "unphilosophical speculation," yet argues that general rules aid us in separating causes from accidental circumstances. If all causal inferences are unreasonable, one wants to ask, why does Hurne consider prejudice unphilosophical? How are we supposed to Simultaneously employ and steer clear of general rules?

First, let us consider why Hume might think general rules are unphilosophical. Intuitively, we think that the general rrue "all Cretans are liars" is rash when we have met only three liars who happen to be Cretan. On this view, the unphilosophicality of a general rule is a function of how

GENERAL RULES AND THE NORMATIVITY OF CAUSAL INFERENCES

19

much experience we have had. We tend to approach general rules from an external standpoint - we think that the infer­ ence, II all Cretans are liars/' fails to meet an objective stan­ dard of what constitutes causality. Given more experience, we believe, we should eventually feel secure in our inference.

Burne takes a completely different view: our causal infer­ ences, no matter how much experience we have had, are never reasonable. Even if we had met the entire population of Crete except for one person, and discovered them all to be liars, we would still have no reason to expect the last person to be a liar. Therefore, it would be difficult to ascribe an external standard of causality to Hume - we cannot endorse our general rules on the basis of their past empirical evidence.

Before we consider what Burne's normative stance might be, it is important to examine how we use general rules to modify our beliefs. After accounting for the natural mechan­ ics of how we form general rules, Bume tries to explain how we use general rules to regulate our judgment./1 Returning to the Indiana Jones example, we can see the particular rule of cause and effect that he formed by habit in the Temple of Doom: "If I step on a triangular blue flagstone, a dart will fire at me./I When Indiana Jones is about to step on the flagstone in the corridor, he believes for a moment that a dart will fire at him. But, he has also had countless encounters with triangular blue flagstones in the past, not to mention similar flagstones of different shape and color. From repeated expe­ riences of stepping on flagstones with no ill effect, Indiana Jones has formed the general rule that "In general, if I step of a flagstone, nothing bad is going to happen to me."

Why does Indiana Jones end up acting on the general rule

and not the first causal inference? Hume addresses this

conflict between a general rule and another causal principle

in Section XIII:

/I

But as this frequent conjunction Ie.g. of triangular blue flagstones and darts] necessarily makes it have some effect upon the imagination, in spite of the opposite conclusion from general rules, the opposi­ tion of these two principles produces a contrariety in

20

JOSHUA DERMAN

~------------------~

our thoughts, and causes us to ascribe the one infer­ ence to our judgment, and the other to our imagina­ tion. The general rule is attributed to ourjudgment; as being more extensive and constant. The exception to the imaginationi as being more capricious and uncer­ tain. (Hume 149) According to Bume, we endorse the general rule on account of its "extensive and constant nature," as opposed to the " capricious and uncertain" qualities of the other causal infer­ ence. Therefore, we judge according to the general rule and dismiss the other inference as misleading. It is difficult to develop a consistent reading of these claims - as we saw in the Cretan example, from an external! objective standard, the consistency of our experiences does nothing to change the unreasonableness of our inferences. What can Hume mean, then? If Hume is talking about extensiveness, constancy, capriciousness and uncertainty, he must not be referring to how general rules categorize our past experience.

Before we give up on the external/ objective stance on causality, it is worth considering w hat Bume says to suggest it. When referring to the ways in which general rules modify other beliefs, he claims if we take a review of this act of mind" when we make inferences, we can"correct this pro­ pensity (towards particular rules] by a reflection on the nature of those circumstances," and "learn to distinguish the accidental circumstances from the efficacious causes" (Hume 150, 148, 149). Taking a review, correcting, learning, and reflection all imply an active frame of mind, a process of self­ conscious reasoning. It almost suggests that we have reasons for the conclusions we make after reflecting back on past experience through general rules.

Despite this evidence, the external! objective stance seems untenable. A major aspect of Bume's methodology has been to show that what we, as first person observers, take to be reasons, a "scientist of human nature" could account for as simply the work of custom. If we accept his theory of causal inferences, it becomes clear that we can never justify any particular causal inference on an external/ objective ground II

GENERAL RULES AND THE NORMATIVITY OF CAUSAL INFERENCES

- His unreasonable to believe it, butitis our nature to believe it anyway. We might as well replace our usage of "causal belief" with custom."

But perhaps we can take a different stance from the external/ objective, and regard ourselves from the internal/ subjective perspective of minds that already have customs.

Given a set of customs which we hold, some recent and others long-standing, when is a new custom good or bad? From this perspective, we can make better sense of Hume's emphasis on consistency and constancy in making normative judg­ ments. Let us say that we have a set of customs A,B,C... ,Y, and suddenly we develop custom Z. Should we endorse it?

According to this interpretation of Hume' s passage on p.149, Z would be a good custom if it does not conflict with customs A through Y. What happens if custom Y conflicts with custom Z? We already know, on the basis of having a set of customs, that custom Y is consistent with customs A through Z. Therefore, based on its consistency and constancy, we should accept custom Y. Custom Z is capricious and uncer­ tain with respect to our other customs; therefore, we attribute it to the mechanical workings of our imagination.

Hume appears to say something to this effect in his description of the conflict between general rules. General rules can allow us to "compare [a new belief/custom] with the more general and authentic operations of the under­ standing, [and] find it to be of an irregular nature, and destructive of the most established principles of reasoningsi which is the cause of our rejecting it" (Hume 150). By the "most established principles of reasonings" we could inter­ pret Hume to mean our most established customs, since as he has argued before, all causal reasonings are the work of custom.

On one level, this might be considered circular reasoning.

Why should we have any faith in customs A through Z to begin with? This kind of objection forces us to change stances - in order to hold the intemalist/subjective stance, we have to assume that lithe curtain rises" with our mind holding certain customs. This does not necessarily imply that we are born with innate customs, only that we must refer our norma­ II

DERMAN

tive judgments about new customs to our old ones. Another objection might be that judging on the basis of prior customs is not normative. Saying that we ought to do something suggests that there is a good reason for doing it: how can a custom provide a reason? This objection can be best an~ swered by example. Let us say that after having spent a semester studying in England, I developed the custom of driving on the left side of the road. As soon as I return to America, my newly~formed custom draws me towards driv­ ing on the left. Should I endorse this custom? The answer is no, since it conflicts with my much deeper-held custom of driving on the right in America. This is a normative judg­ ment about what I should do, even though it rests on achiev~ ing a consistency of customs.

What happens, one might ask, when a new belief sup~ plants an old one? Why ought we to endorse a new custom over an old established custom? According to the internal! subjective interpretation, we can only replace an old custom if the new one provides greater consistency among the other customs. When the physicist Paul Dirac discovered that one of his equations presupposed the existence of the positron, a sub-atomic particle which had not been experimentally con­ firmed, he chose to believe his equation over the experimen­ tal evidence. Why? For the reason that his custom of trusting mathematical equations was deeper entrenched than his custom of trusting experiments - the positron, if it existed, would provide a much greater consistency among all his other mathematical beliefs than any other accepted theory.

The implication of this view is that even the most unset­ tling experience cannot upset all your beliefs, since you need some deeply held belief to endorse this new experience. One could argue that even Descartes the rationalist was aware of this, when he refused to commence his descent into skepti­ cism without a few provisional principles. Our reading of Hume's paradoxical idea - that a general belief can only be subverted through recourse to another general principle _ can help to explain a particularly cryptic passage from Chap_ ter XIII:

GENERAL RULES AND THE NORMATIVITY OF CAUSAL INFERENCES

Mean while the sceptics may here have the pleasure of observing a new and signal contradiction in our reason, and of seeing all philosophy ready to be subverted by a principle of human nature, and again sav'd by a new direction of the very same principle.

The following of general rules is a very unphilosophical species of probability; and yet 'tis only by following them that we can correct this, and all other unphilosophical probabilities. (Hume 150) We can understand this passage in the following way. From what we can gather from Hume's examples, it seems that most of the causal inferences we make in day-to-day experi­ ence are general rules, not "pure" causal inferences. For instance, Indiana Jones' inference that triangular blue flag­ stones in the Temple of Doom were deadly was also a general rule, since we can be fairly sure that those flagstones were not completely identical. Likewise, his inference that most flag­ stones are safe was a general rule, as was shown earlier.

Indiana Jones arrives at his judgment through an equilibrium process, weighing all his beliefs/ customs in order to attain maximum consistency. Any new general rule will have some inconsistencies with his other beliefs/customs - in this regard, it is an unphilosophical probability, in that by its general nature it is bound to contradict some very refined customs. But it is also philosophical, in that it provides us with opportunity (as Dirac's equation did) to achieve an even greater level of consistency. The push-and-pull of conflicting general rules leads us from our gross customs towards in­ creasingly refined ones.

Thus, by adopting an internal/ subjective standard to­ wards general rules, we can better understand why Hume believes general rules can lead to normative positions. But in what way is it a consequence or boon to skepticism? The above-cited passage shares a similar tone with the final passages of Book I, in which Hume arrives at his "happy-face skepticism." SW'prisingly, skepticism provides a solid foun­ dation for normative judgments about causality - this con­ clusion parallels in opposite direction the one on page 269, where Hume realizes that "in yield[ing] to the current of

23

DERMAN

nature, in submitting to my sense and understanding ... I shewmost perfectly my sceptical disposition and principles."

This question, whether a skeptical project can create a sci足 ence of human nature," rises beyond causality and frames the interpretation of the entire Treatise.

If

WORKS CITED

Rume, David. A Treatise of Human Nature, Ed. L.A. Selby足 Bigge, revised by P.R. Nidditch. Oxford: Oarendon Press, 1978.

EXPLANATORY PERVASION AND THE UNITY OF SCIENCE

Andrew Miller

Wartburg College

The reductionist program arose out of the logical positiv­ ist era, when philosophers of science judged all other sciences against the natural sciences, which were believed to be ideal models of what science should be. Taking this idea one step further, reductionists held that all other sciences were, at least theoretically, unnecessary. The basic premise of reduc­ tionism is that all the laws of science can be explained, and thus rendered superfluous, by physics (Garfinkel 443). Un­ derlying this assertion (often implicitly) is a belief in the symmetry of explanation and prediction; thus, standard reductionism believes that all the laws of science can be derived from the laws of physics.

The reductionist thesis no longer stands as pure and unassailed as it did during the heyday of the logical positiv­ ists. Many authors have uncovered fatal flaws in the reduc­ tionists' views. Some philosophers take their arguments a stepfurther than merely destroying the idea of the unity of laws, though: they attempt to cast doubt upon the unity of science itself, asserting that the specialized scientific disci­ plines are necessarily autonomous from one another. This view is, I believe, too harsh. I intend to argue that while the sciences may not be unified by law, there is every reason to believe that they are unified by explanation.

Explanatory Pervasion Due largely to the notion of what constitutes a law, reductionism can be shown to be untenable. 1 But what of a reduction of sorts in the opposite direction? What if, instead Miller is a junior mathematics and physics major at Wartburg College. In May of 1996, he presented this paper to the Augustana Undergraduate Philosophy Conference and won Honorable Mention for Best Presentation.

EPISTEME - VOLUME

vm-S

EPTEMBER 1997

ANDREW MILLER

of predicting a particular event solely from the laws of physics, I wish merely to explain it? Do such explanations exist?

At least two authors do not think so. Alan Garfinkel and David Owens argue against what David Owens defines as

explanatory pervasion: If the occurrence of an Sl-event explains the occur­ rence of an 5 2-event then there are physical predicates Pl and P 2 such that (a) the occurrence of a P1-event is sufficient for the occurrence of an 51-event, (b) the occurrence of a P2-event is sufficient for the occur­ rence of an 52-event, and (c) PI-events causally ex­ plain P2-events. (Owens 65)

This is not identical to reductionism: for explanatory perva­ sion to be true, P1 and P 2 need not be related by law. All that is required is that we can explain the higher-level events in virtue of the physical mechanisms that connect them.

Garfinkel posits that this is not possible. He illustrates his argument with the use of an example from ecology: a system of fluctuating populations of foxes and rabbits. Given that X(t) denotes the number of foxes at time t and Y( t) denotes the number of rabbits at time t, we can describe the behavior of such a system with a pair of differ­ ential equations2:

'ff = aXY - bX,

dY

dt =pY -qXY

Using these equations we can put forth some basic expla­ nations for phenomena in the system. For example, if the fox population is high, the rabbit popUlation will most likely be decreasing. Thus, if a particularrabbit gets caught and eaten, we can say that [t]he cause of the death of the rabbit was that the fox population was high" (Garfinkel 446). According to Garfinkel, this is the macroexpla11.ation of the rabbit's death, for it appeals to high-level scientific laws. We can attempt a microexplanation, of the event as well; Garfinkel suggests If

EXPLANATORY PERVASION AND THE UNITY OF SCIENCE

L/[r]abbit rwas eaten because he passed through the capture space of fox f' (Garfinkel 447).

