Aristotle on Norms of Inquiry

Author(s): James G. Lennox

Source: HOPOS: The Journal of the International Society for the History of Philosophy of
Science, Vol. 1, No. 1 (Spring 2011), pp. 23-46
Published by: The University of Chicago Press on behalf of the International Society for the
History of Philosophy of Science

Stable URL: http://www.jstor.org/stable/10.1086/658482 .

Accessed: 28/08/2015 08:53
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

The University of Chicago Press and International Society for the History of Philosophy of Science are
collaborating with JSTOR to digitize, preserve and extend access to HOPOS: The Journal of the International
Society for the History of Philosophy of Science.

http://www.jstor.org

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

A R I S T O T L E O N NO R M S O F I N Q U I R Y
James G. Lennox
Where does Aristotle stand in the debate between rationalism and empiricism? The locus
classicus on this question, Posterior Analytics II. 19, seems clearly empiricist. Yet many
commentators have resisted this conclusion. Here, I review their arguments and conclude
that they rest in part on expectations for this text that go unfulfilled. I argue that this is
because his views about norms of empirical inquiry are in the rich methodological passages
in his scientific treatises. In support of this claim, I explore such passages in On Parts of
Animals and De anima. I argue that they reach distinct, though complementary, conclusions about the norms governing zoological and psychological inquiries.

1. Introduction
A classic question that has divided scholarship on Aristotle from the Greek
commentators forward is whether, when it comes to scientific first principles,
James G. Lennox, professor of history and philosophy of science, Department of History and
Philosophy of Science, University of Pittsburgh, 1017 Cathedral of Learning, 4200 Fifth Avenue,
Pittsburgh, PA 15260 ( [email protected]).
Ernan McMullin took up a 2-year residence as a visiting fellow at the Center for Philosophy of Science in
197879, shortly after I was appointed assistant professor of history and philosophy of science at the University
of Pittsburgh. We eventually became friends. I've never forgotten his kindness and encouragement toward me
in those years, and his work served us all as a model for our field. This article is dedicated to his memory.
I thank the participants in an eponymous seminar at the University of Pittsburgh for helpful discussion,
and especially Keith Bemer, Peter Distelzweig, and Allan Gotthelf. The line of argument here first took
shape as a presentation to the Philosophy Colloquium of Duquesne University. I thank Ron Polansky for
the invitation and the audience for probing questions. During my time as Biggs Lecturer at Washington
University, St. Louis (April 2010), I presented some of this material to the Faculty Seminar, during which
Eric Brown and Mariska Leunissen pressed helpfully on a number of my contentions. The occasion for this
publication was a presentation to an Aristotle Session at the 2010 meetings of the International Society for
History of Philosophy of Science in Budapest ( June 2010). On that occasion, I received helpful feedback
from Istvn Bodnar, Boris Hennig, John McCaskey, Pierre Pellegrin, and Tiberiu Popa. Finally,
encouraging discussions over the last couple of years with James Allen, David Charles, Alan Code, Allan
Gotthelf, Devin Henry, Aryeh Kosman, Mariska Leunissen, Greg Salmieri, and Joel Yurdin have helped
this project in innumerable ways, though I am confident none of them agrees entirely with its conclusions.
HOPOS: The Journal of the International Society for the History of Philosophy of Science, vol. 1 (Spring 2011).
2152-5188/2011/0101-0002$10.00. 2011 by the International Society for the History of Philosophy of Science.
All rights reserved.
23

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

Aristotle is an empiricist or a rationalistor put slightly differently, an inductivist or a coherentist.1 In a recent critical survey of various attempts to make
Aristotle out to be an empiricist on this question, Michael Ferejohn has legitimately questioned whether there is an inevitable anachronistic distortion that
arises from putting the issue in these terms (2009). Of course the rationalism/
empiricism divide has ancient roots in Hellenistic medicine, but all of the interpreters Ferejohn takes to task, he argues, are in one way or another framing
the issue in ways that presuppose the concerns of post-Cartesian epistemology.2
I will add that one has the impression that the answer given by a particular commentator stems more from a principle of charity than from positive evidence
for the attribution: Aristotle is a profound philosopher, and a profound philosopher should hold that first principles are grounded in the appropriate
way.
At first blush, it would seem obvious that Aristotle is on the empiricist/
inductivist side of this issue. After all, the text that is often taken to state his
definitive position on the question, Posterior Analytics II. 19, claims that there
is a path that leads from perception to the first universal in the soul, and
from there to first principles, a path described as coming to know by induction
(APo. II. 19 100a3b4). And it appears this path is characterized in very similar
terms in the first chapter of the Metaphysics. More generally, as we will see, there
appear to be explicit proclamations of empiricist commitments in his works in
natural science.
There are, however, good reasons why many commentators on Aristotle
have resisted attributing this position to him. In this article, I review those
reasons and suggest that they stem from looking in the wrong place for
Aristotles views on inductive inquiry. Aristotle, as it turns out (and as he tells
us repeatedly), is a localist when it comes to scientific first principlesnot just
1. The full name of Aristotles works will be used the first time a work is mentioned; after that, the
following abbreviations will be used unless context demands otherwise: Posterior Analytics (APo.), Prior
Analytics (APr.), On Parts of Animals (PA), Historia Animalium (HA), De anima (De an.), Nicomachean
Ethics (EN ), Metaphysics (Metaph.), Meteorology (Mete.), and De caelo (Cael.). All translations are mine,
except where otherwise indicated.
2. While I am sympathetic to Ferejohns critical evaluations of the work he surveys, I am less so with
his central argument for distancing Aristotles project from that of the empiricists of the seventeenth and
eighteenth centuries. Ferejohn rightly sees their central concern as how legitimately to abstract universal
knowledge from sensory particulars, while he reads a notoriously mysterious sentence in APo. II. 19 as
evidence that, for Aristotle, this is not a concern since the content of perception is (already) universal. As
Ferejohn puts it, Aristotle is an immanent realist and thus need not concern himself about abstracting
universal content from particulars (2009, 71). I think there is ample evidence that Aristotle is not an
immanent realist in this sense (cf. Metaph. . 5 1071a1920, Z. 13 1038b812, 1038b3439a2) and
that he is as concerned about how one grasps universals on the basis of an experience of particulars, as
people in the seventeenth century were.
24

