Essence and E ND I N Aris Totle
Essence and E ND I N Aris Totle
Essence and E ND I N Aris Totle
22 hours page 73
E SSE N C E A ND EN D IN ARISTOTLE
JACO B RO S E N
. Introduction
Three [modes of cause] often converge upon one thing: the what-is-it and
the for-the-sake-of-which are one, and the primary source of change is one
in kind with these; for a man generates a man. (Phys. . , a)
Jacob Rosen
also most enduring legacies. By bringing together these two aspects of his thought, the claim promises to illuminate both. Thus,
for example, when scholars are puzzled as to how an end should
be thought of as a cause, they often appeal for help to forms, or
to some drive or irreducible potential for form (these things, at
least, actually exist at the time of the explanandum). Conversely,
when they wonder in what way a form should be thought of as a
cause, they often look for answers in the role of form as a goal of
generation and development.
What is more, the sameness claim promises a major simplication
of Aristotles doctrine of the four causes. It suggests that, in the natural world, the number of causal factors can be reduced from four
to three, or even just two: matter and form. Suggestions for such a
simplication are found already among the earliest and best extant
commentaries on Aristotle, and continue to be put forward today.
But despite these promises, the claim has done more to obscure
than to help. If its upshot really is, as Ross puts it, that the nal
cause has been completely identied with the formal, then it cannot represent Aristotles considered view. Its uncritical acceptance
by many commentators has led to error and confusion, not only
about Aristotles substantive beliefs, but about his very conceptual
apparatus for thinking about causation.
After talking of confusion and error, I want to emphasize the con
Gotthelf proposes an inuential denition of for the sake of in terms of irreducible potentials for form in A. Gotthelf, Aristotles Conception of Final Causality,
in A. Gotthelf and J. Lennox (eds.), Philosophical Issues in Aristotles Biology (Cambridge, ), at and n. . Hankinson writes: Final causes, then,
are parts of reality in the sense that the drive for form that they represent is written directly into the structure of things (R. J. Hankinson, Cause and Explanation in
Ancient Greek Thought [Cause and Explanation] (Oxford, ), ).
According to Bostock, in biological contexts Aristotle thinks that a form functions as a cause by way of being the goal towards which the animal develops (D.
Bostock, Aristotle on Teleology in Nature, in id., Space, Time, Matter, and Form:
Essays on Aristotles Physics [Essays] (Oxford, ), at ). For a similar
thought see Hankinson, Cause and Explanation, .
For example, in a study of modal concepts, Stalnaker writes: One claries such
notions, not by reducing them to something else, but by developing ones theories
in terms of them (R. Stalnaker, Ways a World Might Be (Oxford, ), ). Gideon
Rosen describes a related strategy in his discussion of metaphysical dependence:
The plan is to begin to lay out the principles that govern this relation and its interaction with other important philosophical notions (G. Rosen, Metaphysical Dependence: Grounding and Reduction, in B. Hale and A. Homann (eds.), Modality:
Metaphysics, Logic, and Epistemology (Oxford, ), at ). Rosen does not
claim to be elucidating the concept in question, but only to be arguing that it is legitimate; still, I think he clearly contributes to the former project as well as the latter.
Only by and large. Sometimes our judgements about a specic case are tentative,
or divergent. Then the account that best deals with the clear cases can be allowed
to settle the unclear case (spoils to the victor). Sometimes our pattern of specic judgements turns out to be incoherent or otherwise in need of revision, and an
account of the concept can guide this revision. Again, sometimes a concept is introduced by stipulatory denition; in this case the denition presumably settles the
standards of correct use.