He goes on to argue that these explanations are not equivalent, because their objects are not the same: the macro­ object is the death of the rabbit, whereas the micro-object is the deathofthe rabbit at the hands offoxf, atplacep, at time t f etc. Thus, this micro explanation is inadequate in that it is hyperspecuic - it fails to account for small perturbations in the initial conditions. One can suppose that had the rabbit escaped death at the hands of foxf, he probably would have suffered the fate at the hands of some other fox, assuming the fox population remained high (Garfinkel 447). But our microexplanation does not tell us this. Neither, then, does it tell us how the rabbit might have avoided being caught, and, as Garfinkel points out, an explanation is inadequate if it does not lend itself to use for prediction and prevention (Garfinkel 448).

From these assertions, Garfinkel sets forth an alternate conception of explanation. He associates with each event a sort of phase space of initial conditions, where, small changes in the initial conditions do not result in major changes in the qualitative results except at certain critical points.

The crucial thing we want to know is how this set of critical points is embedded in the substratum space, for that will tell us what is really relevant and what is not. Therefore, what is necessary for a true explana­ tion is an account of how the underlying space is partitioned into basins of irrelevant differences, sepa­ rated by ridge lines of critical points. (Garfinkel 452) Is this impression of explanation really all that different from the standard account? To answer this question, we need look only so far as the boundary lines in Garfinkel's explana­ tion space. Now, there are two ways of interpreting these boundary lines: 1) we can explain, in lower terms, why these boundary lines determine the changes in the system, or 2) these critical points are an inexplicable property of the sys­

ANDREW MILLER

tem itself. If the first interpretation is true, though, we really don't have a new definition of explanation at all: Garfinkel's idea collapses into a more standard conception of explana足 tion, in which higher-level laws can be explained, if not derived, using more basic laws. So Garfinkel's use of his phase-space model of explanation to support the disunity of science hinges upon whether these critical points are unac足 countable features of high-level systems. I assert that this is not so, for reasons which I shall now explore.

The untenable nature of Garfinkel's arguments becomes clearer when we consider an idea which led him to the phase足 space model of explanation: redundant causality. Garfinkel feels that in systems such as the fox-rabbit ecology, citing a particular physical mechanism of the resulting event (e.g., citing how the rabbit died) is irrelevant because [s]ystems which exhibit redundant causality ... have, for every consequent Q, a bundle of antecedents (P) such that: 1. If anyone of the Pi is true, so will be Q. 2. If one Pi should not be the case, some other will. (Garfinkel 448) Now, I don'ttake issue with (1) - it is just a reformulation of what David Owens calls the multiple realization point, the idea that each non-physical state can have associated with it many subvenient phYSical states. I must, however, disagree with (2). This aspect states that Q is in some way inevitable - no matter what happens, the universe will contrive to assure that some physical mechanism sufficient for Q will take place. To me, this is excessively speculative; it ascribes to the universe more order (indeed, almost an intelligence) than it could reasonably have. Furthermore, it flies in the face oHact. !tis easily conceivable (though perhaps unlikely) that our rabbit could have escaped all of the foxes and gone on to live a long and happy life; his fate was never a foregone concl usion.

This whole issue can be resolved if one realizes that Garfinkel's redundant causality is nothing more than a thinly

EXPLANATORY PERVASION AND TIlE UNITY OF SCIENCE

29

veiled theory of probabilistic laws. Based on the high fox population, the rabbit's death was highly probable, but not guaranteed. Hence, we could say'at any given time t, the probability that the rabbit will be caught by a fox is high.'

This probabilistic formulation captures all of the important features of Garfinkel's redundant causality: (1) is assured because we know that every time t will have associated with it a certain physical setup, and among those setups will be many that are sufficient (but not necessary) for the death of the rabbit; (2) is covered, and improved, by the fact that it will be highly improbable, but still possible, that the rabbit should survive for a long period of time.

Now that we recognize redundant causality as a probabi足 listic system, we can attempt anappropriate micro explanation ofthefox-rabbitecology. To begin, we calculate theprobabil足 ity of an individual rabbit r's being caught and eaten in a given period of time t (denoted by P(rEt)). The differential equations tell us that in a given unit of time, the number of rabbits that are eaten is -qXY. The probability of a single rabbit's being eaten is that number divided by the total number of rabbits at that time, that is: qXY == qX P(rEt) = y From this, we see that the probability of an individual rabbit's being eaten depends only on the number of foxes in the system!

We can explain this by taking the original luicroexplanation (that the rabbit was eaten because he en足 tered the fox's capture space) as a starting point and asking why this mechanism was likely to occur, i.e., asking why it was nomologically expectable that the rabbit should enter a fox's capture space. It is reasonable to suppose that the size of an individual fox's capture space (relative to rabbits) is a function of both the rabbit's and the fox's attributes (their individual speeds, reaction times, senses of smell, etc.). So we let S(AR' AF) denote the size of this capture space, where AR and AF are the relevant attributes of rabbits and foxes. Then

ANDREW MILLER

the total area covered by the capture spaces of all the faxes at any time t is less than or equal to X S(AR' A F) ("less than or equal to" because of the possibility of separate faxes having overlapping capture spaces). Assurrring that the foxes and rabbits are confined to a region with area R, we can say that the probability that a rabbit will be in the capture space of a fox, and hence the probability of his being eaten, is given by P(rEt) s; X·S (AR' AJ

R which is, as we predicted, a function of the number of foxes in the system.

And so we have a formula, stated in terms of probabili­ ties, spatio-temporal coordinates, and the biological attributes of foxes and rabbits, that is a potential explanation of the behavior at the sociological level. The adequacy of this explanation can be tested by applying the conditions stated earlier by Garfinkel - namely, that an adequate explanation must lend itself to prediction and prevention. Clearly, with this model I am able to predict how likely it is for anyone rabbit to die, and from there to derive the overall results at the sociological level. This model also shows me what I must do if I wish to prevent the death of rabbits (or at least slow down the population decline). I can do one of three things: decrease the number of foxes; decrease the size of the foxes' capture spaces by improving my rabbits or somehow handicapping my foxes; or increase the region inwhich the foxes and rabbits are confined. Of these three, only one is demonstrated in the original differential equations - the other two are "hidden" in the constant q. It appears that I have a successful micro explanation that actually has more descriptive content than the macro-explanation!

So we see that Garfinkel's arguments are insufficient to show the absence of explanatory links between the higher­ level scientific laws and physics. The case is far from closed, however. The systems considered so far by Garfinkel and myself are a good deal less complicated than many systems which arise in the higher-level sciences. Whereas the ex­ amples so far have been ones in which the physical realiza­

EXPLANATORY PERVASION AND THE

UNITY OF SCIENCE

tions of the higher-level events have been largely homoge­ neous, there are many cases in which the physical mecha­ nisms are quite different from one another. In these cases, one would need to formulate a separate exp lanation for each type of mechanism.

David Owens makes this point when he considers (in a more direct fashion than Garfinkel) the thesis of explanatory pervasion in light of an economic example: a monetary exchange. Owens looks at Fisher's Law, which states that if there is an increase in the money supply in an economic system, then there will be an increase in prices (Owens 62).

Now, increases in the money supply and increases in prices can be realized in many ways. The discovery of a new gold mine, the exploitation of a new industry; the lowering of interest rates, and so forth are examples of an increase in money supply. Similarly, raises in prices can be reflected in the real estate market, the agricultural industry, the enter­ tainment industry, and every other market in the economic system. Thus, the PI and Pz in the explanatory pervasion thesis shall have to be broadened to allow us to apply them to heterogeneous sets of physical mechanisms. The multiple physical realizations of these economic predicates shows us that a necessary condition for the truth of explanatory perva­ sion is agglomerativity: ,/ AgglOlnerativity of Causal Explana­ tion: If A causally explains Band C causally explains D, then A&C causally explain B&D" (Neander et al. 460). Further­ more, since there will be no direct explanatory linkage be­ tween the physical realizations of an increase in the money supply and an increase in prices, we will have to instantiate a chain of explanations to connect the two. On the basis of this argument, one can conclude that an additional necessary condition for explanatory pervasion is transitivity: "Transi­ tivity of Causal Explanation: If A causally explains Band B ca usally explains C, then A causally explains C" (Neander et al. 460). Owens proceeds to attempt to show that both agglomerativity and transitivity do not always apply to explanation, casting doubt upon the thesis of explanatory pervasion.

ANDREW MILLER

First, we examine agglomerativity. To attack this thesis, Owens gives the following Aristotelian example: A man eats spicy food, and gets thirsty, and so goes out to the well, where he meets some ruffians who happen to be passing and who kill him. They did not come for the purpose of finding him, nor did they lure him there" (Owens 72). Herewehave a consequent (the man's murder) which is the result of joining two component physical processes (the arrival of the ruffians at the well at time t, and the arrival of the man at the well at time t) each of which have causal explanations: the man's eating spicy food, and the ruffians wishing to rest at the well.

However, it cannot be said that the joining of these explana足 tions causally explains the joint arrival of the man and the ruffians at the well. This joint arrival has no proper explana足 tion; it is a coincidence.

Owens claims that the reason that agglomerativity fails in this case is that the individual explanations have no cornman element, just as the explanation of how lowering interest rates causes inflation has no (or very few) elements in com足 mon with an explanation of how the discovery of a new gold mine accomplishes the same effect (Owens 72). He contrasts these examples with a situation in which agglomerativity does hold because of such a common element: ~hejointarrival of all of the members of an orchestra at an orchestra hall for a rehearsal can be explained by the fact that each member heard the conductor announce the time of the rehearsal (Owens 72).

Karen Neander and Peter Menzies agree with Owens that the agglomerativity of causal explanations does not always hold, but they disagree with his arguments to that effect. To illustrate their point, they redescribe the Aristotelian ex足 ample with different constituent explanations: II

The victim left home at 6 a.m., walked a distance D at an average velocity of V, and consequently arrived at the well at noon. The ruffians decamped at 9 a.m., walked a distance of D /2 at an average velocity of V, and also arrived at the well at noon. (Neander et al.

461)

EXPLANATORY PERVASION AND THE UNITY OF SCIENCE

On the basis of these explanations, the simultaneous arrival of the man and the ruffians ceases to be coincidental; it is predictable from our explanations. But these explanations share no common element, for surely the man's motion is independent of any causal factors that brought about the ruffians l motion. Hence, Owens's arguments are deficient.

As an alternative reason to reject agglomerativity, the authors develop the idea that explanations are often invoked relative to a contrast class3 :

If we schematize the simplest kind of event as an object al s having property F at time tl then a request for a causal explanation of this event may be calling for different things: it may be calling for an explana­ tion of why the object a, rather than some other object, has F at t; or it may be calling for an explanation of why a has property FI rather than some other prop­ ertyl at t; or it may be requesting an explanation for why a has F at time t, rather than at some other time.

(Neander et a1. 462) Relative to this idea, we can illuminate the differences be­ tween these two descriptions of the Aristotelian example.

The most important factor of the event that we are consider­ ing is that the ruffians and the man arrived at the well at the same time. So we seek an explanation which will tell us why they all arrived at the well at that particular time, rather than some other time. In Owensls explanation, the component explanations tell us nothing about why the ruffians and the man arrived at tha t specific time; they merely tell us why they went to the welll as opposed to going somewhere else.

Neander and Menzies's explanations, on the other handl give a precise explanation of the times of arrival, and joining them does explain the two partiesl simultaneous arrival. Accord­ inglYI the authors conclude l as far as we can see, agglomerativity holds provided that the contrast class of the compound explanation matches the contrast classes of the individual explanationsll (Neander et a1. 463). /I

34

- - - - - - - - -ANDREW - - - -MILLER ---_.-_._---­ On this account, we can see that the explanations of the

individual instantiations of Fisher's Law still do not agglom­

erate: just as citing the specific mechanism for the rabbit's

death did not explain why he was caught, citing the mecha­

nisms through which the rise in prices is achieved is insuffi­ cientto show why this inflation took place. We are not asking

why the prices rose at that time, in that place, etc., but rather why they rose at all. If we could formulate some explanation that accounted for the reasons behind the inflation, as I did with the foxes and rabbits, I see no reason to believe that agglomerativity would not hold for those explanations.