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

Lennox

l S P R I N G 2011

in the sense that each science has first principles peculiar to it but in the sense
that those principles will be discovered only by attending to facts that are specific to the domain that they govern. What one can say at the level of complete
generality about how one grounds the basic concepts, definitions, and causal
principles of a science is quite limited and provides very little guidance on how
one goes methodically from ones initial, unsystematic experience with a specific domain of the natural world to a systematic knowledge of that domain that
rests on true, immediate, causally primary principles.
There are two reasons for this localism. First, the objects in different domains are different, and since Aristotle is antireductionist, these differences
are nontrivial. It is his view that understanding those different objects will only
come from attending to what differentiates them from other things, as much as
from attending to what all natural objects have in common. Second, we stand
in different epistemic relationships to different kinds of objects. The difference
Aristotle remarks on most often is that between the eternal, but remote, beings
in the heaven and the mutable and perishable plants and animals all around
us, which are more like us and which, in any case, can be studied to our hearts
content. In On Parts of Animals I. 5, for example, he comments on the epistemological differences between these two branches of natural science as
follows:
Among the substantial beings constituted by nature, some are ungenerated
and imperishable through all eternity, while others partake of generation
and perishing. Yet it has turned out that our studies of the former, though
they are valuable and divine, are fewer (for as regards both those things on
the basis of which one would examine them and those things about them
which we long to know, the perceptual phenomena are altogether few).
We are, however, much better provided in relation to knowledge about
the perishable plants and animals, because we live among them. For anyone wishing to labor sufficiently can grasp many things about each
kind. Perishable beings take the prize with respect to scientific knowledge because we know more of them and we know them more fully.
(644b2230, 645a12)
With less eloquence, but to the same effect, he introduces his attempt to
account for the variations in direction of motion of the heavenly bodies in
De caelo as follows: Since circular movement is not opposed to circular movement, we must investigate on what account there are many motions, even if we
are attempting to make the inquiry from a great distancedistant not [merely]

25

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

in respect of location, but much more in respect of having perceptual awareness

of far too few of the attributes that belong to these things (Cael. II. 3 286a38).

2. Framing the Issue

A good place to begin is with one of the many passages in Aristotle that could be
cited to establish his credentials as an empiricist. Lack of experience [ ]
is a cause of the relative inability to comprehend the admitted facts. Wherefore
those who have dwelt more among natural things [ o o
o ] are better able to postulate principles of the sort that can connect
many things together; while those who, from engaging in many arguments, have
failed to study things as they are, readily show themselves capable of seeing very
little (Generation and Corruption I. 2 316a510). Note that this is not just
a description of how one acquires principles that can connect many things
togetherit is a passage shot through with normative language. Lack of experience accounts for a failure of comprehension; the more you dwell among
natural things, the better able you will be to identify appropriate principles. It
sounds as if Aristotle has some view that sufficient experience of the subject
matter for which you seek understanding is necessary to guide you to appropriate
first principles.
Nevertheless, for reasons that I will explore momentarily, scholars of Aristotle
who are perfectly familiar with this text see no such implication. Here are two
examples: To the extent that this [acquisition of concepts] is a natural process
based on perception, the relation between our perceptions and our knowledge
of first principles is a natural, a causal, rather than an epistemic relation. And
this is how Aristotle can be an extreme rationalist and still constantly insist on the
fundamental importance of perception for knowledge (Frede 1996, 172; italics
added). His conception of demonstration embodies a foundationalist conception of justification. The right sort of justification must avoid infinite regress
and vicious circle; and Aristotle can meet this requirement only if he recognizes self-evident first principles grasped by intuition (Irwin 1998, 134; italics
added).
On the basis of their readings of the Posterior Analytics, and especially its closing chapter on the path to first principles, Michael Frede depicts Aristotle as an
extreme rationalist, Terrance Irwin as an intuitionist.3 What leads them to these

3. Irwin will go on, of course, to argue that Aristotle abandons this position, to which earlier
commitments in APo. force him, and creates the discipline of First Philosophy to allow for a justification of first principles grounded in strong dialectic (see Irwin 1988, 14854); that is, Aristotle
trades in intuitionism for coherentism. I will not consider this move here since I agree with those
who do not think Aristotle is in any way committed to intuitionism about first principles in APo.
26

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

Lennox

l S P R I N G 2011

antiempiricist readings of Aristotles views about grounding first principles?

First of all, it is clear that each of these scholars approaches the question with
at least two unquestioned assumptions borrowed from modern epistemology:
(1) justification must follow discovery; the latter may be a suitable subject for
psychology or biography, but only the former is normative and a suitable topic
for epistemology, and (2) norms and standards are denizens of the space of reasons, not of the natural, law-governed world. From the standpoint of a person
wedded to these assumptions, any proposal that epistemic norms are built into the
inductive, principle discovering process of inquiry is ill conceived from the start.
Out of respect, perhaps, Frede refuses to attribute such a view to Aristotle, reading
passages that sound like such a view as characterizing a natural, a causal, rather
than an epistemic relation, and Irwin argues that Aristotle recognizes the error of
his intuitionist ways and invents metaphysics and strong dialectic as a way out.
The problem is not merely one of reading the Posterior Analytics through the
assumptions of analytic philosophy, however. The Analytics account of induction ( ), the way to first principles, is (and, I would argue, must be)
strikingly devoid of what any self-respecting modern inductivist is looking for.
In particular, it is devoid of one feature that, at least since Francis Bacon, everyone expects to be the centerpiece of a theory of induction: a set of norms or
standards to keep your inquiry on track and self-correcting. Aristotle is almost
obsessively concerned with the errors of his predecessors, and so he ought to
have been concerned to articulate such norms. But where, in APo. II. 19, for
example, does one find them? Norms for such things as proof, explanation, or
proper definition abound, but what about inquiry, the subject of the entire
second book? Even those who wish to paint Aristotle as an empiricist of some
sort make excuses for him on this point.
I shall argue that they need not do so, however. Both those who want to
depict him as a rationalist of one sort or another and those who depict him
as an empiricist have been looking in the wrong place for his views on norms
that need to be followed in order for an inquiry to be successful. A passage in the
Prior Analytics that is often cited, by those with empiricist leanings, to support
their case makes, I believe, a very different point. The majority of principles for
each science are distinctive []

to it. Consequently, it is for our experiences

concerning each subject to provide the principles. I mean, for instance, that it is
for astronomical experience to provide the principles for the science of astronomy (for when the appearances had been sufficiently grasped, in this way astronomical demonstrations were discovered; and it is also similar concerning
any other art or science whatsoever) (APr. I. 30 46a1727). This passage is
not, primarily, a general endorsement of empiricism about first principles
(although I think it implies that). Its primary messages, I believe, are (1) most
27

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

of the causal principles and definitions that inquiry seeks are specific to distinct
areas of knowledge, and (2) the norms and standards for searching for them
are thus, in an important sense, also specific. Very little can be said in the
abstract about this topic. The implication I have drawn from these messages
is that Aristotles views on the question of what standards and norms are
needed to ensure successful inquiry will be found in the methodologically
normative passages in his scientific works, especially, but not limited to,
the introductions to these works.
In the remainder of this article, I will build a preliminary case for this conclusion in two steps.4 First, I will explore Aristotles use of the concept of
oo in a number of key passages, focusing on those that are in the passages
that introduce an inquiry. That will lead rather naturally into a more detailed,
although still somewhat sketchy, examination of two such passages, where it is
quite clear that Aristotle is raising questions about what standards are appropriate for the investigation on which he is embarking and what norms of inquiry those standards call for.