Jacob Rosen
It is therefore important to settle whether or not Aristotles sameness claim accurately reects his core understanding of formal and
nal causation. The second guiding idea suggests a method for addressing the question. We cannot, it is true, follow exactly the same
methodology that David Lewis and others do. We have no body
of folk intuitions to work from; all of our data consist of theoretical statements made by Aristotle. Nevertheless, I think, we can
distinguish between statements in Aristotle that are more and less
authoritative for our purposes. Some statements gure in reasonably workaday explanations of concrete biological or other natural
facts. Some statements derive from, or are used to derive, claims
that are clearly central to Aristotles thought, and are thus tightly
bound into his web of beliefs. Such statements are weighty. Other
statements are comparatively free-oating, or they are sweeping generalizations of the kind that philosophers often get wrong. These
statements are less weighty. I have no overall system on oer for
measuring the evidential weight of a statement. This paper follows
the motto: rst do it, then think about how to do it. It is an experiment in methodology, which will, I hope, lead to useful reection
on methodology.
Jacob Rosen
Important here is Aristotles remark that the account of the essence, i.e. the formal cause, and the for-the-sake-of-which as an
end, i.e. the nal cause, should pretty much be regarded as one;
and, again, his statement that these two things are the same.
A further passage to consider is found in Metaphysics :
When someone seeks the cause, since causes are spoken of in many ways,
one must state all the possible causes. For example, of a man: what is the
cause as matter? Is it the menses? What as mover? Is it the seed? What as
form? The essence. What as for-the-sake-of-which? The end. Perhaps these
are both the same. (Metaph. , ab, emphasis added)
Ross, Physics, .
While it is left somewhat vague quite how often this triple coincidence occurs, it presumably is intended to apply at least to all living things, which are Aristotles primary examples of substances (D. Bostock, Aristotles Theory of Form,
in id., Essays, at ). Bostock adds in a footnote (n. ) that Aristotle would
probably wish to identify the form and the purpose of many manufactured objects,
including houses, ships, walls, and saws.
Jacob Rosen
primary caregiver. I only mean that one person often occupies both
roles.
In the other direction, we should note that Aristotle is not merely
saying that the formal cause of one thing is often the nal cause of
some other thing. Rather, he should be understood as saying that
the formal cause of a given thing is often the same as the nal cause
of that same thing. The rst, and weaker, claim is without doubt
true for Aristotle, but it would not naturally be expressed by saying, straight out, that the formal cause and the nal cause are one
or the same. For comparison, I believe that everything that is to the
east of something is also to the west of something, and vice versa;
yet I would not say, what is to the east and what is to the west are
the same. I also believe that, often, the father of one person is also
the brother of another person; but this is not well expressed by saying, the father and the brother are often one.
Finally, we must consider what is meant in these passages by one
and same, given that Aristotle famously distinguishes several different uses for each of these terms. In our passages, Aristotle does
not explicitly qualify or restrict his assertion of unity or sameness
between formal and nal cause. This makes it natural to read him
as asserting sameness and oneness in their strictest and most dominant sense, which we may express by saying that a things formal
cause and its nal cause are the same per se and in number. To spell
this out: let A stand for a term picking out some item under some
description, and consider the best, most canonical way of lling in
the blanks in the following two sentences:
is the formal cause
of A;
is the nal cause of A. Aristotles sameness claim is
naturally read as asserting that, in many cases, the very same term
The loci classici for Aristotles distinctions are Top. . and Metaph. and .
The sorts of qualication we might look for and do not nd include in number
(but not in being), in kind (but not in number), accidentally, or, most vaguely,
in a way (). It is true that Aristotle gives what might be taken as signs of hesitation or qualication, namely the word often () in Phys. . , pretty much
() in GA . , and perhaps () in Metaph. . However, the rst pertains
only to the range of cases, not to the kind of sameness at issue; and the latter two
words are frequently used by Aristotle to soften his tone without indicating any real
limitation or uncertainty (cf. H. Bonitz, Index Aristotelicus (Berlin, ), b .
and a .).
Cf. Top. . , b. My strictest and most dominant corresponds to Aristotles . Aristotle omits the phrase per se ( ) in his gloss of strict
sameness in Top. . , but it is obviously intended: see Top. . , a, and the
fact that his tests for strict sameness would clearly be failed by true statements of
accidental sameness such as the seated man is the same as the cultivated man.
should ll both blanks. Or, at the very least, that the blank-llers
should be related as synonyms (like himation and lpion, two words
for a cloak) or as a word to the corresponding denition (like human and biped land animal).