To challenge the idea of the transitivity of causal explana­ tions, Owens looks at a well-known nursery rhyme:

For want of a nail the shoe was lost,

For want of a shoe the horse was lost, For want of a horse, the rider was lost, For want of a rider the battle was lost, For want of a battle the kingdom was lost, And all for want of a horseshoe nail. (Owens

75) Owens casts doubt upon the validity of the inference in the last line. Neander and Menzies again agree with Owens that the transitivity of explanation does not always hold, and again they make strong claims against the efficacy of Owens's arguments. Since the details of Owens's position are neither essential nor illuminative of the issue at hand, I will consider only the stance of Neander and Menzies on the issue.

The authors invoke another facet of explanation to sup­ port their assertion: explanatory relevance. They note that "in asking for the causal explanation of some event ... one is asking for an explanatorily relevant causal condition" (Neander et a1. 464). Thus, the question arises: Are we justified in saying that PI-events of the type described in the thesis of explanatory pervasion are explanatorily relevant to P2-events?

Neander and Menzies contend, correctly, in my opinion, that the answer is no. In most cases, the explanatory chain

II •

I

EXPLANATORY PERVASION AND THE UNITY OF SCIENCE

will be too long for this to be true: transitivity will fail in those cases in which the probability of the explanandum event falls below [aJ required threshold when we trace its causal ancestry far enough back" (Neander et al. 465). The nursery rhyme is one such case. The loss of a battle could reasonably bring about the loss of a kingdom; and one could suppose that the loss of a particular rider would be sufficient for the loss of the kingdom as welt if that rider were the king or some other vitally important political figure. But tha t is as far as we may go, say Neander and Menzies, for the loss of a horse does not make the loss of the kingdom sufficiently probable for us to say that it is explanatorily relevant.

Is this true? Surely, one would think that if in reality the battle was lost for lack of one rider; and that rider was lost only bcause his horse collapsed, and that this in tum hap­ pened only because of a loose shoe, then the missing nail would indeed be necessary in an explanation of the event. To resolve this conflict, one must distinguish between explana­ tory relevance and causal relevance. The missing nail is causally relevant to the loss of the kingdom in the same way that "rabbit rentered into the capture space of foxfat time til is causally relevant to the death of the rabbit. But it is not explanatorily relevant, for the same reason that the rabbit's movements were not explanatorily relevant to its death: an explanation based on the missing nail is not useful for predic­ tion and prevention. Had a blacksmith taken more time to ensure that the horse's shoe was firmly affixed, would that have prevented the loss of the kingdom? Common sense tells us no: if the kingdom were so fragile that the death of one rider would determine its fate, we would guess that some other minor factor (a rusty sword, for example) would have caused its downfalL There are other, more important, factors in the kingdom's history (its military strength, recent politi­ cal happenings, economic stability, etc.) that acted to bring about this fragility, and a true, useful, explanation would cite these factors.

It is interesting to note that this explication of the transi­ tivity of explanation is very closely related to the vindication 11

36

ANDREW MILLER

of agglomerativity. Again, we are dealing with contrst classes. The missing nail answers the question, "Why did the kingdom fall at this time, in this way, in thise as opposed to falling tomorrow, or to a different country, etc.?" An expla­ nation of the kingdom's collapse would more likely be seek­ ing hte answer to the question "Why did the kingdom fall at all, as opposed to continuing on for fity prosperous years?"

Thus, a particular physical realization of the kingdom's downfall is not explanatorily relevant.

Now that we see that explanatory trnasitivity is ot true in all cases, we are forced to reject the original concept of explanatory pervasion inasmuch as it is too strong. All is not lost, though, for we may still hold on to agglomerativity as long as we pay adequate attention to the contrast classes of high-level explanations. Thus, we can formulate a weaker version of thethesis. Neander and Menzies call it explanatory

pervasion: If the occurrence of an 51-event explains the occur­ rence of an 52-event then there are physical events PI and P2 such that (a) the ocurrence of the Pt-event is sufficient for the occurrence of the 51-event; (b) the occurrence fo the P2-event is sufficient for the occur­ rence of the S2eventi and (c) there is aseries ofexplana­ tions linking the Pl-event with the P2event. (Neander et al. 466) Conclusion

Now we see that those authors who would reject a thesis of explanatory pervasion are being overly critical of the positivists'views. Indeed, a belief in the disunity of science and the autonomy of the spedal sciences goes against both commonsense and common practice in the scientific commu­ nity. This does not mean that traditional formulations of explanatory pervasion emerge from this debate unscathed.

On the contrary, as I have shown above, in light of the failure of the transitivity of explanation it is necessary to reformulate

EXPLANATORY PERVASION AND THE UNITY OF SCIENCE

the thesis in the manner of Neander and Menzies. This new conception of explanatory unification, as opposed to theory unification, provides a promising common ground for phi足 losophers and scientists alike to share information across disciplines.

WORKS CITED

Boyd, Richard, Philip Gasper, and J.D. Trout, eds. The Philosophy ofScience!..Cambridge: The :MlTPress, 1991.

Canlap, Rudolf. "Logical Foundations ofthe Unity ofSdence," Boyd et al. 393-404.

Fodor, Jerry. "Special Sciences," Boyd et al. 429-41.

Foss, Jeffrey. "Materialism, Reductionism, Replacement, and the Place of Consciousness in Science," The Journal ofPhilosophy 92 (1995): 401-29.

Garfinkel, Alan. "Reductionism," Boyd et al. 443-59.

Neander, Karen and Peter Menzies. "David Owens on Levels of Explanation," Mind 99 (1990): 459-66.

Oppenheim, Paul and Hilary Puhlam. "Dnity of Science as a Working Hypothesis," Boyd et al. 405-27.

Owens, David. "Levels of Explanation," Mind 98 (1989): 59足 79.

Popper, Sir Karl. "Scientific Reduction and the Essential Incompleteness of All Science," The Open Universe: An Argument for Indeterminism, Totowa: Bowman and Littlefield, 1982.

"Further Remarks on Reduction, 1981." The Open

37

ANDREW MILLER

Universe: An Argument for Indeterminism

L

Totowa:

Rowman and Littlefield, 1982.

Trout, J.D. "Reductionism and the Unity of Science: Introductory Essay," Boyd et aL 387-92.

NOTES

1. Jerry Fodor, in "Special Sciences", does an especially good job of arguing against reductionism on appeals to ideas of lawlikeness.

2. In these equations, the product XY represents the number of encounters between foxes and rabbits; a, b PI and q are constants whose values are determined by fitting the observed datain an actual system. These equations are called the Lotka-Volterra equations, and readers who wish to ex足 plore their derivation inmore detail are referred to Garfinkel's sources: Braun, Differential Equations and Their Applications (New York: Springer-~erlag, 1975) and E.c. Pielou, An Intro足 duction to Mathematical Ecology (New York: Wiley-Interscience, 1977).

3. This idea was, I think, implicit in Garfinkel's argu足 ments. I believe his reasoning for why microexplanations often seem to miss their mark lines up perfectly with this idea. l

PERSONS, MATERIALISM AND CONSCIOUSNESS

Brendan Neufeld

University of Saskatchewan

While reductive materialist theories of mind are suscep­ tible to devastating objections, it seems one cannot outright deny some sort of genuine cormection between the mind" and the brain. It is a fact that neurophysiology has enjoyed significant empirical success; testimony to this fact can be seen, for instance, in the pharmaceutical industry, where drugs which affect the chemistry of the brain can be used to alleviate numerous mental disorders. Nevertheless, in this essay I will argue that the materialist thesis is problematic, particularly at the linguistic level. I will then propose a model which salvages some of the significant empirical insights provided by neurophysiology, yet avoids the linguistic con­ fusion of materialism proper by drastically curtailing the neurophysiologist's role in understanding and explaining "the mind". /I

Materialism and Reduction "Whenever a new science achieves its first big successes, its enthusias.tic acolytes always fancy that all questions are now soluble by extension of its methods of solving its questions" (Ryle 76). On this model, the neurophysiologists claim that questions about the mental states and events of humans and other obviously conscious animals can be reduced to ques­ tions about states and events of the brain and nervous sys­ tem. This scientific stance finds its philosophical counterpart in a cluster of views known broadly as "reductive material­ ism" (which I shall simply call "materialism"). Accordingto materialists, what we call lithe mental" is either identical with, or entirely determined by, the physical. Typically, physical" is meant to refer to the brain and nervous system, 1/

Neufeld is a senior philosophy major s t the University ofSaskatchewan. He will be pursuing a master's degree in philosophy in the fall with the long-term goal of obtaining a doctorate.

EPlSTEME • VOLUME VIII· SEPTEMBER

1997

BRENDAN NEUFELD

although many materialists believe that the mental can also be realized by certain non-biological physical systems (for instance, AI theorists hold that a sufficiently complex system of silicon chips implementing a sufficiently complex com­ puter program can be said to have Umental" properties and capabilities). In any case, the materialists "wish to deny the existence of any irred ucible mental phenomena in the world" (Searle 27). Thus, both the neurophysiologists and the mate­ rialists propose an explanatory reduction: the mental can be explained solely in terms of the neurophysiological because the mental is nothing over and above the neurophysiolo gical; every "mental" state or event is reducible to some neuro­ physiological correlate.

I characterize these views as "reductive" whether or not they deny the existence of mental phenomena such as, most significantly, consciousness. The existence of mental phe­ nomena can be denied by such theories (Churchlandian eliminative materialism being the most explicit and extreme example), but it need not be. Materialists can hold that there are such things as consciousness, imagination, beliefs, desires, sensations, etc. However, the materialist position is reductive in that for every mental particular, there is some neurophysiological particular that is identical with, causes, or otherwise wholly determines the nature of its correspond­ ing mental particular. To give a complete neurophysiologi­ cal story of a human being is to give a complete "mental" story of that human being. The neurophysiologist's" aim is to explain what thinking, perceiving, etc. are by reference to the 'thinking' or interpreting', 'inferring' or 'hypothesizing', allegedly engaged in by the brain and its parts (Hacker 149)...." Hence, there is an explanatory reduction (of the mental to the neurophysiological).

I

Linguistic Objections

In Philosophical Investigatiol1s Wittgenstein makes the follOWing point: "[O]nly of a living human being and what resembles (behaves like) a living human being can one say: it L

1:'EFlSONS. MATERIALISM AND CONSCIOUSNESS

has sensations; it sees; is blind; hears; is deaf; is conscious or unconscious (Wittgenstein 97)." This list could be extended to include all of the vocabulary which refers to the mental, including thinking, believing, desiring, remembering, the various moods and emotions, etc.; all these words refer to mental states and events. According to Wittgenstein, these words are predicable only of human beings because we use these words to refer to what human beings do, viz., perceive, think, believe, etc1 : "[T]he criteria for the application of such [words] consist in behaviour patterns in specific contexts against a background of widely ramifying complex capaci­ ties manifest in behaviour (Hacker 147)./1 We attribute the various mental (viz., psychological) properties to persons (to adopt Straws on's terminology) because only a person could manifest the contextually situated behaviour which consti­ tutes the criteria for attributing said properties.

The materialist claims that the psychological properties of a person are entirely dependent upon, and thus can be explained solely in reference to, that person's neurophysi­ ological properties; the neurophysiological explanation is said to be "basic". But this entails that the various psycho­ logical predicates must be ascribed to physiological mecha­ nisms, since they and nothing else are invoked in the expla­ nation of the psychological property. The materialist is thus bound to speak of brains thinking, imagining! being in pain! seeing (in conjunction with the eyes), etc. At this point! the materialist's explanation has become nonsensical. It makes no sense to think of a brain exhibiting the criteria by which we apply psychological predicates. A brain cannot cry out in pain! express a belief, argue for an hypothesis, watch a sunset, or read a book. This is not due to the obvious fact that abrainhas neither a vocalnora visual apparatus; to the extent that the brain is cited as the causal origin of all these activities, it can be said to be what is acting, and hence is the bearer of the psychological predicates. Yet it is these criteria alone which enable us to say of a personthat she feels pain, believes, thinks, sees, or understands written language. Since only persons themselves can exhibit the criteria for ascribing

BRENDAN NEUFELD

psychological predicates, they must be presupposed in any story of the psychological. The neurophysiological explana­ tion, which talks on one hand about brains and on the other about psychological attributes which these brains presum­ ably manifest in various biochemical states and events, is not merely incomplete; it is incoherent.

I have used scarce quotes when speaking about "the mind" because there is a tendency to suppose that in denying that one can attribute psychological predicates to neuro­ physiological mechanisms, one is thereby committed to at­ tributing them to "the mind". hnmediately one is con­ fronted with the ontologically dubious entity made famous

by Descartes; indeed, it is the very absurdity of the notion of a thinking, immaterial substance somehow inhering in and animating a body which provides much of the impetus for the more scientifically acceptable materialist explanation.