3. Discourses on oo
Aristotle may actually have composed a discourse on method. Early in the Rhetoric,
in the context of explaining that rhetoric makes use of techniques that have their
counterparts in dialectic, he remarks that it is also apparent that each form of
rhetoric has its own good; for what has been said in the discourses on method
[ o oo (oo ?)], applies equally well heresome forms of rhetoric appeal to exemplars, others to persuasive arguments, and the same goes for
rhetoricians (Rhetoric I. 2 1356b203).
And in its catalog of Aristotles works, Diogenes Laertiuss Lives of the Philosophers (V. 23; Hicks 1925) refers to Mo
` . That this
might be an alternative way of referring to either the Topics or the Analytics is
unlikely since both works are explicitly referred to by those names just a few
lines before this passage. In any case, we can tell from the context the sort of
thing that was apparently discussed there: the use of exemplars and enthymeme in
rhetoric corresponds to the use of induction and deduction in dialectic. That is,
it sounds as if, if there were such a work, it discussed, among other things, the ways
in which different disciplines proceed by different although related methods
in this case, different although related methods of argument.

4. The full case will be made in a book on which I am currently working, for which this article can be
seen as an advertisement.
28

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

Lennox

l S P R I N G 2011

The word oo appears with remarkable frequency in the opening sentences of Aristotles major treatises, a fact that is rarely noted. Take the following
nonexhaustive set of examples (since the meaning and use of the term will be
under review, I will simply transliterate oo in these passages): knowledge,
and in particular scientific knowledge about every methodos (Physics I.
1 184a1011); regarding every study [theoria] and methodos, the more humble
and the more valuable alike (PA I. 1 639a12); every craft [techn ] and every
methodos, and likewise every action and decision [ proairesis], seems to aim at
some good (Nicomachean Ethics I. 1 1094a13); it remains for us to study
a part of the same methodos, which everyone prior to us has called meteorology
(Meteorology I. 1 338a256); the theme proposed for this work is to discover a
methodos by which we will be able to reason from accepted opinions about any
proposed problem (Topics I. 1 100a18); since nature is a source of motion and
change, and our methodos is about nature, we must not overlook the question,
What is motion? (Physics III. 1, 200b1213).
One reason for the failure to comment on this potentially revealing fact
about Aristotles opening paragraphs is suggested by the lack of agreement on
how oo is to be translated. Take the following sample of translations of the
opening words of Physics I. 1 184a10, for example. The relevant Greek phrase is
`

` o, which is variously translated: in all disciplines
(Charlton 1992), toutes les recherches (Pellegrin 2000), in any subject
(Waterfield 1996), in any department (Hardie and Gaye 1930), in every line
of inquiry (Irwin and Fine 1995, 83, with glossary note, 594), and in every
inquiry (Bolton 1991, 2). I have no doubt that a wider sampling would turn
up other options. There appears to be no general agreement among translators
about what Aristotle has in mind by the term in any particular application. Moreover, as I will explore in some detail, Aristotle deploys the concept in two very
different ways. In one, it feels natural to offer transliteration as translation, for
Aristotle seems to be referring to a method, a way of proceeding (which is unsurprising since the root of the Greek word is o , which refers primarily to a
path or route). This seems to be the sense carried by the term at the beginning of
the Topics, quoted above. But the term is also used with some frequency in a way
that suggests reference, not to the method by which a human activity is done but
to the activity itself. Most of the translations of the opening phrase in Physics I. 1,
for example, take it that way, although they differ about whether it refers specifically
to inquiry or research or to a field of knowledge more generally. Without begging
any questions, I am going to proceed on the hypothesis that there is a significant
difference between these two uses and explore two texts, each of which seems to
be a rich and self-conscious discussion about the nature of an inquiry in which
Aristotle is engaged but which uses oo in two somewhat different ways.
29

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

4. How to Proceed with Animals

Aristotles Meteorology opens with a review of what the course of study of the
natural world has covered so far (338a2025): the primary causes of nature, all
natural motion, the movement of the ordered system of stars, the elements
of bodies, how many there are and what sort, their changes into one another,
and generation and decay of the most common kind. He is now about to begin
an investigation of a part of the same oo, which all previous investigators have called meteorology (338a2526). After outlining what is involved in that investigation, he goes on: Having dealt with these subjects,
we will study whether we are somehow able, according to the recommended
manner [
` o o],5 to give an account of animals and
plants, both in general and separately. For having spoken of these things we
would pretty much have reached the goal of our original plan in its entirety
(339a59).
As I have argued elsewhere (Lennox 2010), this question about whether we
are able to give an account of animals and plants according to the same method
adopted for the previously recounted branches of natural science is a genuine
concern for Aristotle. Moreover, it is not primarily a question about how to
study animals and plants; rather, it is a metalevel question about whether the
methods used elsewhere in the study of nature suffice for living nature. It is, I
believe, the question that lies behind the need for PA I.
PA I, which is generally recognized as a philosophical introduction to the
study of animals, begins with the following words:
` ` oo, o o o `
, o `o o , ` `
o
o oo, ` o o

. (1 639a15)
[Regarding every study and every investigation, the more humble and
more valuable alike, there appear to be two sorts of state, one of which
may properly be called understanding of the subject matter, the other a
certain sort of educatedness.]
PA I is a guide to acquiring that second state, this special form of paideia, which
provides one with the ability to make critical judgments about what is well or
5. Compare ` ` oo (Politics I. 1 1252a1718). These two phrases
are very close in meaning.
30

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

Lennox

l S P R I N G 2011

poorly expressed.6 Aristotle goes on to distinguish one who has a very general
form of this skill and one who has it about a specific discipline, say, the art of
medicine (as in the passage from the Politics quoted in n. 6) or natural science.
Thus, at the end of this paragraph, he makes a transition to the business at hand
with the following words: So it is clear, for natural inquiry too [`
o], that there is need of some such standards [oo] (639a12
13). The ability to make such judgments requires certain standards, and if
ones paideia is about a specific field, then it will be standards appropriate
to that field that one needs to acquire. This will be the topic of discussion
for the five chapters of PA I.
Let us now return briefly to the appearance of the concept of oo in the
first sentence of this book. I want to draw attention to the fact that and
oo are conjoined not simply by and () but by both and ( ),
which makes it quite clear that these terms are conveying different ideas and
are not merely synonyms. It could be that Aristotle has two different categories
of cognitive endeavor in mind. However, it could also be, and I want to suggest
that it is the case, that he wants to make reference to two different aspects of a
cognitive endeavor, aspects that are picked up when he identifies two different
states associated with themscientific knowledge ( ) and that general
critical judgment he identifies as a certain sort of paideia. The evidence that I find
compelling for this reading comes from the last paragraph of chapter 4, which is
a summary of what has been accomplished. We have said, then, how the investigation of nature [` ` oo] should be appraised, and in
what way the study [ ] of these things might proceed on course [o ]
and with greatest ease. Further, about division [` ] we have said
in what way [ o] it is possible by pursuing it [o ] to grasp things
in a useful manner, and why dichotomy is in a way impossible and in a way
vacuous (PA I. 4 644b1722).
The repetition of oo and here is almost certainly intended as a
conscious echo of the opening five words of PA I. 1, and here one can see that
the two words are aimed at emphasizing different aspects of a single study. Provisionally, I want to suggest that to refer to a oo of X is to emphasize the
way in which the investigation should be carried out, while to refer to a