This reading is natural in the light of the use to which Aristotle puts his sameness claim. For he uses it in the introduction to
Generation of Animals in order to justify the assertion that, having
spoken of the formal causes of animals, he has thereby also spoken
of their nal causes. This works most easily if he is thinking that
the questions what is its formal cause? and what is its nal cause?
should receive precisely the same answer, and not answers related
as, say, the woodworker and the mason are related when Johnny
is both woodworker and mason.
The natural reading may turn out not to be the right reading. But
it is, I think, the usual reading (for example, Ross says that Aristotle
has completely identied the formal and nal cause), and a reasonable place to start. If we think that a dierent kind of sameness
is in play, then we owe an explicit account of what kind of sameness
this is. I do not know of any such account in the recent literature.
(On the other hand, I have received some interesting suggestions in
spoken discussion, and will report them later.)
Ross, Physics, .
Jacob Rosen
Now, scholars sometimes say that a things work is the same as its
formal cause. But even these same scholars seem, as if led by the
truth itself, to contradict themselves, and rightly say that the two
are dierent. One way to establish the dierence is by observing
that a things work is something that need not actually be there in
order for the thing to be there. For example, the work of my eyes is
an activity, seeing, and my eyes are still there when no seeing is taking place, for example in the dark. Likewise, the work of an axe is
chopping, or perhaps chopped wood, and the axe is still there while
it hangs in the shed and no chopping or chopped wood is present.
By contrast, it seems evident that a things formal cause is something that necessarily is there so long as the thing is there. Taken
together, the last two points imply that it is possible for a things
formal cause to be there while its work is not there, and this implies
that the two are dierent. Since the work is a nal cause, it follows
that for a wide class of things, each thing has a nal cause that is
dierent from its formal cause.
In a moment I will oer a more formal version of the argument,
and provide evidence that Aristotle is committed to all the relevant
premisses. In preparation for that, I need to oer a few clarications.
First, a note about the term work. This corresponds to the Greek
ergon, also commonly translated function. It signies an activity or
product which a thing has the task of doing or making. I emphasize
that a things work is an activity or product, not the having it as ones
task to do or make this activity or product. The latter property
(perhaps job or functional state is an apt term for it?) will be
The soul is the characteristic functions and activities that are essential to the
organism (T. Irwin, Aristotle: Nicomachean Ethics, Translated, with Introduction,
Notes, and Glossary, nd edn. (Indianapolis, ), ); a full identity between
form and function . . . obtains in the case of fully realized forms (M. Leunissen,
Explanation and Teleology in Aristotles Science of Nature [Teleology] (Cambridge,
), n. ).
Aristotle therefore identies substance and form with rst actuality, the permanent state of the organism, as opposed to the intermittent vital activities (T. Irwin,
Aristotles First Principles (Oxford, ), ); Functions . . . exist on top of
the realized forms that constitute the rst type of nal cause. For instance, a completed house is the nal cause and the fully realized form of the art of house-building,
while shelter is the function and nal cause of the realized house (Leunissen, Teleology, ).
Jacob Rosen
Jacob Rosen
M. Nussbaum, Aristotles De motu animalium: Text with Translation, Commentary, and Interpretive Essays (Princeton, ), and .
Jacob Rosen
(The sense in which the body is for the sake of the soul is, I take it, a
special one; see Section ., especially n. .) Moreover, in particular cases such as teeth, he refers to the same things both as the work
and as the nal cause of the part in question. He also sometimes
combines talk of work with talk of nal causation, for example in
Parts of Animals:
Animals also have the nature of a mouth for the sake of these works, as well
asin those animals that breathe and are cooled from outsidefor the sake
of breathing. (PA . , a)
Teeth are for the sake of nourishment, in some animals also for defence, and
in humans for speech (GA . , b); the work of teeth is the preparation of
nourishment, in some animals also defence (PA . , b; see also PA . ,
b; GA . , a; . , b).