Denying materialism does not mean one must speak in terms of "minds" in any more than a metaphorical sense; one can simply talk about human beings or persons, entities which cannot present any serious ontological uncertainty. Once persons have been posited, one can simply proceed to at­ tribute them psychological predicates of perceiving, believ­ ing, etc., based simply on the fact that these are things that persons (not their brains or "minds") do.

Brains Matter Nevertheless, the materialist will insist that it is undeni­ able that persons require brains. This, of course, is true.

Remove the brain from a person's head, and you no longer have a living person; you are left with a corpse. But the linguistic objection outlined above denies none of this; it is a grammatical objection not an empirical one. In his exegesis of Wittgenstein's Investigations, Hacker explains what this means: [W]e know what [psychological] verbs mean only in so far as we have mastered their existing use, which

43

PERSONS, MATERIALISM AND CONSCIOUSNESS

does not license applying them to the body or its parts, save derivatively .... [A] lthough neurological complexity (crudely speaking) is empirically requi­ site for possession of perceptuat volitional, and cog­ nitive faculties, the kinds of features ... that underlie, and constitute criteria for such faculties and their exercise to a [person] are quite different from this.

(Hacker 148,162) These criteria are what persons, not brains, do and say. But just as it does not follow from accepting this that one must deny that there is a mental process" ,it also does not follow that one must posit an unbridgeable gulf between con­ sciousness and brain-process" (Wittgenstein 1022,124).

The neuroscientist proposes an explanation which com­ bines the terms, and hence the rules of use, from the person­ story" and the "brain-story". This combination "produces a conflict of rules and hence incoherence in the neuroscientists' use of these terms" (Hacker 148-9). Hacker emphasizes that the neuroscientist erroneously uses these terms, not that there is anything wrong with these terms per se. Accordingly he admits that the use of neurophysiological terminology in explanations of psychological predicates could be a coherent possibility, though not on the present model of such explanc'l­ tions: 1/

1/

U

If neurophysiologists ... or philosophers wish to change existing grammar, to introduce new ways of speaking, they may do SOi but their new stipulations must be explained and conditions of application laid down. (Hacker 148)

I have no intention of introducing any "new ways of speak­ ing", but I would like now to introduce and develop a grammatical distinction which, I shall argue, picks up on the ontological connection between the psychological and the neurophysiological, a connection which all non-Cartesians (Wittgenstein and Hacker included) seem willing to admit exists at some level. The details of this grammatical distinc­

BRENDAN NEUFELD

tion constitute the stipulations by which this ontological connection can be explained. Although this might presently sound like an attempt to vindicate the materialist position, I am quite sure that those philosophers and scientists who enthusiastically endorse neurophysiological explanation will be less than satisfied with my conclusions.

Transitive and Intransitive Consciousness

Consciousness, according to Searle, lIis the central men足 tal notion" (Searle 84). Furthermore, he argues that the subjective, qualitative character of consciousness cannot, even in principle, be accounted for by a purely objective neurophysiological explanation. Searle is certainly correct in emphasizing the importance of consciousness; any account which cannot explain consciousness is grossly incomplete. I shall assume for now that materialists do not wish to deny the existence of consciousness, but believe (if falsely) that an ideal neurophysiological explanation will be able to account for consciousness. 3 There are two relevant senses of the word conscious足 ness". Norman Malcolm provides a general picture of this distinction: /I

There is a grammatical difference between two uses of the word'conscious'. In one use this word requires an object: one is said to be conscious of something, or to be conscious that so-and-so.... There is another use of the word conscious' in which it does not take an object. If we think that a person who was knocked unconscious has regained consciousness, we can say, 'He is conscious', without needing to add an 'of' or a 'that'. (Armstrong & Malcolm 3) I

Consciousness with an object is called" transitive conscious足 ness", while the more generic sense of consciousness, in which one either is or is not conscious or awake", regardless of what one is conscious of or that, is called "intransitive consciousness" .4 II

PERSONSr MATERIALISM AND CONSCIOUSNESS

Although the general idea should be obvious, there is clearly more to be said about this distinction. It is particularly helpful to note that there are numerous sorts of transitive consciousness. I can be transitively conscious of, for instance, stones, cats, the weather, and countless other things in the physical world, via the senses; I can see a stone or smell a cat, and by doing so I am conscious of the stone or the cat (or perhaps the cat' s smell; the distinction is irrelevant here). I can be conscious ofmyself or bits of myself; when I am in pain because I have burnt my finger, I am conscious of my finger in a particularly unpleasant (viz., painful) way.

Further types of transitive consciousness can be under~ stood by using different prepositions. In these cases, the object of the preposition will be identical with the object of consciousness. Thus, when thirsty (which perhaps involves, among other things, consciousness of one's dry mouth), one can have a desire for a glass of water. One can be attentive to what a pontificating orator is saying; and one might be angry at, or perhaps even in love with, the orator. In all these cases, one is in one way or another conscious of something or someone, i.e., transitively conscious.

Malcolm suggests that whatever we can be conscious of, we can also be conscious that; for instance, I can be conscious afsomeone playing the piano, and also conscious that some~ one is playing the piano. These may seem like the same thing, but there is "a difference between a concept.free mental state (e.g., an experience) and a concept-charged mental state (e.g., a belief)" (Dretske 263). In the above example, then, con足 sciousness of someone playing the piano is, say, seeing a person sitting at the piano and hearing certain sounds ema足 nating from the instrument. Consciousness that someone is a playing the piano, on the other hand, is a belief which necessarily involves certain concepts, viz., concepts of pi足 anos, music, and persons. An animal which lacks these concepts cannot be conceptually conscious that a piano is being played, but it can be conscious of the piano being played (I.e., it can hear it).

45

BRENDAN NEUFELD

Dretske characterizes this distinction as one between consciousness of facts and consciousness of things, but I think that the more general and relevant distinction is be足 tween conceptual and non-conceptual consciousness, and that the word that best captures the nature of the more conceptual forms of consciousness. Thus one can be said to believe (veridically or not) that Santa Claus exists; remember that one has to take the garbage out, or understand that chat" means"cat" in French. Whatever the status of the distinction between consciousness of and consciousness that, it is clear that both are types of transitive consciousness, viz., consciousness with an object.

Intransitive consciousness seems to be a far less complex notion than transitive consciousness; it is consciousness" tou t court" ,not consciousness ofor that anything (Armstrong & Malcolm 3). Intransitive consciousness is consciousness without an object. I think it is best thought of as a necessary "background" for transitive consciousness.s That is, one mustbe intransitively conscious, inthe sense of being "awake", in order to be transitively conscious. Intransitive conscious足 ness can be better understood by examining a misconstrual of what it is. Armstrong says, "Suppose it is true to say of somebody that he is seeing a horse. Normally at least, this is an intransitive idiom" (Armstrong 117). Not at alL One sees a horse; here we clearly have a type of consciousness (seeing) which has an object (a horse), and hence is a form of transitive consciousness. I am uncertain why Armstrong would say this, given that he later claims (correctly) that there is no intransitive perception"; presumably "He saw a horse" as an idiom suggests a type of intransitive conscious足 ness, but how it does so is entirely unclear (Armstrong 117).

Transitive consciousness presupposes intransitive con足 sciousness, in the sense that to be able to say of a person that she perceives her surroundings, has sensations, expresses beliefs, etc., presupposes her "being awake".6 Intransitive consciousness is not, however, a mere static state; there are degrees of intransitive consciousness, as Searle illustrates: /I

11

PERSONS, MATERIALISM AND CONSCIOUSNESS

47

-------------~~~~~--------------------------~~

If during sleep I have dreams, I become conscious, though dream forms of consciousness in general are of a much lower level of intensity and vividness than ordinary waking consciousness. Consciousness can vary in degree even during our waking hours, as for example when we move from being wide awake and alert to sleepy or drowsy, or simply bored and inat足 tentive. Some people introduce chemical substances into their brains for the purpose of producing altered states of consciousness, but even without chemical assistance, it is possible in ordinary life to distinguish different degrees and forms of consciousness. (Searle 83)

The types of consciousness Searle is describing in the above passage can all be understood as intransitive;7 they do not take objects. One cannot be awake of or that anything, no matter how alert, attentive, drowsy, or distracted one may be in that particular state of "awakeness". By affecting the brain with alcohol, one cannot be said to be drunk of or that an object, fact, etc. Nevertheless, intransitive consciousness (of whatever sort) is a necessary condition of transitive con足 sciousness; it is the requisite "background" for the more seeing, hearing, feeling emotions, thinking, and all the other interesting things that persons do.

Consciousness and Materialism The empirical findings so lauded by materialism seems to provide an asymmetrical understanding of these two types of consciousness, with intransitive consciousness enjoying the more thorough explanation. According to the materialist, the fact that everything can be explained physically means that everything can be explained causally, viz., in terms of causal physicallaws. Our sense ofintransitiveconsciousness as a background state fits the causal model quite well. To be /I awake" in the most basic sense, one requires a properly functioning, oxygenated brain in a comparatively normal biochemical state. Modifications of this biochemical state

BRENDAN NEUFELD

produce different types of intransitive consciousness. For instance, a brain which is being affected by the chemicals found in tranquilizers will produce a lower level" of intran­ sitive consciousness (i.e., will cause a person to be drowsy); on the other hand, a brain being flooded by endorphins causes an intransitive state of consciousness characterized by an overall sense of elation (though one is not, in this sense, elated of or that anything). Different general IImoods" as different types of intransitive consciousness such as being ecstatic or depressed (though, to distinguish moods from emotions (which are types of transitive consciousness), not ecstatic or depressed of or that anything), seem quite ame­ nable to causal explanation: II

t

Moods are pervasive, they are rather simple, espe­ cially because they have no essential intentionality, and it looks like there ought even to be a biochemical account of some moods. We already have drugs that are used to alleviate clinical depression. (Searle 140-1) A neurophysiological (Le., physical) explanation of the vari­ ous types of intransitive consciousness seems to be a genuine possibility, because intransitive consciousness seems to be a largely (I will not say entirely) causal notion; one is caused to be alert, drunk, or depressed because one's brain is well­ rested, permeated by alcohol, or in some state of chemical imbalance. Indeed, neurophysiologists have been able to explain a great deal in this domain.

Transitive consciousness, on the other hand, has not been nearly as well explained by neurophysiology. 8 Explanations of remembering, believing, and other such cognitive types of consciousness seem particularly impoverished; in particular, explanations of how the content of particular memories and thoughts is stored" in particular parts of the brain is highly theoretical at best. Explanations of perception fare a bit better, but still seem insufficient to account for such things such as the Gestalt structure of perception in purely neuro­ physiological terms. Neuroscientists readily acknowledge 1/

PERSONS, MATERIALISM AND CONSCIOUSNESS

the high level of theoriticity in many of their explanations, and, for some types of consciousness, the lack of any reason­ ably complete or concrete explanation. However, they tend to dismiss these problems by invoking a future when, as a result of continued successful empirical research, all such difficulties will allegedly disappear.

Transitive consciousness is characterized by its having an object. It is consciousness of or that something, which is to say that it is consciousness directed towards an object. Hence, transitive consciousness involves an intentional relation to its object. 9 These intentional relations cannot be reduced to causal relations, simply because they are two fundamentally different sorts of relations; the essence of intentionality is its directedness" or aboutness", while this element is not present in causality. Therefore, transitive consciousness is not explainable on the neurophysiologist's model of an en­ tirely causal, physical system. Any appeal to future empirical discoveries about the brain is futile, because these will only be discoveries of causal events and relations in the brain; intentionality, however, is not reducible to causality.

Ii

/I

Consciousness and Language

EYen if empirical discoveries allowed the neuroscientist to establish some definite level of psycho-physical parallel­ ism in cases of transitive consciousness}O it does not follow that one can give a reductive explanation of psychological predicates. The linguistic points made above still hold: Persons are the bearers of psychological predicates, and it is what persons do that constitutes the criteria for ascribing these predicates; hence, an explanation of the psychological must be based in language about persons and what they do, no matter how much is known about brains and what they do. It is clearly the case that persons bear the psychological predicates which correspond to the various types of transi­ tive consciousness. Persons see, feel and understand; brains do not. Furthermore, it is the actions of persons that serve as criteria for attributing the various types of transitive con­

49

BRENDAN NEUFELD

sdousness. We say that someone believes that such-and-such is the case because tha t person verbally affirms the belief, acts in accordance with it; etc.; to explain that someone believes something because his brain carries some sort of chemical code presupposes that we have used the regular criteria of what he says and does to determine that he does, in fact, believe such-and-such.