6. Compare the following comment: Hence just as a court of physicians must judge the work of a
physician, so also all other practitioners ought to be called to account before their fellows. But physician
means both the ordinary practitioner, and the master of the craft, and thirdly, the one who is generally
educated [o o] about the craft (for in almost all the arts there are some such people, and we
assign the right of judgment [ ] just as much to the generally educated [o o] as
to those with knowledge) (Politics III. 6 1282a1). I suppose this third class corresponds to a philosopher
of science or medicine.
31

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

of X is to emphasize that the investigation is directed on a particular object

or class of objects. A proper study of X requires that one proceed in a proper
mannerthe use of o , in close proximity to its derivative oo, provides further support for this thought.7 What Aristotle has been discussing
for the previous four chapters are the norms that must guide animal investigation if it is to stay on course or on track toward the goal of scientific
knowledge.
This passage, then, suggests a hypothesis about the connection between
these two uses of the term. To ask, as Mete. I. 1 does, Is X part of the same
oo as Y? is to ask whether the same norms or standards of inquiry are
appropriate for X and for Y. That is, it is to ask whether the same oo is to
be used. If so, then you are licensed to refer to them as instances of the same
oo. When the term is used in this way, its rules for application will be
similar to those for inquiry, investigation, and discipline. However, it will
pick out tokens of a type of investigation by attending to the norms and standards that are in play, rather than to the object of investigation.
To make this hypothesis plausible, we shall now turn to the opening questions of PA I. 1, with an eye to what Aristotle is trying to accomplish. In the
following section, we will turn to Aristotles methodological introduction to an
investigation into the soul at the beginning of the De anima. What will be clear
from doing so is that the norms being discussed there, and even the way they are
being discussed, are very different and entirely appropriate.
During the introduction to PA I. 1, Aristotle discusses the special skill of
critical judgment we need to acquire, in a way that suggests it will be primarily
of use in judging what others have said or written. However, as he turns to the
questions that need to be considered in acquiring this critical ability, as we saw,
he refers to certain standards that need to be in place in the inquiry into nature,
strongly suggesting that these are standards that ought also be referred to by one
engaged in an ongoing investigation, and when the first question is taken up,
that suggestion is confirmed. Wherefore how we ought to carry out the investigation [ o] should not be overlooked; I mean, whether
one should study things in common according to kind first, and then later their
distinctive characteristics, or whether one should study them one by one straight

7. The noun o refers to a road, path, or track and is used metaphorically in much the same
way as those English expressions are. The dative form used here often has adverbial force, conveying the
idea of staying on the road to your destination, thus my on course. M oo is formed from that noun
and a prepositional prefix,
, which when used as a prefix carries the sense of going after and in
quest of . The basic idea, then, is a path taken in quest or pursuit of something. It is already used in
Platos Sophist (218d, 235c, 243d) to refer to an inquiry, and Republic VII. 533c refers to the dialectical
method ( ` oo) as the only way to advance to first principles.
32

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

Lennox

l S P R I N G 2011

away (639b45; emphasis added). I have highlighted three features of importance in this formulation:
1. The question is about how one ought to investigate: that is, it is a normative question about inquiry.
2. This is reflected in an explicit question about the proper order of investigation. That is, Aristotle appears to be concerned that if one does not
investigate things in the correct order, the investigation will inevitably go
off track.
3. To be specific, Aristotle begins this discussion of proper method for investigating animals by raising a question about the level of generality that
should be targeted at the beginning of such an inquiry (639a1619):
should we study individual species one by one or begin with attributes
possessed in common by many kinds of animal? He has now converted
this question into an interesting alternative: (a) pursue a two-tiered investigation, in which one begins by investigating features that belong in
common to kinds, and then later moves to investigating characteristics
that are distinctive to specific forms of those kinds, or (b) begin investigation immediately with specific forms, studying each of them one by one.
These alternatives reflect the results of the intervening discussion (639a25b4) in
which Aristotle articulates a distinction between two different kinds of commonly
possessed features: (1) commonly possessed undifferentiated featureshe gives as
examples sleep and respirationand (2) features that belong in common to many
animals but are differentiated according to form, a distinction you are far more
likely to notice if you begin at the more general level. Here, he gives as an example
one that is directly reflected in one of his own studies, De incessu animalium.
Locomotion belongs in common to a wide range of animals; however, some of
them fly, some walk, some swim, and some crawl.
Although he spends a great deal of time introducing and then fine-tuning
the question here, he concludes by noting, with regret, that neither it nor the
question to which he is about to turn has been adequately answered. He does,
however, reintroduce this first question in chapter 4, after he has defended a
new method of division that grows out of scathing demolition of Platonic dichotomous division in chapters 23. That new method is designed to identify
levels of differentiation of multiple, coextensive features that belong in common
to general groups of animals (as, e.g., feathers and beaks are common to all birds),
groups he will elsewhere refer to as great kinds (megista gen). But such a method
of multiple differentiation presupposes that kinds constituted of such coextensive
features can be identified, and chapter 4 lays out a set of standards to be used in
33

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

identifying such kinds at this general level. This is not the place to go into details
about these standards,8 but the narrative structure of PA I explains why Aristotle
introduces the question early in chapter 1 but does not fully answer it until
much later. It is a question that can receive a proper answer only after a proper
method for tracking divisions of multiple features that belong in common according to kind has been described and defended.
There are multiple examples in the Historia Animalium of Aristotle displaying this method at work. I have chosen the opening of his discussion of the
cephalopods, in the first chapter of book IV.
Among the animals called soft-bodies these are the external parts: 1. the
so-called feet; 2. the head, continuous with the feet; 3. the sac, containing
the internal organs, which some mistakenly call the head; 4. the fin,
which encircles the sac. In all of the soft-bodies the head turns out to be
between the feet and the belly. Moreover, all have eight feet, and all have
two rows of suckers, except for one kind of octopus.9 The cuttlefish, and
the large and small calamary have a distinctive feature, two long tentacles,
the ends of which are rough with two rows of suckers, by which they capture food and convey it to their mouth and fasten themselves to a rock
when it storms, like an anchor.10 (523b2133; cf. PA IV. 9 685a33b2)
Here we see Aristotle capturing external anatomy at the level general to the kind
first and then gradually moving to features that are peculiar to subkinds. This
framework governs the rest of the discussion, which goes on for two more
Bekker pages, concluding with a discussion of the many kinds of octopuses,
itself a subkind of the cephalopods (
`
; lit. the softies).
This is, then, a methodological norm that is introduced in the form of a
question about how to proceed with the investigation, which is gradually articulated and defended in PA I, and is consistently implemented in the presentation
of the results of his empirical investigation of animals in HA. Moreover, it
8. I have discussed these in Lennox (2005, 87100); a different, more theory-laden reading of the
same text can be found in Charles (2000, 31216).
9. Eledone cirrhosa, the lesser octopus.
10. Aristotles account and those found in contemporary textbooks are very similar: Their [cuttlefish
and calamari] preferred diet is crabs or fish, and when it is close enough it opens apart its eight arms and out
shoots two deceptively long feeding tentacles. On the end of each is a pad covered in suckers that grasp hold
of the prey (Dunlop 2003). One principle difference is that Aristotle consistently mentions two functions
for these tentacles, feeding and mooring during storms, while modern texts I have consulted only mention
feeding. One audience member, who attended a lecture in which I mentioned this discrepancy, claimed to
have repeatedly observed the mooring behavior described by Aristotle, while snorkeling in the
Mediterranean.
34