DA . , b (the parts of plants too [sc. like the parts of animals] are instruments); PA . , a (the lung is an instrument of breathing); . , a
(hands are an instrument); . , b (nature makes instruments [sc. body parts]
with a view to their work); GA . , a (the parts of the body are instruments
for an animals powers).
. , . I have some reservations
about relying on this passage, because, strictly speaking, Aristotle says that waking
activity is an end for animals, not that it is an end of animals.
, (GA . , a).
Another passage: Metaph. , b (something is a hand only if it is
capable of accomplishing the work: ).
Jacob Rosen
According to this passage, some given matter will actually constitute a given thing only when the appropriate form is actually there
informing the matter. Aristotle does not state that this is necessarily so, but it is plausible that he intends to be oering a scientic
truth here, and, according to Aristotle, scientic truths are necessary truths.
The thesis is also suggested in Aristotles explanations of particular cases. For example, when he analogizes in De anima . between
animals, axes, and eyes, he says of the formal cause of each of the latter two that if it were separated o (or went away), there would
For one example from the secondary literature see J. L. Ackrill, Aristotles
Denitions of Psuch [Psuch], in M. Nussbaum and A. O. Rorty (eds.), Essays
on Aristotles De anima [Essays] (Oxford, ), at : The form is what the
matter has to get or have if it is to become or be an X; for the matter, to become or
to be an X is precisely to get or to have the form. Presumably the has to in has to
get or have expresses necessity.
DA . , b, .
See e.g. Metaph. , b; , a; , a; , a.
For discussion see M. Frede, The Denition of Sensible Substances in Met. , in
D. Devereux and P. Pellegrin (eds.), Biologie, logique, et mtaphysique chez Aristote
(Paris, ), .
I assume that there is a unique formal cause for each kind. By unique I mean
specically uniquewe need not decide whether all ks share numerically one formal
cause; perhaps each individual k has its own individual form.
For the rst claim see Post. An. . , a (in conjunction with a, ); . ,
b. For the second claim see Post. An. . , b, , , .
Jacob Rosen
of the eye, the relevant terms would be seeing and sight: seeing
is per se a nal cause of the eye, sight is per se the formal cause of
the eye, and seeing is not strictly the same as sight. I emphasize that
my conclusion concerns sameness in the strictest and most dominant sense of the term. For all I have proven, it may yet turn out
that a things nal cause and its formal cause are the same in some
weaker sense. Indeed, for all I have proven, it could even turn out
that they are identical (in todays sense of identical, which may or
may not have been a concept employed by Aristotle himself). Carrying on with the eye, it could be that, although seeing is not strictly
the same as sight, seeing is identical with sight. I am condent that
such an identity does not in fact hold, but I have not proven that it
does not hold. We will meet this point again (Section .).
Jacob Rosen
mals form has good title to be called the nal cause of the animal.
(This line of thought is addressed to whole organisms only; it is not
designed to identify the formal and nal causes of parts or instruments.)
Such a view is endorsed, for example, by Gareth Matthews:
Now if the soul of a living thing is the cause of its living, and its living is
naturally directed towards the preservation of its species, then the souls
powers (the psychic powers we have been talking about) are presumably
powers naturally directed toward the preservation of the species of that
particular thing.
What are we to think of this view? To begin, let us grant for the sake
of argument that the working of an animal or plant always contributes to survival or reproduction. Then we are faced with a circle:
the form promotes the work (since it grounds the ability to work,
or is this ability), and the work promotes the form (since it preserves and reproduces the form). In this mutual furthering of form
and work, which is for the sake of which? Or, to shift the question,
which is the more intrinsically valuable, and which rather derives
its value from that of the other?
It seems to me that preference must go to the work, not to the
form. The situation is analogous to what we nd in the Ethics in the
relation between virtue and virtuous activity. Virtue is a state which
provides, or is, a disposition to act virtuously; acting virtuously
develops or reinforces the state of virtue. Aristotle is insistent
When one thing is for the sake of another, normally the latter is intrinsically
better than the former. This comes out, for example, in the reasoning about goods
and ends in NE book : see especially NE . , a, and . , a; also
EE . , a.
against Platothat the higher good is the activity, not the state.