Intransi tive consciousness might seem to be exempt from this point if, as I have suggested, it can largely be understood in terms of a causal, neurophysiological explanation. How­ ever, it is still persons, not their brains, who are said to be awake, asleep, elated, depressed, inebriated, etc.; persons are the bearers ofthe psychological predicates which correspond to. the various types of intransitive consciousness. Even thoughintransitive consciousness can largely be explained in neurophysiological/biochemical terms, such an explanation is secondary to an explanation in terms of the persons to whom the predicates of intransitive consciousness are as­ cribed. Empirical knowledge of a certain aspect of persons, viz., the influence of their brains on their intransitively conscious states, cannot absolve an explanation of this aspect from presupposing persons. As brains are no more than parts of persons, so too is neurophysiological explanation no more than a part of the explanation of persons.

Conclusion

The fact that persons are more than their neurophysi­ ological makeup does not imply that neurophysiology is irrelevant to an explanation of persons and what they do and say. As we have seen, neurophysiology is particularly rel­ evant in understanding and explaining intransitive con­ sciousness, while it has a far lesser role in explaining transi­ tive consciousness. However, a purely neurophysiological explanation is not sufficient to explain any sort of conscious­ ness, precisely because it leaves out of the explanations those very things which are conscious, persons.u Hacker warns that crossing the language of persons with the language of

MATERIALISM AND CONSCIOUSNESS

neurophysiology "produces a conflict of rules and hence incoherence", but only if one attempts to reduce the "person story" to the "brain story" (Hacker 148-9). If the "brainstory" is viewed as no more than a supplement to the prior, basic Ifperson story" and if stipulations of applicability are care足 fully laid out, then the crossing of languages need not result in incoherence; rather, it might lead to a more inclusive and unitary account of persons - for whatever else they are, persons are beings which possess brains, and these brains are necessary for their personhood. What I have attempted to do here is provide a basic outline for how such a project might proceed. f

WORKS CITED

Armstrong, D.M. and Norman Malcolm. Consciousness and Causality: A Debate on the Nature of Mind. Oxford: Blackwell,1984.

Dretske, Fred. Conscious Experience." Mind 102 (1993): 263足 283.

It

Hacker, P.M.S. "Men, Minds and Machines./I Wittgenstein, Meaning and Mind. Oxford: Blackwell, 1990. 147-170.

Ryle, Gilbert. The Concept ofMind. London: Hutchinson,1949.

Searle, John R. The Rediscovery ofthe Mind. Cambridge: MIT, 1992.

Wittgenstein, Ludwig. Philosophical Investigations. Trans.

G.E.M. Anscombe. 3rd ed. New York: Macmillan, 1953.

52

BRENDAN NEUFELD ---------------------------------------­

Notes 1. Animals can, of course, perceive things and have sensations, but it is dubious whether or not they can engage in more complex cognitive activities.

2. This is (admittedly) a quick rejoinder to anyone who would assume that a Wittgensteinian approach dispenses with the subjective aspect of personhood in favour of a behaviorist or verificationist stance.

3. Eliminative materialists willfully deny its existence; I can only dismiss this as absurd a philosophical position as one could ever take.

4. Malcolm and others go on to use this distinction in debates about the nature of introspection and self-conscious­ ness. While it is true that these issues follow from the distinc­ tion, my project here is to show how these concepts of consciousness relate specifically to the materialist thesis.

S. This is notthe same as Searle's notion of Background" (Searle 175ff.).

6. The literature seems equally divided on whether intransitive consciousness presupposes transitive conscious­ ness; I am inclined to say that it does not, although it is almost always the case that manifestations of transitive conscious­ ness constitute the criteria by which we attribute intransitive consciousness.

7. Dreaming constitutes an exception; we dream of things, hence it is clearly a type of transitive consciousness.

Nevertheless, I believe dreaming can be accommodated on this model; as a sketch, I suggest that one could posit a very low level of intransitive consciousness (" awakeness which would facilitate the transitive consciousness of dreaming (which itself seems to be a peculiar species of imagination).

8. I do not claim to be abreast of state-of-the-art neuro­ physiology, butthis is the sense I get of the discipline's status.

9. I do not wish to identify transitive consciousness and intentionality; there seem to be many cases of intentionality without transitive consciousness. For example, a belief is intentional even if one is not presently conscious of that belief.

10. The discovery that certain areas of the brain are "modules" for certain types of transitive consciousness (es­ pecially percpetion) is an example of a small step towards such a arallelism.

II

lt

)

MATERIALISM AND CONSCIOUSNESS

11. To say nothing of the Nagelian objection that a reductive materialist account could not possibly account for the subjective, qualitative nature of consciousness; such sub足 jectivity seems to be an essential aspect of consciousness (transitive or intransitive).

V ALUE OF NATURAL KINDS FROM A KRIPKEAN PERSPECTIVE: A CRITIQUE OF ERIC KATz'S, "ORGANISM, COMMUNITY, AND THE 'SUBSTITUTION PROBLEM'"

THE

Jessica H. Mitchell

Newcomb College, Tulane University

Saul Kripke' s influential theory of reference examines the relationship between the name given to a referent and various descriptive statements that can be made aboutit. In this essay, I extend this theory of reference, as developed by Kripke and Hilary Putnam, in an effort to critique and also develop Eric Katz's argument in, "Organism, Community, and the 'Sub· stitution Problem'."

I begin by explaining how Katz rejects an organism model of species within an ecosystem because it allows for the possibility of 'the substitution problem', and instead favors a community model of associated individuals who are valued both intrinsically and instrumentally. I then maintain that Katz's distinction between the organism and the COIn­ munity models is merely one of degree and is largely seman· tic. The difference between the models is the way in which they serve Katz's goal of differentiating between intrinsic and instrumental value. I analyze this goal within the context of Kripke' s theory of reference in a way that illumina tes some of the issues contributing to the dispute between valuing species intrinsically and instrumentally. Finally, I conclude that Katz's counterfactual exercise and 'the substitution prob­ lem'i if approached from the perspective of the Kripke's theory of reference, can bring to light new considerations for an ethic to address.

Mitchell is a senior philosophy/political science major at Newcomb College, Tulane University. Her interests include social and political theory as well as applied philosophy.

EPISTEME • VOLUME VIII· SEPTEMBER

1997

THE VALUE OF NATURAL KINDS

Organism vs. Community To analyze Eric Katz's holistic environmental ethic, it is important first to understand his conceptual distinction be足 tween the organism and community metaphors often used to describe natural systems. In the next section, we will see tha t there are significant problems with this distinction. The organism model conceives of individual species as organs in a larger organism, organs that cannot exist apart from the organism. The community model conceives of individual species as members of a larger group, but also acknowledges that they have a measure of independence. For Katz, an organism model is not desirable because it fails to value a species for itself, intrinsically, and instead values a species solely in terms of its functional, instrumental worth (Katz 249).1 Katz believes a feasible environmental ethic must establish an acceptable balance between and include both intrinsic and instrumental value. The community model, for Katz, seems to fill this requirement.

Katz argues (incorrectly it would seem) thatthe organism model fails to strike this balance between values, because it imagines a species merely as a component, part, or unit whose existence [in the natural system] is due to the con足 tinuous functioning of the organic whole of which they are a part" (Katz 245). A species is conceived of as an organ necessarily dependent upon all the other organs in the sys足 tem. By overemphasizing species' interdependence, Katz argues, the organism model denies a species' independence, its intrinsic value. The independence is lost, according to Katz because, "an entity valued intrinsically requires no relationship with any other entities" (Katz 249). But all species require relationships with other entities to survive. It seems that within this dependence there does exist an ele足 ment of value which is granted to the species itself, to its presence in the system as a whole; intrinsic value does seem to be present.

II

55

56

H. MITCHELL

Katz argues that another undesirable result of the organ­ model is •the substitu tion problem'. If a species is valued only instrumentally, it is logically possible to replace that species with either another species or a machine capable of iuUilling its functional role in the natural system. Katz explains, "what is really important is the role, not the species" (Katz 251).: If this is true, the organism model allows for a species to be replaced without remorse] because what is most important is the ecosystemic organ, to which the species is whollv subservient.

Eric Katz claims, however, that there is a model that successfully embraces both the intrinsic value of a species as well as the instrumental role it plays in the natural system.

This model is that of the community, which "focuses on both functional value and autonomous intrinsic value of natural entities in a sy-steml) (Katz 241). In this model, an entity exists both as an individual (in its own right) and as a member (constituting a unit) of a functioning community. A species, as an entity in the community model, has a relatively inde­ pendent existence and value over and above its value as a member of the community. According to Katz this relatively independent status makes a species in this model"similar to an entity with intrinsic value; it possesses some value in itself without regard to other entities" (Katz249). Itis clear that this model solves 'the substitution problem' by recognizing that species cannot be conceived solely as a functional unit in an ecosystemic whole, by recognizing their intrinsic worth. In recognizing both instrumental and intrinsic value, this model creates the conw tions necessary for an effective environmen­ tal ethic which recognizes and protects a species both for itself and for its function.

A Degree Relationship: Organism and Community Now that Katz's basic argument has been explicated, we can tum to an examination of its seemingly unfounded distinction followed by a Kripkean critique of this distinc­ tion. A close reading of the article shows that the difference

THE VALUE OF NATURAL KINDS

57

between an organism and a community is nothing more than a difference of degree, one that weights intrinsic and instru­ mental value. Therefore, Katz's criticism of the organism model in favor of a supposedly more balanced version, the communitymodel,is inaccurate in suggesting that the organ­ ism model wholly excludes intrinsic value. Katz acknowl­ edges that his distinction is merely a semantic differentiation of degree in saying, the model of community permits the consideration of both intrinsic and instrumental value to a greater extent than the model of organism. Since anorganism is primarily concerned with the functions of interde­ pendent parts, it emphasizes instrumental value. (Katz 249-50)3 In saying that what is important is the extent to which value is acknowledged in the community model, Katz ex­

presses his displeasure with the lesser degree to which the organism model grants a species value. Katz favors an ethic with a more even balance of-values, both intrinsic and instru­ mental. But the organism model need not lack intrinsic value entirely. In advocating the community model as a balanced ethic, Katz makes it clear that a species necessarily has intrinsic value merely by virtue of being natural: [T]he intrinsic value of natural entities is their source or origin - what causes them to be what they are. A natural entity possesses intrinsic value to some extent because it is natural, an entity that arose through processes that are not artificially human. This 'natu­ ralness' is one of the properties that gives it its value.

(Katz 254)4 Evidently, it is not that the organism model overlooks or

excludes intrinsic value, rather, Katz finds the degree to

which that model acknowledges this value inadequate for

forming a balanced environmental ethic. But because he has

explicitly said that species have intrinsic value merely by

58

JESSICA

H. MITCHELL

virtue of beingnatural, the substitution problem is a mislead­ ing construct. Substituting a species valued both instrumen­ tally and intrinsically, for that is what Katz claims a desirable ethic would accomplish, would seem morally wrongeven for a holist. If a species is valued intrinsically even to a dimin­ ished extent, then it would be hypocritical for the valuers to replace it with another intrinsically valued entity, since sub­ stitution seems to presuppose thatneither the replaced species nor the one replacing it has intrinsic value. Hence the 'substi­ tution problem' amounts to a "straw man", since even most holists would balk at the idea that species are nothing but replaceable organs.

Katz makes clear what aspects of an environmental ethic ~e considers are important, namely, that it incorporate both intrinsic and instrumental value into its framework, thereby eliminating the possibility of the 'substitution problem' be­ ing actualized. In whatfollows, I show how Kripke's theory of reference can provide the basis for a thought experiment leading to such an ethic. In addition, I indicate why an application of Hilary Putnam's work with counterfactual worlds provides additional reasons why 'the substitution problem' is an implausible concern. Finally, I argue that their work can also give the sort of rationale for preservation that Katz believes a feasible ethic could provide.

The Theory of Reference

Kripke's theory of reference as explained in Naming and Necessity is concerned with the extent to which descriptions are related to names in the real world as well as in counterfactual worlds. In reworking reference theory, Kripke criticizes and ultimately overcomes the previous conclusions concerning names and descriptions as suggested by such philosophers as Frege and Russell. Frege for example be­ lieves that names, spoken and written linguistic terms, are synonymous with their ontological definite descriptions/' statements that can be made about a referent's properties.

For Frege, names have the same meaning as descriptions; II

THE

V ALUE

OF NATURAL KINDS

that is, names can'stand in for' /,be substituted for' descrip­ tions. The particular description the speaker intends the name to stand for is used to determine the referent of the name" (Kripke 28).