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

Lennox

l S P R I N G 2011

appears to depend on another set of standards for (a) identifying kinds that
share a large number of features in common and (b) producing divisions by
means of noting differentiations of those general features within the many
forms of those kinds. For example (concretely), cephalopods are a good entrylevel kind because they all share an overall distinctive bodily organization (head,
eight legs, visceral sac, circular fin, etc.), many of those features are found to be
differentiated in distinct ways in different subkinds (cuttlefish, squid, octopuses),
and these differentiations can be tracked and correlated by the multiaxial method
of division characterized in PA I. 23 (e.g., tentacles with one vs. two rows of
suckers). Moreover, these are norms that arise out of extensive experience with
the animal world and that are designed with an investigation of that world specifically in mind.11
APr. I. 30, which I quoted above, asserts that the distinctive principles of a
domain arise from the specific experiences of the objects of that domain and uses
astronomy as its example. It is for astronomical experience to provide the principles for the science of astronomy (for when the appearances had been sufficiently
grasped, in this way astronomical demonstrations were discovered; and it is also
similar concerning any other art or science whatsoever) (46a 2327). At that
level of generality, perhaps the discovery of principles is similar in any other art
or science. But in PA I. 1, Aristotles next question centers on this very issue:
Whether, just as the mathematicians explain the phenomena that concern astronomy, so too the investigator of nature [ o], having first studied the
phenomena regarding animals and the parts of each, should then state the reason why and the causes, or whether he should proceed in some other way
(639b710). Now one might think that if anything is an axiom of Aristotles
philosophy of science it is that you must first study the phenomena before going
on to determine the reason why and the causesWhat other way could an inductivist proceed? The investigation of animals and their parts is, however,

11. An extremely important issue that I cannot take the space to address here, but which I want to flag,
turns on distinguishing between Aristotles actual methods of empirical investigation and his written reports
of the results of those investigations. When historians of science investigate many post-Renaissance natural
philosophers/scientists, they have access to such things as field notes, letters, laboratory notebooks,
autobiographical accounts of investigations, and so on. This allows a principled distinction between the
published book or journal article and the day-to-day struggle of collecting information, analyzing it, interpreting it, and integrating new findings with old that is often suppressed in the publications that result.
Not only do we not have these two different kinds of documents to compare in the case of Aristotle, but
we are not even entirely sure whether what we have is more like a published book or more like an investigators notes of an ongoing investigation. This poses an intriguing metalevel problem for inquiry like that
in which I am currently engagedIs it possible to distinguish in Aristotles texts norms for the presentation
of the results of an inquiry from norms for carrying out an empirical inquiry? This footnote is intended to
signal only an awareness of the problem, not a solution to it.
35

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

complicated and very different from that undertaken by the mathematical astronomer. Although it may not be immediately obvious, it turns out there are
two normative questions embedded in this passage: (1) Should the natural
scientist proceed in the same way as the astronomer? (2) Should the natural
scientist, in studying animals, proceed from a study of animals and their parts
to a study of their causes?12 Why does Aristotle raise these concerns here? To
answer that question, it helps to remind ourselves of Aristotles views about the
objects that mathematical astronomy studies and our cognitive access to them.
The objects of astronomy are the heavenly bodies, and Aristotle argues, in
De caelo, that they are eternal. The only change they engage in is movement
in place, and it is because they move in changeless, circular patterns that the
mathematician can provide explanations of their motions. We can be sure of
very little about their material constitution or the physical causes of their motions since our access to them is extremely limited.
It is thus not surprising that the philosophical discussion of PA I is structured
around a distinction between two sorts of natural beings: those that are eternal
and governed in their movements by unconditional necessity and those that
come to be and pass away and are governed in their changes by conditional
necessity (639b2127, 640a12, 644b22645a11). Animals are a special class
of natural substances that come into being in a particularly complex and yet
coordinated manner that is evidently goal directed. Moreover, there are a vast
number of different animals, and the number of ways in which they differ is
even more vast. Among their differences is a vast range of coordinated activities
that constitute their distinctive ways of life.
Thus, while the prescription to study the phenomena first and then the causes
may be unproblematic for mathematical astronomy, it is highly problematic for
a norm-governed inquiry concerning the animals ( ` oo;
On Length and Shortness of Life 6. 467b79). To cite one class of problems on
which we will focus shortly, should one study the phenomena related to the process
of coming to be as well as the phenomena related to the anatomy, physiology, and
behavior of fully developed animals? Might the process of coming to be actually be
the cause of the fully developed animal and its features? If so, can we really distinguish a noncausal from a causal stage of investigation in the case of animals?
These concerns, I would suggest, lie behind the fact that this question about
the proper way to order the investigation of animals, like the first question

12. Note that, stated in this way, what proceeding in this manner amounts to will be very different,
depending on whether you think that the material and generative conditions are the only or primary
causes of animals and their parts or you think that the form of the completed animal determines the
nature of its parts and the order in which they must come to be. I believe it is Aristotles sensitivity to this
difference that leads him to delay answering it.
36