As in the practical realm, I say, so in the biological. An activity such
as perception or thought is of basic intrinsic value; its value is akin
to the value of gods activity. The form of an embodied animal is
valuable because it is or provides the ability to perform such valuable activity. Aristotles higher valuation of activities over capacities
seems to be quite general, extending outside his ethical works into
the physical and metaphysical. For example, in Metaphysics
he argues that whenever a capacity is good, the corresponding activity is better.
I started out by granting the claim that, for Aristotle, the working of a living thing always contributes to its own survival or reproduction. I do not think that this should really be granted. In the
human case, theoretical activity is an obvious and acknowledged
sticking point for the claim. In other animals and in plants, there
are no such obvious counter-examples: the activities described in
Aristotles biological works all seem to be connected with getting
food, mating, and protecting oneself. Nevertheless, it is important
to see that Aristotles theoretical framework leaves room for animal
activities which are performed simply for their own intrinsic value,
and not for the sake of any contribution to survival or reproduction. Correspondingly, it leaves room for teleological explanations
of animal traits or parts in terms of their usefulness for intrinsically
valuable activities, without any regard to considerations of survival
or reproduction. This is a crucial dierence between Aristotles
framework and the prevailing Darwinian framework of today, and
it would be a shame to obscure it or cover it over.
.. Coming to be
Aristotles way of referring to nal causes consists in a mere fragment of a clause, that for the sake of which, and we might wonder
how to complete this fragment: that for the sake of which . . . what?
I have been taking the view that, in the case where A is an object
such as an animal or a plant, a nal cause of A is something for
the sake of which A is there. But one might argue for completing
Jacob Rosen
Metaph. , a: but this sort of cause [i.e. the ecient cause] is sought
for coming into being and perishing, while the other [i.e. the nal cause] is also
sought for being.
It has become a familiar point in discussions of Aristotles Ethics that his for
the sake of relation is more inclusive than a purely instrumental meansends relation. Even so, in standard examples, such as when I putt for the sake of playing golf
or play golf for the sake of having a good holiday (J. L. Ackrill, Aristotle on Eudaimonia, in A. O. Rorty (ed.), Essays on Aristotles Ethics (Berkeley, ), at
), I achieve the end because of or in virtue of the thing that is for the sake of the
end. I am playing golf in virtue of the fact that I am putting; I am having a good
holiday in virtue of the fact that I am playing golf. Now, a cat is there in virtue of
its forms being there, not vice versa, so it is implausible that the cat is there for the
sake of the forms being there.
Jacob Rosen
rst two, more elaborate ones, we should not use it in a way that is
ambiguous between them. We strive for clarity and precision, and
avoid vague or ambiguous expressions. As for Aristotles usage, it
is hard to be certain, but it seems unlikely that he meant B is a
cause of A to be ambiguous between B is a cause of being to A
and B is a cause of coming into being to A. Aristotle makes a great
many distinctions concerning the ambiguities of causal claims, and
he does not indicate that there is any ambiguity of this particular
kind.
Finally, to complete my argument, there are passages in which
Aristotle appears to use cause of A interchangeably with cause of
being to A. If the appearance is correct, and given that we and
Aristotle both wish to avoid ambiguity, then cause of A should
not be used as a proxy for cause of coming into being to A. In the
case of nal causation, this means that we should call B a nal cause
of A only if it is a nal cause of As being, that is, only if A is there
for the sake of B, and not if it is merely a nal cause of As coming
into being. So, although an organisms formal cause is a nal cause
of the organisms generation and maturation, this is no good reason
for calling the form a nal cause of the organism itself.
.. Benet
In a handful of passages, Aristotle distinguishes between dierent
senses in which one thing can be for the sake of another. It is worth
considering the distinctions he makes, to see if they deliver some
sense in which things can plausibly be said to be for the sake of
their forms.
The clearest elaboration of a distinction is found in Generation
of Animals . . Here Aristotle distinguishes between, on the one
hand, somethings being there in order to generate or produce a
given thing, and, on the other hand, somethings being there in
order to be used by the thing (GA . , a). For example,
a ute teacher is there for the sake of a autist in the rst sense,
whereas a ute is there for the sake of a autist in the second sense.