By contrast, a name for Kripke is a rigid designator; that is, it is nothing more than a strict label or a tag given to a referent, to a unique identity. As Kripke states, "Let's call something a rigid designator if in every possible world it designates the same object ... [N]ames are rigid designators" (Kripke 48). 5 The name, then, is rigidly tagged onto an object across all possible worlds. However, a description, in Kripke' s theory of reference, is not synonymous with a name; that is, a description is not something that'stands for' a name or is equivalent to it. If a name is 'x' in a given sentence, then in Kripke's analysis, a description can not be substituted for' x' and still maintain the same meaning (Kripke 48).6 A descrip­ tion is a contingent statement from which a reference is fixed or determined. Hence, lithe description used is not synony­ mous with the name it introduces but rather fixes its refer­ ence" (Kripke 96n).

In the new theory of reference, this description is most directly aimed at proper names. In cases involving individu­ als, the theory is easily applied, seeming almost commonsensical. Kripke gives many such examples, one of which concerns Aristotle. If the name i Aristotle' is the name affixed to a specific object, then the statement 'Plato' s greatest student' is, as a description, not a rigid designator but instead is a statement about the contingent properties associated with that name. It is contingent because if Aristotle were not Plato's greatest student he would still have been called ,Aristotle'. The name is rigid, but the description in this case is not.

Of special interest here is tl1e case in which a statement is a definite description which can be used to isolate Aristotle' as a unique individual. Such a case would be a statement in which Aristotle were to state that he was born in'x' month to 'such and such parents'. The descriptions in this case point only to the being named 'Aristotle' asa unique individual. In II

I

H. MITCHELL this caset the description indicates fundamental qualities/ essential properties that could only be held by one unique individuaL When essential properties are used in a descrip足 tive statement in this waYt that descriptiont tOOt is a rigid designator. Then and only thent just as a name wouldt the description tags the individual with its essential properties (Kripke 28). These essential properties are exclusively neces~ sary characteristics or scientific facts about the individual that are known definitely.

Common Nouns and General N ames As They Are Associ足 ated With Natural Kinds Another interesting relationship between name and de足 scription that is especially significant in this analysis is how Kripkets theory of names applies to common nouns and general names. By applying the theory to a collection of individuals t such as a species t the theory of names becomes significant for environmental ethics. In these casest the name used represents a group that has some commonalityt unlike the previous cases where the names belonged to unique individuals. Within these cases His the class of general names associated with natural kinds - that iSt with classes of things that we regard as of explanatory importance; classes whose normal distinguishing characteristics are 'held together' or even explained by deep-Iyingmechanismll (Putnam 102). It becomes clear that when general names are usedt a wrinkle is added to Kripkets theory of reference. Do the individual members of the group lose some form of individuality by being represented by a collective rigid designator? Whatis the shared essence of the members in the group such that they can all be tagged by the same nameT What implications does this have for groups such as species and how does this affect an environmental ethic? Hilary Putnam addresses many of these unanswerable questions inhis development of Kripke s theory of reference as it applies to groups. l

l

THE

V AWE OF NATURAL KINDS

Putnam's investigation leads to a focus on essences of groups with something in common. As he explains, "what the essential nature is is not a matter of language analysis but of scientific theory construction.. . (Putnam 105). Such scientific constructions would resemble chemical breakdowns or chromosomal structures of objects. Kripke suggests that, II A priori, all we can say is that it is an empirical matter whether the characteristics originally associated with the kind apply to its members universally, or even ever, and whether they are in fact jointly sufficient for membership in the kind" (Kripke 137). As Putnam points out, there is one obvious problem with such essential descriptions. Because they represent a stereotype of the object, they do not account for abnormalities. Despite this drawback, these descriptions, have the status of rigid designa tion because they are based on as the actual nature of the particular things as they can be known by humankind. Kripke grants that, "In general, science attempts, by investigating basic structural traits, to find the nature, and thus the essence (in the philosophical sense) [a priori] of the kind" (Kripke 138).

An example of essence, understood as what a thing necessarily is according to science, is explained by Hilary Putnam in an article entitled "Meaning and Reference," Putnam complicates the theory of reference and names a hypothetical' substitution problem' in a counterfactual world.

This world contains a liquid possessing the same superficial properties as Earth's water: "it is indistinguishable from water at normal temperatures and pressures," except it is composed of XYZ as opposed to H 20 (Putnam 121b). The substance is known as and called 'water' inthis counterfactual 'Twin Earth' , but according to Putnam's analysis, largely based on Kripke's, this substance is not water. It is a scientifi足 cally known fact that in our world 'H20 is water' is a true identity statement. That is, this statement is not contingently known because it is an identity statement between two names. In this case, if the statement "H20 =water" is true and if they are both names (that is, they are rigid designators), then "H20 = water" is a necessary truth. It is interesting to /1

62

JESSICA H. MITCHELL

note that even though the statement is not contingent, it still must be true aposteriori. Because this non-contingent claim is known aposteriori it involves an element of uncertainty. Even though we know water to chemically be H 20, it is possible in the future that, another scientifically determined composi­ tion will be discovered thatis more accurate. The new theory of reference, :insofar as it considers the essences of things, rests on the dynamic shoulders of empirical science.

Species, Katz, and The New Theory of Reference Many of the above mentioned examples and concepts prove helpful when examining the main issues in Eric Katz's article. The implications of the new theory are important in giving additional reasons for the implausibility of/the substi­ tution problem' as well as for providing a basis for arriving at an environmental ethic similar to the community model' from Katz's essay.

The substitution problem, as was explained earlier, is not a plausible problem. If it is acknowledged that species have intrinsic value by virtue of being natural, as was suggested earlier, then it is not plausible that this"serious moral prob­ lem," of replacing intrinsically valued species with other entities capable of carrying out the original's understood function in an ecosystem, could arise. Indeed, Ka tz admits in his article that such a substitution has never really occurred.

After suggesting that such a thing could happen, how­ ever, he also stated that if the substitution problem occurred it would be morally wrong, because it directly"involves the ideas of identity, integrity, or intrinsic value applied to individual organisms and species" (Katz 253). He later says in the article, "a technically adequate functional substitute, because it is not an outgrowth of the original natural pro­ cesses of the system, does not possess the same intrinsic value as the original entity" (Katz 254,245). Therefore, the identi­ ties of species are separate from and independent of each other.

Such natural independence is in part why something is intrinsically valuable. This is very similar to the statement I

THE VALUE OF NATURAL KINDS

describing Kripke' s conception of identities being a combina­ tion of origin and substance I both of which are essential. As Kripke surmises, "If a material object has its origin from a certain hunk of matter, it could not have had its origin in any other matter" (Kripke 114). Because Kripke considers origin andsubstanceuniquecharacteristicsofanindependentiden­ tHy, it seems that these are the elements that cause a being to be separate and independent.

It follows that, even if some other entity or mechanical object were capable of replacing a species' functional rolet the replacement would never 'be' the species. It would not acquire the identity of the replaced species by mimicking the functional role. Hilary Putnam gives a concise accountofthis in the example of water on Twin Earth. Putnam explains that even if XYZ fulfills the"operational definition," that is, has the same superficial properties as water, it is not water because it does not have the same essential properties as the stuff on Earth (Putnam 129-30b). Therefore, by usingPutnam's example I maintain that in Katz's 'substitution problem', even if the functional role is fulfilled and the thing looks superficially like the original species, it is not and cannot be that species, since its essence is necessarily different. The origin and substance of the two species are different; there­ fore, the identity of the replacement species is different from that of the original species: A species cannot be replaced.

Although on the surface this link between Katz and Kripke's analysis seems tenuous, it is important to under­ stand, because Kripke guarantees the identity and individu­ ality of unique beings. It follows, in Kripke's analysis, that a separate entity has also a separate identity and essence, since identity is tied up in origin and substance (Kripke 114). It is unique in its fundamental properties which set it away from and distinguish it from other entities. Although many of these properties may be shared with other entities (such as function which could be mimicked), science and philosophy grantthatthereissomething, an essence, thatisuniqueto that specific entity. It is helpful here to repeat Katz's claim that, the intrinsic value of natural entities is their source or origin.

II

64

JESSICA

H. MITCHELL

.. " (Katz 254, 249-50). These conclusions provide the base for a balanced ethic by privileging origin and uniqueness over mere "role" within an ecosystem. A combined consideration of all three of these properties helps locate the intrinsic and instrumental value of each species.

As we have seen, by attempting to extend this principle to common names, Putnam also extends the analysis of essence. If the theory of names is applied to a collective entity (a species for instance), then that group should have some fundamental, essential property that grants it indiViduality and distinction. In Katz' s article this is explained as justifica­ tion for the preservation of rare species which, it seems, no longer have any instrumental worth in an ecosystem. Be­ cause a Panda Bear is fundamentally unlike any other spe­ cies, since it maintains its own identity and essential proper­ ties, it is intrinsic value that justifies preserving the species.

As Katz explains, They have no instrumental value [supposedly], since the ecological system seems to function quite well without them. Thus, if they are to be preserved or protected, as environmentalist policies universally dictate, it must be because of their intrinsic value.

(Katz 255) One concern that could arise from applying theory of name principles to groups is that anindividual ina group will lose its individuality and merely represent the 'group es­ sence' as a whole. This difference is not however lost, His,in part, indicated in a semantic specification - the names are different. Let's take the case of a herd of deer. Because they are a group how is it possible that each member in the group is a distinct individual if they do not have different names? It is possible if, when referring to a specific deer, we say, 'that' deer. 'That', in this case, is a rigid designator intended to indicate only one deer from the herd.

THE

V ALUE OF NATURAL KINDS

Another way to support Katz's search for an adequate environmental ethic is to examine his use of counterfactual worlds in view of Kripke's thought. In so doing, we can arrive at a theoretically-based appeal which leads to a ratio­ nale for preservation that is very similar to Katz's community model. Hence, Kripke's theory can inform the future devel­ opment of a feasible environmental ethic that advocates species preservation.

By applying Kripke's philosophy, we come to a rather astonishing realization. As was mentioned earlier, because of the uncertainty of science necessary truths, essences, and essential properties cannot always be known apriori. If, then, there is still much to learn about relationships and essences and composition, any description that is not 'analytic's car­ ries with it an element of doubt. Kripke is careful to deal with this topic in Naming and Necessity. In a footnote he indicates that he did not try to deal with the delicate issues involving analyticity here, but does acknowledge Putnam's work (Kripke 122-23). Later, however, in the summary, he warns that, "For species, as for proper names, the way the reference of a term is fixed should not be regarded as'a synonym for the term" (Kripke 135).

Following this realization, we can see the danger in considering fixed references to be analytic truths. It is still true in our world that definite descriptions, descriptions which supposedly capture the essence of a thing or species, can change as our knowledge increases and develops. We first had to learn that water =H 20. It may still be the case, as was mentioned earlier, that H 20 may not be the most accu­ rate representation of water's composition. Just so, then, that our understanding of what constitutes a species' eSSel'lCe, is based on society's understanding of its chemical composi­ tion, its chromosomal makeup, and even its ecosystemic role.. This description, it seems, is not necessarily analytic, but is just a description that at least temporarily 'fixes the referent!. Our knowledge about a species' essence may

65

H.MITCHELL

change in the future. In view of these empirical scientific matters, we may doubt that a successful analytic statement about a species in possible.

These arguments about a species' value and its essence are fundamental in developing an understanding of the relationship between, not only a species and its role, but also one species with another. It is only possible, once this relationship is understood, to make a determination about the continued existence or obliteration of a unique identity.

Although this investigation of Katz from Kripke's perspec­ tive does not neatly solve the problems surrounding species valuation, it informs the debate, developing the issues in question. If there is no way for us to 'know' what a species' r~al essence is, or its role is not a part of its essence, it follows that Kripke's theory of names could assist both the impetus for preservation and the justification for destruction. Preser­ vation would be based on the arguments I have illuminated in this essay, and destruction could be justified because of what is not known. There is still a personal judgment involved that theory can only influence, not dictate. It is, however, helpful to note that our understandings of a spe­ cies, as was shown above, is merely dependent upon a fixed reference based upon the knowledge we have, not the knowl­ edge we might gain. Simply, preservation need not be based on an instrumental futurity because of the knowledge we lack or could lack - even if we do not consider the potential use-value of a species, because it is valued intrinsically by virtue of being naturat it seems as though it should not be destroyed or replaced.

WORKS CITED

Katz, Eric. "Organism, Community, and the 'Substitution Problem'," Environmental Ethics 7 (Fall 1985): 241­ 256.

THE VALUE OF NATURAL KINDS

Kripke, Saul. Naming and Necessity, Cambridge: Harvard University Press, 1972.

PutnamJ-Iilary. "ls Semantics Possible? ," inNaming, Necessity, and Natural Kinds, edited by Stephen P. Schwartz.

Ithaca: Cornell University Press, 1977. p. 102-118.