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

Lennox

l S P R I N G 2011

about whether to begin with an investigation of specific kinds or features common to many kinds, is not immediately answered. Before this second question
can be answered, the complexity of causal explanation in the study of animals
needs to be addressed. Thus, after asking, and not answering, this question
about whether the zoologists investigations should be patterned on those of
the mathematical astronomer, Aristotle asks another question about the order
of investigation. Since we see more than one cause of natural generation, e.g.,
both the cause for the sake of which and the cause whence comes the origin of
motion, we need also to determine, about these causes, which is naturally first
and which second (PA I. 1 639b1114). Unlike the first two questions, however,
Aristotle turns immediately to the task of answering this one. From 639b1521,
he defends the causal priority of the cause for the sake of which (the inappropriately named final cause) in things composed by nature and by art. In these
cases, he insists, the nature of the process of coming to be is determined by the
goal toward which it is directed, not the other way around. This can be seen in
the fact that it is the knowledge of the building that is to be built that dictates to
the builder the materials to be used and the precise order of the steps in the
building process. He concludes by stressing that that for the sake of which
and the good ( ) are present more in the works of nature than in those
of art. He then articulates and defends a distinction between unqualified necessity and conditional necessity that is implicit in his account of the priority of the
final over the efficient cause. In things that come to be, whether in art or in
nature, what is to be, the end of the developmental process, necessitates that
certain materials and motive causes be present. If a certain house is to be, bricks,
mortar, and timber must be present and must be acted on in specific ways by
builders. If a frog is to come to be, the female must provide an egg with the
appropriate developmental capacities, and the male must convey the appropriate species-specific nutritive/generative heat to that egg. Aristotles central point
is that the necessities are imposed by the goal on the material and efficient
causesthe material and efficient causes do not necessitate the production
of the goal.13 With these key ideas in place, he then argues, from 640a19,
for a distinct manner of demonstration for contexts in which conditional necessity is operative and where the starting points and definitions will identify
the goals of the processes that are productive of those goals.14
13. Compare Physics II. 9 200a3134: Plainly, then, the necessary in things which are natural is
that which is given as the matter, and the changes it undergoes. The student of nature should state both
causes, but particularly the cause which is what the thing is for; for that is responsible for the matter,
whilst the matter is not responsible for the end (Charlton 1992).
14. Again, it is important to read this passage together with Physics II. 9 200a15200b8, where
Aristotle contrasts such teleological demonstrations with those in geometry.
37

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

And, finally, with the machinery of teleology integrated with conditional

necessity, Aristotle has prepared the ground for answering his second question.
But he proceeds to the answer in what appears, at least if read without that
preparation in mind, in a puzzling manner. He reminds us that those before
him (he cites Empedocles in particular) have sought to understand the way things
are, by investigating how they came to be that way, without asking whether that
is the appropriate way to proceed. Then, as if it were a response to such people,
he says: It seems we should begin, even with generation, precisely as we said
before: first one should get hold of the phenomena concerning each kind, and
then state their causes (PA I. 1 640a1214). And as if it explicates this claim
about proceeding from grasping the facts to asserting their causes, he immediately goes on: For even with house-building, it is rather that these things [i.e.,
what happens during coming to be] happen because the form of the house is
such as it is, than that the house is such as it is because it comes to be in this way.
For generation is for the sake of being; being is not for the sake of generation
(640a1519). But how does this reminder of the conditionally necessary nature
of goal-directed processes help us to understand that in studying generation,
too, we should first grasp the phenomena and then search for their causes? This
appears to be a non sequitur.
The solution to this puzzle is to place the argument in its context. What
stands between Aristotles statement of the question and its answer is his insistence that natural generation has more than one cause, that goal causation has
priority over motive causation, and that the necessity of the materials and motive causes in generation is conditional: they do not necessitate the goal, but
they are necessary conditions for its achievement.15 Given that context, then,
it is natural that he is, in the passage we are currently investigating, criticizing
his predecessors for explaining the attributes of the developed animal as a necessitated outcome of the motive causes that produced them and insisting that
coming to be happens as it does because of the goal toward which it is directed.
And this understanding of causal priority will only be achieved by doing a preliminary historia, of the sort that is reported on in the Historia Animalium,
which after four books systematically surveying the internal and external anatomy
of all the major animal kinds, spends three of the remaining books surveying the
facts about generation.16 Thus, after reviewing what Empedocles had missed by
assuming that one accounts for (e.g.) the segmented backbone of vertebrates by

15. There is a considerable literature around the question of the nature of this priority. For a recent
survey of the issues at stake that takes a somewhat different view from that defended here, see Leunissen
(2009, 99108).
16. For an example, see the discussion of HA IV. 1 above.
38

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

Lennox

l S P R I N G 2011

explaining how they happen to twist as they come to be (640a1922), Aristotle

summarizes what he takes to be the appropriate method.
Hence it would be best to state that, since this is what it is to be a human being, on account of this it has these [features]; for it cannot be [a
human being] without these parts. If one cannot say this, one should say
the next best thing, that is, either that in general it cannot be otherwise, or
that it is at least good [that it is] thus. And certain other things follow.
And since it is such [as it is], its generation necessarily happens in this way
and is such as it is. (This is why this part comes to be first, then that one.)
(PA I. 1 640a33640b2)
Notice that in this passage, the causal relationships among the features of the
actual members of a kind must be understood in order to explain the patterns
that one sees in generation. Herein lies the complexity in studying animals and
their generation that prevented Aristotle from simply affirming that inquiry
should proceed here in exactly the same manner as in mathematical astronomy.
One actually needs to have a causal understanding of the facts about the actual
animals one is studying in order to move on to a causal understanding of generation. Without that, you will not understand the order in which parts unfold
in development at all.
As with our discussion of question 1, it is important to see that these methodological norms reflect, or are reflected in, his practice. In Generation of
Animals II. 6, having presented his theory of the causal contributions of male
and female to sexual generation, Aristotle is ready to explain the formation of
the parts in development. As is so often true at such transitional points in his
scientific writing, however, he opens by worrying over the best way to proceed.
Thus, as many of the instrumental parts as are generative in nature must
be present first (for such [parts] are for the sake of others as their origin),
while as many [parts] as are among things for the sake of another but not
generative in nature must be later. Wherefore, it is not easy to distinguish
which among the parts that are for the sake of another are prior, or what it
is these parts are for the sake of. The motive parts, being before the end in
generation, intervene, and distinguishing between the motive and the instrumental parts is not easy. And yet, it is necessary to inquire what comes
to be after what according to this method [
` ` oo

]; for the end is posterior to some things, but prior to others.

(742b311)

39

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

Because biological generation is goal directed, the order in which the parts
appear is dictated by teleologically established priorities; if an instrumental part
(e.g., the heart) plays a crucial role throughout the generative process, it must
come to be early on; if some part has a teleological role to play only in the
completed animal, it will develop later. One cannot simply read off causation from the temporal sequences that one observes taking place during
development.17
It is now time to step back from these first three questions, which set the
agenda for the rest of PA I, and reflect on the way in which they establish a
set of norms of inquiry for the study of animalsnorms that, while fully consistent with Aristotles theory of inquiry in APo. II, are specific to zoological
inquiry. First, note that these norms are established by means of asking a set
of domain-specific, normative questions. By calling them normative, I have
in mind that they are questions about how an inquirer, or inquiry, should proceed. They assume an inquiry has alternativesand indeed, Aristotle often has
in mind ways in which his predecessors have proceeded that he thinks are inappropriate in various ways and have inhibited progress. This is what I will refer
to as establishing an erotetic framework for the ensuing inquiry. In the next
section, we will see precisely the same feature in the methodologically oriented
first chapter of De anima. Second, it is also noteworthy that the three questions
are, among other things, about the order of inquiry. Should we start by grasping
attributes common to many kinds and then proceed to the more specific;
should we, in investigating animals, study the phenomena before inquiring into
their causes, as the astronomers do; should we study the actual animal and its
parts before studying its development? Finally, question 1 concerns inquiry into
the defining natures of things, while questions 2 and 3 concern the order of
inquiry aimed at causal understanding. As I noted, these two themes dominate
PA I. Of equal interest is that it is if it is/what it is inquiries aimed at definition
and that is is/why it is inquiries aimed at causal explanation that Aristotle seeks
to integrate in APo. II. The overarching picture of the epistemic quest has not
changed, but how to keep a quest on track toward understanding requires
norms that are largely domain determined. In this case, we are dealing with
17. As an aside that underscores Aristotles point here, William Harvey takes Aristotle to task in
his Exercitationes de Generatione Animalium for thinking that the first appearance of a pulsing bloody
spot in the developing chick embryo is the heart. For other reasons, Harvey believes that the blood
carries the causal agency guiding generation and forms before the heart and then produces ithe thus
argues that what Aristotle was actually observing was blood pulsing with the power of life. Aristotle was
in fact correct. More important, Aristotle thought the blood was instrumental in distributing the generative capacity throughout the organism, while the heart was truly generative. On this disagreement,
see Aristotle, HA VI. 3 561a415; William Harvey, Exercitationes de Generatione Animalium, 96 ex. 17,
241 ex. 51 (Whitteridge 1981).
40