A much more condensed statement of a distinction is found in De
anima . . Here Aristotle tells us that that for the sake of which is
twofold; it encompasses (to translate in minimal fashion) that of
which and that to which. It is dicult to be sure what he means
here. There is fairly wide agreement nowadays that that of which
means an end to be attained or realized, and there is wide verbal
agreement in saying that that to which means someone or something to be beneted. But the verbal agreement masks a great disparity in understandings of benet. Some scholars think that to
benet someone is to bring him or her into a better condition.
Others think that our enemies can properly be called beneciaries of the measures we take to frighten them in battle. For these
scholars, benet has a touch of the gangsters euphemism about it,
or at any rate a rather broad meaning. Finally, some seem to think
that the notion of beneting something is equivalent to (or encompasses) the notion of being useful to it.
The third understanding of benet has the advantage of making Aristotles distinctions in De anima and Generation of Animals
line up pretty well with each other. I have no objection to it considered as an interpretation of the De anima passage, but I would
like to plead for more dierentiated terminology. There are many
reasons for keeping the notion of benet clearly separated from the
notion of usefulness. For one thing, it is widely assumed that something can be beneted, or beneted, only if it is changeable. But
it is possible to be useful to an art, which is presumably not (per se)
changeable. Moreover, as Plato has Socrates argue in Republic ,
Stephen Menn writes of being for the sake of something as the to-benetwhom, as an is for the sake of the art or the artisan (Menn, Programme, ).
Menn emphasizes that an art can be the to-benet-whom (as he calls it) of an
instrument (Programme, ).
Jacob Rosen
make the soul come out as the formal cause and a nal cause of one
and the same thing. Perhaps it can be done, but I am not optimistic.
What of the benet proposal? Aristotle does seem to hold that
an animals being is a benet to it: he says that being is choiceworthy
and lovable, and that a childs being is a great service done to it by
its father. It is not implausible to assign the enjoyment of benets
specically to the animals soul, and thus to say that the animals
being is a benet to the animals soul. However, it is dicult to go
further than this, and to claim that we have here a partial explanation of why the animal exists. Though it is plausible that each
animals existence is a benet to the animals soul, it is neither intuitively plausible nor (to my knowledge) ever asserted by Aristotle
that each animal exists for the sake of this benet to the animals
soul. Perhaps this strategy can somehow be carried o, but again I
am not optimistic.
.. Sameness of activity and capacity
I would like to mention one last strategy for upholding a kind of
sameness between formal and nal causes. I have not seen the strategy pursued in print, but it has arisen often in conversation about
the argument presented in Section above. An interlocutor begins by granting that a things formal cause and its nal cause are,
in a way, dierent. In particular, he says, typically a things formal
cause is a capacity while its nal cause is the corresponding activity.
For example, an animals formal cause, its soul, is a complex capacity for certain life activities, and the animals nal cause is those
life activities. But then, the interlocutor proceeds, a capacity and
the corresponding activity are the same.
It remains to spell out what kind of sameness is at issue. As an
opening move, the interlocutor notes that a capacity and an activity can typically be referred to by the same linguistic expression.
For example, if Aristotle says that something sees (horai), this can
mean either that the thing is able to see, or that it is actively seeing.
If something lives (zi), this can mean either that it is alive, or that
it is actively performing life activities (cf. NE . , a). So we
can answer the question of essence and the question of end with the
NE . , a; . , a.
The strategy addressed in this section has been defended in discussion (whether
from conviction or for dialectical purposes) by Stephan Schmid, Christian Pfeier,
Antonio Vargas, Jonathan Beere, Gavin Lawrence, and Calvin Normore.
Jacob Rosen
Metaph. , a .
Presumably there are several ways one could go from here, but what follows is
the only concrete proposal I have heard.