"Meaning and Reference," in Naming, Necessity, and Natural Kinds, edited by Stephen P. Schwartz. Ithaca: Cornell University Press, 1977. 119-132.

NOTES

1. Katz makes it clear that an ethic based solely on a species' intrinsic worth is equally unacceptable. It seems that this intrinsically valuable ethic would lead to a completely individualistic ethic denying species interdependency, since An entity valued intrinsically requires no relationships with any other entities."

2. Italics added.

3. Italics added.

4. It is interesting to question Katz's meaning when he says 'artificially human'. For instance, is an animal that is conceived through artificial insemination in a zoo not consid足 ered natural because its origins are in some fundamental way artificial human processes?

5. Italics added. See also p. 58 6. A description is not a rigid designator that is just like a name "unless (of course) we happen to use essential prop足 erties in our description" (57).

7. There is extensive literature which investigates these concerns, much of the work of Hilary Putnam, but this topic is for the most part beyond the scope of this paper.

8. Kripke understood analyticity to be descriptions that were both necessary and a priori.

11

FREUD ON WRITING: SOME HISTORICIST PERSPECTIVES

Sanja Perovic

McMaster University

The link between psychoanalysis and writing is inextri~ cable, especially since psychoanalytic theory both explains and justifies itself through analogies with writing. In particu­ lar, I wish to examine two such analogies: the textual analogy of the psyche's content in Freud's Interpretation ofDreams and the writing machine analogy of the psyche's structure in Freud's Note on the 'Mystic Writing Pad'. The ramifications of using the written text as a paradigm to explain psychical behavior affects theories of how writing produces meaning, of how the psyche negotiates its own textual nature and how this textual basis of psychoanalysis effects the institutional nature of psychoanalytic practice. I wish to focus on the relation between writing and the interpretation of psychical phenomena at their j oint intersection with history by looking at three readings of Freud by Derrida, de Certeau and Deleuze/Guattari. In brief, the following question will be considered: how does writing produce meaning and what does this historicizing function imply about the psyche in general?

Even a cursory glance at Freud's dream interpretation reveals a unique approach. Freud both uses his own dreams as scientific material and refuses to refer to a fixed universal key of translation. It is the latter that allows Freud to provide an innovative explanation of the relation between the signi­ fier and the signified in dreams. As Derrida points out: The absence of an exhaustive and absolutely infallible code means that in psychic writing which thus prefig­ ures the meaning of writing in general, the difference Perovic is a 1996 graduate ojMcMaster Universihj, graduating wi tit a combined honours degree in philosophy and comparstive literature. Perovic is currently pursubtg Jhe first year oj a: two year Master's degree in philosophy.

EPISTEME • VOLUME VIII· SEPTEMBER

1997

FREUD ON WRITING

between signifier and signified is never radical.

(Derrida 210) Indeed, Freud further suggests that dreams completelyeman­ cipate the signifier from the signified. This results in the treatment of words as things in themselves to the extent that words and plays on words are almost solely responsible for conveying meaning. However this treatment of signifiers immediately poses a problem for translating dreams. If the signifiers in dreams manufacture their own significance, the very arbitrariness of this production resists the possibility of translation. This is because translation requires a permanent code enabling the substitution or transformation of signifiers while retaining the same signified, always present. despite the absence of any specific signifier" (Derrida 210). It is clear that object of contention is the presupposed presence of meaning itself. On a prima facie level, by postulating both latent and manifest contents Freud seems to follow the paradigm of translation: transforming the latent signifiers into manifest signifiers which better express the always present but buried meaning which persists through the substitution game.

However, a closer look at Freud's project reveals something different. The meaning of dreams does not so much satisfy the checklist of metaphysical attributes of immutability ­ enduring presence, atemporality, unity - but is itself a pro­ duction and hence fundamentally historical. This interpreta­ tion of Freud, somewhat against his intentions but without betraying him, now needs to be validated.

The motivation for postulating a two-tiered conception of meaning itself follows from Freud's conviction that mani­ fest meanings are generally deliberate distortions which dissimulate the wishes that the dreamers want to but are 'uncomfortable with' fulfilling. Freud first explainS the psy­ chical basis for such dissimulation via an analogy with writ­ ers. Dream wishes are like books whose meaning needs to be distorted to circumvent political censorship. One notable feature of this analogy is that Freud distinguishes between oral and written pronouncements. Oral pronouncements are 1/

70

PEROVIC

repressed after they have been spoken and so exhibit silence rather than distortion whereas written pronouncements are suppressed beforehand if they intend to make it into print.

Thus, the "stricter the [social] censorship, the more far­ reaching will be the disguise and the more ingenious the means employed for putting the reader on the scent of the true meaning" (Freud 224). With this analogy, Freud not only abrogates individual/social distinctions by using social phe­ nomena to explain matters of the individual psyche but he also privileges the written text over the phonetic one as the closest approximation of the human psyche.

For Derrida, this is crucial. It evinces a radical departure from the logocentric nostalgia which privileges speech as unmediated (logos as both words and reason; particularly the Christian adaptation of logos for the original Word, uniting word and deed in completeness and self-sufficiency) over writing as a mediated and hence corrupted, 'after the fact' supplement to the original meaning. If Freud conceives the psyche as a written text, then he already sees it as deferring the originally intended meaning. Moreover if repression precedes articulation, then all meaning and subse­ quently all understanding of ourselves is always already a dissimulation and interpretation. It is always historical and never original. Freud himself further validates this view of the historicized unconscious by describing the self as initially fractured and fundamentally plural. Dream distortion, like all conscious thought, is the handiwork of two agencies whereby the distressing dreams do in fact contain some­ thing which is distressing to the second agency, but some­ thing which at the same time fulfills a wish on the part of the first agency" (Freud 228). Thus, the basic epistemic unit of Freudian psychology, the individual, consists of two mutu­ ally conflicting agents instead of the unified Kantian indi­ vidual which is the basis of enlightenment views of con­ sciousness. Instead of Kant's transcendental subject (who is the fixed, unified organizer of his sensations) Freud empha­ sizes the irrational. the unseen, the involuntary controlling the voluntary. In short, everything outside the scientific II

FREUD ON WRITING

domain of reason. Not only does Freud render the passions as motive agents, he also considers language as the joint production of desire and force. Desire expresses itself textu­ ally to bypass self-censorship. Writing is born out of desire's repression and so too is always originally repressed.

Derrida explains the significance of considering meaning as the differential between the two forces of desire and repression: There is no present text in general and there is not even a past present text, a text which is past as having been present. The textis not conceivable in an originary or modified form of presence. The unconscious text is already a weave ofpure traces, differences in which force and meaning are united - a text nowhere present, consisting of archives which are always already tran­ scriptions. Originaryprints. Everything begins with production. Always already: repositories of meaning which was never present, whose signified presence was always reconstituted by deferral. (Derrida 211) There is no present text because repression (the social censor) always defers and by deferral produces meaning. The 'originary prints', then, are the product of two actions. Be­ cause meaI').ing is only articulated after repression, it is al­ ways already deferred (desire only achieves articulation through negotiation with the censor). Furthermore, the 'uncorrupted', 'original' presence of desire is always recon­ stituted by this deferral; it is interpreted or produced and not uncovered in its unadulterated state. Freud has thus historicized Kant's transcendental subject. Since we are initially plural and self-alienated, we also possess an inner historicity. Our own existence and self-knowledge is not something freely articulated but through its original defor­ mation implies a production of meaning.

Both the pervasiveness of this deferral as well as Freud's task of founding a discourse out of the nonverbal interaction of psychic forces is encapsulated in his treatment of rhetorical figures. Rhetorical figures in their capacity as constitutive

71

72

SANJA PEROVIC

components of meaning are themselves events (hence his­ torical) by which deformation creates gaps between what is desired, the expression of this desire and its fulfillment. In one of Freud's dreams he realizes that "the wish to go to Rome had become in my dream -like a cloak and symbol for a number of other passionate wishes" (Freud 285). The manifest content is here explicitly identified with the appar­ ently concurrent rhetorical figures of symbol and veil. But for something to qualify as a symbol (as opposed to being a trivial and hence negligible statement) is to confront us with the history of its effects. Thus although the symbol confronts us, it also masks the process by which it gains its importance and why it is meaningful in the first place.

According to de Certeau, this symbol/ mask duality is"at the heart of Freudian discoveries - the return of the re­ pressed" (de Certeau 3). Indeed he states that this 'mecha­ nism': is linked to a certain conception of time and memory, according to which consciousness is both the decep­ tive mask and operative trace of events that organize the present. (de Certeau 3) Thus the symbol is both an archive of the differences between memories forgotten and remembered (why it is meaningful in the first place) - this is the operative trace - and the violence done to meaning to ensure its permanence - this is the deceptive mask. If we define the symbol as a special kind of sign that has practically effaced its referent, we approach Freud' 5 notion of condensation. For example, in the dream about Irma's injection, Freud interprets Irma as a condensed figure who stands for many people: his daughter, the patient who succumbed to poisoning, his wife, her friend. Irma has become the collective image for people who had been sacri­ ficed to the work of condensation" (Freud 406). On the one hand, the use of 'sacrifice' indicates a violence done to mean­ ing - a dissimulation. On the other hand, the condensation creates a proliferation of signifiers because such multiple /I

II

FREUD ON WRITING

determination makes it easier for an element to force its way into the dream-content" (Freud 402). This, at least theoreti­ cally, implies an infinity of interpretations. Aninfinitywhich too has been sacrificed for the sake of intelligibility. In this sense, the intelligibility of the present, its ability to explain itself to itself "takes the place of the history lost to it"; both because its meaningfulness is itself derived from a dissimu­ lation of meaning (the violence done to meaning in order to communicate it) and because intelligibility itself is based on a system of exclusion (de Certeau 29b). Thus, not only is the present itself always reconstituted by deferral (the original violence which communicates meaning) but itself must colo­ nize and empty out the past in order to retain its clarity.

The same implications hold for displacement, the other rhetorical figure Freud emphasizes. The consequence of the displacement is that the dream-content no longer resembles the core of dream-thought and that the dream gives no more than a distortion of the dream-wish which exists in the unconscious" (Freud 417). Distortion is the type of manifes­ tation (dream-content) which is a disguise, a willful substi­ tute which nevertheless fulfills the dream-thought in its fictive, hence rhetorical sphere. However, if the psychic structure of desire and repression is universal, then all articu­ lation of desire will need to circumvent repressive censorship and so will always be removed from the desire it wishes to articulate. This vitiates the possibility of anyone approach­ ing a finalist interpretation of latent meaning. The two rhetorical figures of condensation and displacement are cru­ cial in interpreting the textual fabric of the psyche; both are figures that are not self-evident but require a reading. Con­ densation in its near equivocation with symbols implies a reading concomitant with the rhetoric of optics; its heraldic qualities require a fine-tuned gaze; whereas displacement implies a complex grammar of interpretation.

This rhetoric of optics crucially demonstrates how Freud once again privileges the written over phonetic language, One way Freud emphasizes the written text is by comparing the dream-content with a pictographic script (Freud 381-2).

1/

74

SANJA PEROVIC

Words in their capacity as pictures can mean many contradic­ tory things at once. As Derrida points out, this is in contrast with phonetic writing which necessarily relies on linear execution, always moves"from presentto presentpoint" and so remains in II profound complicity with logos (or the time of logic) which is dominated by the principle of noncontradic­ tion, the cornerstone of all metaphysics of presence" (Derrida 217). Freud's allegiance to nonphoneticwriting, particularly emphasized in its resistance of all logical relations and linear unfoldings of narrative, is further exposed in the following paragraph: The different portions of this complicated structure stand, of course, in the most manifold logical rela­ tions to one another. They can represent foreground and background, digressions and illustrations, con­ ditions, chains of evidence and counter-arguments.

When the whole mass of these dream-thoughts is brought under the pressure of the dream-work, and its elements are turned about, broken into fragments and jammed together - almost like pack-ice - the question arises of what happens to the logical connec­ tions which have hitherto formed its framework.

What representation do dreams provide for 'if', 'be­ cause', 'just as', 'although', 'either-or', and all the other conjunctions without which we cannot under­ stand sentences or speeches? (Freud 422) Aside from the spatialization or optical representation associated with referring to dream-content as pack-ice and hence circumvention of this 'time of logic', we have here a most peculiar situation. The crux of the Freudian position is that the unconscious causes certain conscious behaviors. But here, Freud insists that the whole causal contribution of the unconscious to understanding our desires is not immanent within the dream-content. Instead our discursive under­ standing interprets the unconscious, this pack-ice, as caus­ ally efficacious. For example, whereas the dream-content is Jand', 'and', 'and' ad infinitum we substitute 'either', 'or',

FREUD ON WRITING

'because' etc. In other words, every narrative history, even the personal biography of the analysand involves attributing a motivation and causallinks which are not immanent in the unconscious. For scientific purposes this has a two-fold significance: first it exposes how causal and logical connec­ tions used in constructing historical narrative are substitu­ tions, interpretations; in short, fictions and second implies that the unconscious can never be conceptualized in its original state because understanding it is always already an interpretation. So far, however, Freud has only elucidated various characteristics of the psyche's content by comparing dreams with texts and rhetorical figures. Now, I will turn to the Mystic Writing Pad to investigate another textual meta­ phor which, this time, will express the structure or how the psyche works.