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

Lennox

l S P R I N G 2011

distinctively complex, dynamic, and multilayered investigation of animals in all

their vast variety and of their development. In order to know how to keep such
an inquiry on track, it is Aristotles view that one must consider the specific
nature of the object of inquiry and the nature of our cognitive access to that
object.

5. How to Investigate the Soul

Famously, the De anima opens with praise for the inquiry into the soul ( `
o ). Knowledge of beautiful and honorable things can be
ranked normatively in terms of either their accuracy or the intrinsic value of their
objects. On either count, Aristotle tells us, it is with good reason that the inquiry
into the soul is placed in the first rank. He claims that such an inquiry contributes
greatly to all truth, but especially truth related to nature.
The introduction to the De anima continues, also famously, by warning the
reader that the investigation on which we are embarking is in every way and in
all respects the most difficult about which to get hold of any secure belief
(402a1011). How one reads the rest of this chapter turns on how one interprets what Aristotle says about the character of this difficulty. The difficult nature of the task turns not so much on the fact that we are investigating the soul
as on the fact that we are investigating its substantial being and essence, and
there is a question about the proper method for pursuing such an inquiry. Is the

way forward, he asks, demonstration, division or some other method (

oo; 402a1920)? It is worth recalling, as one reads this, the dialectical
development of APo. II. 110.18 Recall that Aristotle is, there, asking about
how inquiries seeking definition (i.e., knowledge of what things are) are related
to inquiries seeking causes (i.e., knowledge of the reason why). The options on
the table for proceeding from existence to essence are demonstration, which he
discusses and rejects in chapters 3 and 7, and division, which he has already
ruled out in APr. I. 31 as a method of proof and which he rejects as a method
of inquiry leading to proper definition in APo. II. 5.19 From chapter 8 on, he
articulates another method for this inquiry, described, however, at such a level
of abstraction that each domain of study will have to enrich its recommendations by means of a domain-specific set of norms.
And that accounts for the next step in De an. I. 1. Aristotle reminds us that
while we might not need to seek a new method of study for each new investigation
18. This in turn has an Academic background since division, in dialogues such as the Phaedrus, Sophist,
and Statesman, is viewed as the primary tool for inquiry into being and essence.
19. Although in chap. 13, he acknowledges that it can be helpful in hunting for the essence.
41

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

of being and essence, we will still need to seek unique starting points: For there
are different starting points for different things, as in the case of numbers and of
planes (402a2122). Once again, the problem he is facing derives from the fact
that different investigations will have distinct first principles.
All of this reminds us, then, that we are in the philosophical milieu framed
by the Posterior Analytics.20 In any domain of knowledge, we are there told, the
necessary attributes of the objects in that domain are to be demonstrated from a
set of starting points that are (on pain of circularity or regress) indemonstrable
(APo. I. 3 72b1924, I. 6 75a2937; cf. EN VI. 5 1139b2932). Among these
are principles that identify the being and essence of the primary objects in each
domain (APo. II. 10 94a1012, II. 13 96b714). But, since these are not themselves demonstrated, the question arises, how does one investigate and discover
these principles (APo. I. 2 71b1719, I. 3 72b2324, II. 13 96a2023)? This
is precisely the issue being identified as problematic, on the opening page of
De anima, for however we understand the claim, Aristotles baseline assumption
is that to investigate the soul is to investigate some sort of principle ( `) of
living things (402a67).
The next stretch of text lists a number of questions that the person seeking
knowledge of the soul must consider. Here, as in the opening pages of PA I,
Aristotle is establishing the erotetic shape of the investigation by asking a set
of normative questions that the inquiry must answer. I will here only have space
to indicate two of the questions and the striking way that they structure his
positive account of the soul in book II. The primary purpose of doing so is
not to explore the questions and answers in detail but to indicate how different
they are from the questions that shape the inquiry into animals in PA I. 1.

Question 1
First of all, perhaps, it is necessary to determine in which of the kinds [the soul
is found], i.e. what it is; I mean, whether soul is a this and a substantial being
[ ` o ], a quality, a quantity, or even some other of the categories
that have been marked off (PA I. 1 402a223).21 We need to begin by identifying at the most general level what the soul is, which means determining its

20. A point properly stressed in Sisko (1999).

21. This may seem an odd question. However, as one reads through Aristotles careful critique of
previous accounts of the soul in De an. I, it becomes clear that, apart from a number of materialist
theories, the other alternatives are that the soul is a number (404b30), ratio, or harmony of some
sort (407b28ff.; i.e., in the category of quantity) or that it is movement of some sort (405a10ff.). In his
context, determining how to categorize the soul is an important first step.
42

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

Lennox

l S P R I N G 2011

categorical statusprecisely the question he turns to answer at the beginning of

his most common account of the soul, the opening paragraph of De anima:
Now since we call one kind of existent substantial being [o ], spoken of in
one way as matter, which in itself is not a this, another as shape and form, in
virtue of which it already is a this [ ], and third that composed of these
(De an. II. 1 412a69).
That is, he begins by identifying the category in which the soul is found but
then notes that there are a number of senses in which something can be in the
category of substantial being. He argues against seeing the soul as either the
matter or the compound and concludes: The soul must, then, be substantial being
qua form of a natural body that has life potentially (De an. II. 1 412a1921).
That is, the very first positive step in providing his own account of the soul, in
opposition to the views he has reviewed and rejected in book I, is to answer the
first, categorical, question presented in his methodological introduction. And,
by identifying it as belonging in the category of substantial being in the sense
of form, he has affirmed that it is substance in the sense of that which makes
a substance a this, a .