Jacob Rosen
Ackrill, Psuch, .
the form in order that . . . and for the sake of . . . count as answers to why questions, they must somehow be translatable into
statements of the form because . . . or as a result of . . .. But I
think we should keep an open mind about this. Doubtless, in order
that statements stand in inferential connections with certain sorts
of because statements, but these connections are complex and
will most likely not lead to any straightforward translation, denition, or reduction.
What plays the role of nal cause? Well, typically, where a kind is
dened by a capacity, the nal cause will be the activity or the thing
which kind-members are able to do or to make. (I oer this as a general rule, not as a necessary or conceptual truth.) For example,
the art of medicine is a capacity to produce health, and a person is
a doctor for the sake of health: health is the nal cause of a doctor.
Similarly, a house exists for the sake of sheltering bodies and goods,
and an eye exists for the sake of seeing. A living thing exists for the
sake of certain life activities: those activities are the nal cause of
the living thing.
On this picture, although the nal cause of a house is closely related to the formal cause of the house, the causes are not the same.
Its formal cause is the capacity to do something; its nal cause is
that which it is able to do, namely to shelter. Similarly, the nal
cause of a living thing, such as a cat, is related to but dierent from
the things formal cause. The formal cause is a capacity (this is why
the cat still exists while asleep), while the nal cause is the corresponding activity. If the argument of this paper is acceptable, then
we should not let Aristotles sameness claim deter us from adopting the picture I have just sketched; we should consider adopting
it even though it is clear that capacity and activity are dierent
(Metaph. , a).
What, in the end, should we think of Aristotles assertion of the
For example, Aristotle indicates that a nal causal explanation can sometimes be
given in the form because it is better thus ( : Phys. . , b).
Also, some of his arguments presuppose connections between nal causes and ecient causes: for example, he evidently assumes that if a process is eciently caused
entirely by weight and heaviness, then the process does not occur for the sake of covering and preserving (Phys. . , a).
Jacob Rosen
sameness claim? I am inclined to think that Aristotle was speaking loosely, or was making a subtle mistake. Relative to his own
purposes, the mistake is minor and easily corrected. For there are
truths in the neighbourhood of the sameness claim, and the neighbouring truths can do the work that Aristotle needs done. Aristotle
is mainly concerned with the question how a scientist should go
about describing the causes of things, and his main message is that
the scientist need not list formal causes and nal causes separately.
For example, in the passage we saw in Generation of Animals, his
whole point seems to be that, having discussed the formal causes
of animals, he need not give an additional set of lectures on their
nal causes. This point is reasonable provided only that a things
nal cause can be easily inferred from its formal cause, regardless of
whether it is strictly the same as it; and the latter claim is plausible in
Aristotles theoretical framework. Aristotles train of thought goes
through, and relative to this his mistake or loose expression is harmless. Relative to our purposes, on the other hand, when we are trying
to elucidate and reconstruct Aristotles causal concepts, the mistake
is harmful and it is crucial to recognize it as such. It encourages
inaccuracy in the identication of causal relations and relata. And
it obscures the fact that many ends in Aristotles natural world
indeed the highest ends, I thinkare not forms, but rather activities such as perceiving and knowing.
Humboldt-Universitt zu Berlin
BI BL I OG R APHY
Ackrill, J. L., Aristotle on Eudaimonia, in A. O. Rorty (ed.), Essays on
Aristotles Ethics (Berkeley, ), .
Aristotles Denitions of Psuch [Psuch], in Nussbaum and Rorty
(eds.), Essays, .
Bonitz, H., Index Aristotelicus (Berlin, ).
Bostock, D., Aristotle on Teleology in Nature, in id., Essays, .
Aristotles Theory of Form, in id., Essays, .
Space, Time, Matter, and Form: Essays on Aristotles Physics [Essays]
(Oxford, ).
Chellas, B., Modal Logic: An Introduction (Cambridge, ).
Frede, M., The Denition of Sensible Substances in Met. , in D.
Devereux and P. Pellegrin (eds.), Biologie, logique, et mtaphysique chez
Aristote (Paris, ), .
OX FO R D S T U D I E S
IN A N C IE N T
PHIL OS O P H Y
ED IT O R: BRAD I NW O O D
VOLUM E XLVI