Freud begins his exposition by delineating the merits and limitations of two forms of writing: paper-pen and chalk­ slate. Although the advantage of paper-pen writing lies in its ability to maintain a 'permanent trace' the receptive capacity of the writing surface is finite, easily exhaustible and cannot lose whatever trace has been inscribed on it. The chalk-slate method, on the contrary, both has a receptive capacity for an unlimited time" and can erase the memory traces which have become uninteresting and obsolete. It, however, suffers the reverse of the paper-pen method, namely its complete inabil­ ity to maintain a permanent trace (Freud 227b). But with the invention of the mystic writing pad, Freud finds an appara­ tus which has If an unlimited capacity for new perceptions and nevertheless lays down permanent - even though not unalterable - memory-traces of them" (Freud 228b). Freud describes the functioning of the pad: I

To make use of the Mystic Pad, one writes upon the celluloid portion of the covering-sheet which rests on the wax slab. For this purpose no pencil or chalk is necessary, since the writing does not depend on the material being deposited on the receptive surfac.e ...a pointed stilus scratches the surface, the depressIOns

76

PEROVIC

upon which constitute the 'writing' ...If one wishes to destroy what has been written, all that is necessary is to raise the double covering-sheet from the wax slab by a light pulL.the close contact between the waxed paper and the wax slab at the places where it has been scratched (upon which the visibility of the writing depended) is thus brought to an end and does not recur when the two surfaces come together once more. (Freud 229b) This apparatus corresponds with remarkable precision to Freud's radically dualistic model of consciousness. The system Pcpt-Cs (analogous to the covering sheet) receives perceptions but in order for it to Hreact like a clean sheet to each new perception" it retains no permanent trace (Freud 22Sb). Instead the retention of the permanent traces is a function of the mnemic systems which lies behind the per­ ceptual system (analogous to the wax slab) and receive the trace of inscription. Already two important aspects emerge; first the psychical system consists of two separate but inter­ related systems thus once again affirming an original plural­ ity in our psychical constitution and second this psychical dualism itself implies that the depth of the Mystic Pad is simultaneously a depth without bottom, an infinite allusion and a perfectly superficial exteriority" (Derrida 227). In other words, this contraption allows for both a potentially infinite depth of implied meaning and the endless accretion of trace in the' deep' but limitless unconscious which lies behind the perception. It is a potentially infinite writing of the present (one can write on the celluloid ad infinitum) while simulta­ neously avoiding foundationalist pretenses (each system functions separately but in tandem: there is no first innocence in consciousness; to be conscious is to be, from the beginning, plural). Another important implication is that the ability to lift the celluloid and erase the writing is just as integral as the permanent traces which remain on the wax slab. Indeed, it seems tha t the psychical structure (and analogously writing) is contingent both on the contact between the celluloid and wax slab and the lifting of this contact which erases percep­ U

77

FREUD ON WRITING

tion and cleanses the receptive surface to receive sensory data anew. Memory functions to retain the trace of all these present moments. Further, memory points to the differences in what we perceive. It retains the difference between what perceptions are written down, engraved as trace and what perceptions permeate our consciousness without notice. Thus, not only are perceptions themselves cathected in that the only perceptions to contact with the receptive surface are those which interest us, but, derivatively, memory only retains traces of those perceptions cathected in the first place.

Because perception is already an inscription on the celluloid, it follows that the "perceived may be read only in the past, beneath perception and after it" (Derrida 224). What we are actually conscious of perceiving is the cathected perceptions inscribed on our receptive system. Thus, the 'perceived' is always initially delayed. This explains Derrida's claim that IJ memory or writing is the opening of that process of appear­ ance itself" (Derrida 224). Appearance itself is contingent on the writing and memory traces which by differentiating perceptions create the conditions for the appearance.

Freud next continues his analogy by comparing the ac­ tual breaking of contact which o,ccurs when the writing stops in the Mystic Writing Pad" with the IJ periodic, non-excitabil­ ity of the perceptual system" (Freud 225b). Freud postulates that consciousness functions as a series of periodic, discon­ tinuous "cathectic innervations as if the "unconscious stretches out feelers ...towards the external world and hastily withdraws them" (Freud 231b). This periodic discontinuity has several important ramifications. First, on the Freudian schema the essence of our consciousness does not lie outside of time such as the Cartesian cogito, Plato's forms or other classic, timeless specimens. Second, this implies a heteroge­ neous, discontinuous concept of time rather than the homo­ geneous time typical of historiographic writing. Time is not an empty, homogeneous volume through which we progress historically in a smooth continuum between the past and present. Rather consciousness itself is a discontinuous flick­ ering and so implies a different relation between the past and /I

II

78

SANJA PEROVIC

present. The present is disconnected from the past and this vitiates any simple cause-effect relation between them. More­ over, the spatial metaphor implied in the writing pad anal­ ogy further complicates linear cause-effect relations by in­ scribing the past on top of the present. The previous present is erased by the new present so that each present is simulta­ neously discontinuous with the past and yet contingent on its very erasure. The present maintains its necessary virginity by complete expulsion of the past; with gross imperialism it constitutes itself out of the past's exile. Moreover, if the present is always constituted anew by the deliberate forget­ fulness of the past this demonstrates that memory is always originally repressed, that this repression is what allows con­ sciousness to function and that the radical discontinuity of the subject implies that it too is always reconstituted anew.

Like pure perception, a pure subject" does not exist: we are written as we write, by the agency within us which always already keeps watch over perception" (Derrida 226). Thus, we write ourselves in order to be conscious but this is always both a deferring of the previous present and the erasure of ourselves.

But Freud did not .conclude this. Instead, Freud retains the idea of a subject whose potentially radical discontinuity is precariously maintained by the intangible concept of an unconscious retaining these 'permanent traces'. This meta­ physics of presence constitutes Freud's greatest betrayal. As Derrida states: " An unerasable trace is not a trace, it is a full presence, an immobile and incorruptible substance, a son of God, a sign of parousia...that is not a mortal germ" (Derrida 230). Freud neglects the fact that the script of dreams and the mystic pad are only representations and so are static, inferior analogies rather than the foundations of a science. So on the one hand, to use de Certeau's happy phrase, Freud radically "used the dream as a Trojan horse to historicize rhetoric and reintroduce it into the citadel of science" (de Certeau 23b).

But on the other hand, he kills the metaphor, naturalizes the unconscious not as hypothetical construct but as fact and so builds his interpretations and practice on the burial ground

FREUD ON WRmNG

of catachresis. In other words, Freud performs mythology or the science of origins. !tis precisely Freud's desire to treat the unconscious as a first cause that has evinced complaints from critics. Deleuze/Guattari complain that Freud's need for a transcendental guarantee imposes the verb 'to be' on these conjunctions of 'and', 'and, 'and' which is the infinity of interpretation (Deleuze/Guattari 25). The 'to be' of con足 sciousness is complicitous with the metaphysics of presence which permeates Western philosophy; seeking beginnings and erecting foundations instead of nullifying them. !tis this complicity which ties the institutional model of Freudian thought to psychoanalytic practice as institution.1 This curious Freudian junction involves the intersection between historicism and fiction. Once again writing history (individual or social) involves both exile and imperialism or " cannibalistic discourse" (de Certeau 29b). On the one hand, the writing process" establishes at the beginning of writing a separation or exile" (de Certeau 29b). Oedipus begins inter足 preting his origins when he realizes that he is exiled from them. The unknown past pervades his present with a feeling as uncanny as the hidden 'trace' of the past engraved on his ankles. Thus the interpretation of origins, this Oedipal discourse, is founded in absence. Discourse is separated from its referent just as the 'present' of consciousness is always constituted by the forgotten past it colonizes (the writing pad is always erased). On other hand, writing is a " cannibalistic discourse" which takes the place of the 'history lost to it" (de Certeau 29b). In other words, dreams are initially alienated from their meaning because they express desires mutated by their articulation, nevertheless they also retain the trace which calls out to us to interpret them. But when we interpret we also deform the meaning of this trace and so the historical narration is cannibalistic: explanation crossesout the origin in order to speak about it. This is where the betrayal of psychoanalysis occurs. It substitutes the crossing out, the negation, the acknowledgment of the ab足 sence of origin (which is always originally deferred) for survival under reprieve" (Deleuze/ Guattari 125). The can足 JI

80

SANJA PEROVIC

nibalistic discourse equates discourse and reality by impos­ ing its discourse as the law governing the real" (de Certeau 32b). Thus, it legitimates its own interpretation - rhetorical, fictive, itself a vehicle of deferral - by making it 'scientific'.

This is where de Certeau locates institutionalizing ges­ ture of psychoanalysis which differs from fiction precisely because it claims it "might do or become what it says of writing" (de Certeau 30b). So, if writing is the result of the negotiation of desire with the censor then to write this nego­ tiation is essentially to write and construct the unconscious.

Because there is no palpable origin, there is no way to make the unconscious conscious without engaging in the original duplicity of writing. The meaningfulness of the unconscious is produced, not recouped. But although for both Freud and a poet such as Schiller there is a loss of knowledge the outcome is different: "Freud's theoretic production is permit­ ted by a loss of knowledge while for Schiller poetic creation is permitted by a disappearance of being" (de Certeau 30b).

Once again the presence/ absence dichotomy resurfaces. For Freud the loss of knowledge (exile from origin) avoids the nihilism and profound antilogos of this complete loss of 'being' through masking its own production. It creates a discourse that speaks for the referent 'being' which, by the sleighthand of the psychoanalytic institution's own credibil­ ity as curative, is resurrected from the netherworld of noth­ ingness. Freud produces 'being', naturalizes iUn the name of the psychoanalytic institution and then proceeds to inscribe this 'real' unconscious with a truly 'documentary' history.

This nothingness is, in fact, the hinge that legitimates the authoritative position of the analyst. He intervenes in the name of this nothingness and then proceeds to stamp his historicizing of the unconscious with the mark of reali ty. The institution qualifies the slide from historicity to epistemic skepticism (or poetic nothingness) via the authority of his­ tory'as it happened'. Historicism which refuses to recognize its fictiveness dissimula tes itself by the mask of official, grand history. /I

FREUD ON WRITING

Thus, although Freud ventures into historicism, he re足 turns to historiography. But the textual analogies of the psyche, from the textual content of the psyche in the I nterpre足 tation of Dreams to the textual structure of the psyche in the Mystic Writing Pad have done their work. Freud may have betrayed his historicizing activities - historicizing rhetoric, demonstrating the original dissimulation of writing, exceed足 ing the phonetic discourse typical of metaphysics, temporalizing and spatializing writing and the unconscious - but they, in turn, also betray him. To use Oedipal imagery against Freud, the son, the offspring of his work in turn betrays the father. The putative authority of psychoanalysis is undone by the postulates it needs in order to exist.

81

SANJA PEROVIC

82

WORKS CITED

de Certeau, MicheL "Psychoanalysis and Its History," in Heterologies: Discourse on the Other, trans. Brian Massumi. Minneapolis: University of Minnesota Press, 1986. p. 3-16. liThe Freudian Novel: History and Literature," in Heterologies: Discourse on the Other trans Brian Massumi. Minneapolis: University of Minnesota Press, 1986. p.17-34.

L

Deleuze, Gilles and Felix Guattari. A Thousand Plateaus, trans.

BrianMassumi. Minneapolis: University of Minnesota Press, 1987.

Derrida, Jacques. Freud and the Scene of Writing," in Wri ting and Difference, trans. Alan Bass.(Chicago: University of Chicago press, 1978. p.196-231.

II

Freud, Sigmund. The Interpretation of Dreams, trans. James Strachey. London: Penquin Books,1953.

-. "Note on the 'Mystic Writing Pad'," in the Complete Psychological Works of Sigmund Freud, Vol XIX, trans.

James Strachey. London: Hogarth Press, 1961. p.227足 34.

NOTES

1. "The state as a model for the book and for thought has a long history: logos, the philosopher-king, the transcen足 dence of the Idea, the interiority of the concept, the republic of minds, the court of reason, the functionaries of thought, man as legislator and as subject" (Deleuze/Guattari 24).