Question 2
That initial account identifies soul as the form of a body that has life potentially.
If the body has life potentially, what is the modal status of soul? Returning to the
list of questions in book I, chapter 1, the second question Aristotle says we must
answer is precisely that: And further, is soul in potentiality or rather some kind
of actuality [
]? (De an. I. 1 402a256). There are two features
of note in Aristotles formulation of this question that indicate the care with which
he is framing the investigation. First, while Aristotle has two terms that often overlap in reference and are often both translated as actuality and
the one used here is his own coinage, and, while often
simply refers to activity or movement, connotes a state of completion or full realization. The contrast here is, then, between potentiality and its
realization. The second feature of note is the qualification on this term, .
Aristotles next move in providing a common account of the soul at the beginning of book II is to assert that the soul is a special kind of actuality.
Picking up from where we left off, then, Aristotle continues: And matter is
potentiality, while form is actuality [ ]and that in two ways, first as
knowledge is, and second as studying is (De an. II. 1 412a1011). The framing
question in De an. I. 1 thus leaves open not only the question of whether soul is
to be understood as a potency or an actuality of the body but also the precise

43

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

way in which it is an actuality, if that is what it turns out to be. The relevance of
this is that the soul is a first actuality, which, viewed in relationship to the body
of the animal is its realizationthat which makes it a living thing and not
merely a bodywhile viewed in relationship to the animals living activities,
it is the capacity for those activities. Those phenomena associated with soul that
might incline one to classify it on the potential side of the potential/actual distinction are captured by Aristotles distinction between two ways of being actual
or complete. A person who has the expertise to repair an internal combustion
engine is, in a perfectly legitimate sense of the term, an actual auto mechanic
(e.g., compared to me), but that person actually repairing an engine is, in an
equally legitimate sense, fully actualizing his potential as an auto mechanic. All
of this is captured in the conclusion of this first attempt to provide a common
definition of the soul, which provides Aristotles answers to questions 1 and 2:
But substantial being [o ] is actuality [ ]. Soul, then, will be the
actuality of a body of this kind. But actuality is spoken of in two ways, first as
knowledge and second as study. It is clear then that the soul is actuality as knowledge is (412a2021). This idea of the soul as the living bodys capacity for a
kind of activity accounts for a vast range of facts about ensouled beings, such
as that they remain fully capable of the rapid mobilization of coordinated actions
needed to, say, successfully elude a predator while they are at rest and doing no
such thing. A living thing in a state of rest is neither merely a body capable of
life nor fully engaged in living, and Aristotles first actuality captures this important and fundamental fact about life as it had not be captured previously.

6. Conclusion
This essay began by claiming that those who have cast doubt on Aristotles
claims for an inductive path to scientific first principles have made their case
in part by importing relatively recent ideas about epistemic justification into
an alien context. I have not directly made the case that those standards are,
in fact, alien, although their post-Kantian, and more precisely Fregean, origins
make it a plausible claim. But I also claimed at the outset that part of the initial
plausibility of their argument rested on the absence of any obvious norms of
inductive inquiry in those texts scholars have traditionally turned to in search
of such norms, in particular, the Posterior Analytics II. In the past, I have insisted
that there is more there than people have thought, but I agree that there is not
enough. That is, I have here argued, because for Aristotle most of the first principles on which a science is founded are not only specific to that science but
must be searched for through a rich, empirical engagement with the phenomena distinctive to that science. Thus, his most interesting discussions about how
44

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

Lennox

l S P R I N G 2011

inquiries aimed at the formulation of first principles are to be guided and kept
on track are to be found in the methodological introductions to his scientific
works themselves. Here I have only been able to give the briefest sketch of a case
for this conclusion. I have done that, first, by looking at the centrality of the
concept of oo in those introductions and, then, by looking at two such
methodological introductions, one to his study of animals, the other to his inquiry into the soul.22 PA I. 1 and De an. I. 1 share a philosophical outlook, and
both operate by the identification of questions that need to be answered in order
to get their respective inquiries off on the right track. Precisely because they have
this much in common, it is revelatory how very different they areespecially
because one would expect these two inquiries, into the soul and into ensouled
beings, as it were, to have much in common. It is in discussions like these, if
anywhere, that we will find Aristotles norms of inductive inquiry.
22. The term
oo is not used in the Posterior Analytics.

REFERENCES
Bolton, Robert. 1991. Aristotles Method in Natural Science: Physics I. In Aristotles Physics:
A Collection of Essays, ed. Lindsay Judson, 129. Oxford: Clarendon.
Charles, David. 2000. Aristotle on Meaning and Essence. Oxford: Clarendon.
Charlton, William. 1992. Aristotle: Physics Books I and II. Rev. ed. Oxford: Clarendon.
Dunlop, Colin. 2003. Cuttlefish Basics. Deep Intuition. http://www.tonmo.com/articles/
basiccuttlefish.php.
Ferejohn, Michael. 2009. Empiricism and the First Principles of Aristotles Science. In A
Companion to Aristotle, ed. Georgios Anagnostopoulos, 6680. Oxford: Wiley-Blackwell.
Frede, Michael. 1996. Aristotles Rationalism. In Rationality in Greek Thought, ed. Michael
Frede and Gisela Striker, 15775. Oxford: Clarendon.
Hardie, R. P., and R. K. Gaye. 1930. Physica. In The Works of Aristotle Translated into
English, ed. W. D. Ross. Vol. 2. Oxford: Clarendon.
Hicks, R. D., trans. 1925. Diogenes Laertius: Lives of Eminent Philosophers. Cambridge, MA:
Harvard University Press.
Irwin, T. H. 1988. Aristotles First Principles. Oxford: Clarendon.
Irwin, Terence, and Gail Fine, eds. 1995. Aristotle: Selections. Indianapolis: Hackett.
Lennox, James G. 2005. Getting a Science Going: Aristotle on Entry Level Kinds. In Homo
Sapiens und Homo Faber, ed. Gereon Wolters and Martin Carrier. Berlin: de Gruyter.
. 2010. Aristotles Natural Science: The Many and the One. In From Inquiry to
Demonstrative Knowledge: New Essays on Aristotles Posterior Analytics, ed. James H.
Lesher, 125. Published as a special issue of APEIRON: A Journal for Ancient Philosophy
and Science 43, nos. 23 ( June/September 2010).
Leunissen, Mariska. 2009. Explanation and Teleology in Aristotles Science of Nature. Cambridge:
Cambridge University Press.
45

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions

HOPOS

l Aristotle on Norms of Inquiry

Pellegrin, Pierre. 2000. Aristote: Physique. Paris: Flammarion.

Sisko, John. 1999. On Separating the Intellect from the Body: Arisotles De Anima III. 4,
429a10b5. Archiv fur Geschicte der Philosophie 81:24967.
Waterfield, Robin. 1996. Aristotle: Physics. Oxford: Oxford University Press.
Whitteridge, Gweneth. 1981. William Harvey: Disputations Touching the Generation of
Animals. Oxford: Blackwell.

46

This content downloaded from 200.12.134.94 on Fri, 28 Aug 2015 08:53:42 AM

All use subject to JSTOR Terms and Conditions