The History of the Philosophy of Mind
General Editors: Rebecca Copenhaver and Christopher Shields
PHILOSOPHY OF MIND
IN ANTIQUITY
The History of the Philosophy of Mind,
Volume 1
The History of the Philosophy of Mind is a major six-volume reference collection.
covering the key topics, thinkers and debates within philosophy of mind, from
Antiquity to the present day. Each volume is edited by a leading scholar in the
field and comprises chapters written by an international team of specially commissioned contributors.
Including a general introduction by Rebecca Copenhaver and Christopher
Shields, and fully cross-referenced within and across the six volumes, The History of the Philosophy of Mind is an essential resource for students and researchers in philosophy of mind, and will also be of interest to those in many related
disciplines, including Classics, Religion, Literature, History of Psychology, and
Cognitive Science.
Edited by John E. Sisko
VOL.1 PHILOSOPHY OF MIND IN ANTIQUITY
edited by John E. Sisko
VOL.2 PHILOSOPHY OF MIND IN THE EARLY AND HIGH MIDDLE AGES
edited by Margaret Cameron
VOL.3 PHILOSOPHY OF MIND IN THE LATE MIDDLE AGES AND
RENAISSANCE
edited by Stephan Schmid
VOL.4 PHILOSOPHY OF MIND IN THE EARLY MODERN AND
MODERN AGES
edited by Rebecca Copenhaver
VOL.5 PHILOSOPHY OF MIND IN THE NINETEENTH CENTURY
edited by Sandra Lapointe
VOL.6 PHILOSOPHY OF MIND IN THE TWENTIETH AND
TWENTY-FIRST CENTURIES
edited by Amy Kind
€z Routledge
Taylor & Francis Group
LONDON AND NEW YORK
CONTENTS
First published 2019
by Routledge
2 Park Square, Milton Park. .Abingdon. Oxon OX14 4RN
and by Routledge
711 Third Avenue, New York. NY 10017
Routledge is an imprint of the Taylor & Francis Group. an a forma business
<; 2019 selection and editorial matter. John L. Sisko: individual chapters.
the contributors
The right of John F. Sisko to he identified as the authors of the editorial
material, and of the authors for their individual chapters. has been asserted
in accordance with sections 77 and 78 of the Copyright. Designs and
Patents Act 1988.
Notes On contributors
General introduction
British Library Cataloguing-in-Publication Data
Introduction to volume 1
Names: Sisko, John E., editor.
Title: Philosophy of mind in antiquity / edited by John E. Sisko.
Description: New York : Routledge, 2018.1 Series: The history of the
philosophy of mind ; Volume 11 Includes bibliographical references
and index.
Identifiers: LCCN 20170602501 ISBN 9781138243927 (hardback : alk.
paper)IISBN 9780429508219 (c-book)
Subjects: LCSI-1: Philosophy of mind—History.
Classification: LCC ßD418.3 .P484 20181 DDC l28/.209—dc23
LC record available at https://Iccn.Ioc.gov/2017060250
ISBN: 978-1-138-24392-7 (Vol I, hbk)
ISBN: 978-0-429-50821-9 (Vol I, ebk)
ISBN: 978-1-138-24393-4 (Vol II, hbk)
ISBN: 978-0-429-50819-6 (Vol R, ebk)
ISI3N: 978-1-138-24394-1 (Vol III, hbk)
ISBN: 978-0-429-50817-2 (Vol III, ebk)
ISBN: 978-I-138-24395-8 (Vol IV, hbk)
ISBN: 978-0-429-50815-8 (Vol IV, ebk)
ISBN: 978-1-138-24396-5 (Vol V, hbk)
ISBN: 978-0-429-50813-4 (Vol V, ebk)
ISBN: 978-1-138-24397-2 (Vol VI, hbk)
ISBN: 978-0-429-50812-7 (Vol VI, ebk)
ISBN: 978-1-138-92535-9 (6-volume set, hbk)
Typeset in 'Times New Roman
by Apex Co Vantage, LLC
1
JOHN E. SISKO
1
Presocratic interest in the soul's persistence after death
23
JOHN PALMER
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
X
REBECCA COPENIIAVER AND CHRISTOPHER SHIELDS
All rights reserved. No part of this book may he reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented. including photocopying and recording. or in
any information storage or retrieval system, without permission in writing
from the publishers.
Trademark notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identification and explanation
without intent to infringe.
vii
2 Presocratic accounts of perception and cognition
44
PATRICIA CURD
3 Soul, perception and thought in the Hippocratic corpus
64
I3YNEK BARTOB
4 Plato's guide to living with your body
84
RUSSELL E. JONES AND PATRICIA MARECIIAL
5 Plato and tripartition of soul
101
RACHEL SINGPURWALI,A
6 Cosmic and human cognition in the Timaeus
120
GABOR BETEGI-I
7 The power of Aristotle's hylomorphic approach
KELSEY WARD AND RONALD POLANSKY
141
Ci)NTIiNTS
8 Aristotle on the intellect and limits of natural science
160
CHRISTOPHER FREY
9 Aristotle on the perception and cognition of time
175
JOHN nOWIN
10 Aristotle on mind, perception, and body
CONTRIBUTORS
194
JOHN E SISKO
11
Rational impressions and the stoic philosophy of mind
214
VANESSA DE HARVEN
12 Mind in an atomistic world: Epicurus and
the Epicurean tradition
236
FRANCESCA MASI AND FRANCESCO VERDE
13 Galen's philosophy of mind
258
R.J. IIANKINSON
14 Plotinus' theory of affection
279
ANA LAURA EDELHOFF
15 Intellect in Alexander of Aphrodisias and John Philoponus:
divine, human or both?
299
FRANS A. 1. DE HAAS
Hynek Bartos is Associate Professor at Charles University. Faculty of Humanities. He has published in ancient Greek philosophy and medicine, including
papers in Ancient Philosophy, Apeiron, Classical Quarter/•, and Rhi_ai, as
well as a book: Philosophy and Dietetics in the Hippocratic On Regimen (Brill
2015). He co-edits a series of the first Czech translations with commentaries of
the Hippocratic treatises.
Gabor Betegh is Laurence Professor of Ancient Philosophy at the University
of Cambridge and Fellow of Christ's College. His work focuses on ancient
metaphysics, philosophy of nature, and cosmology, as well as the relationship between philosophy and religion. He is the author of The Derveni Papyrus. Cosmology Theology and Interpretation, Cambridge, MA: Cambridge
University Press, 2004 (paperback 2007), and the co-editor with Julia Annas
of Cicero De Finibus: Philosophical Approaches, Cambridge, MA: Cambridge University Press, 2015.
John Bowin is an Associate Professor of Philosophy at the University of California, Santa Cruz. He is the author of a number of articles on Aristotle.
Index
317
Patricia Curd is Professor of Philosophy at Purdue University. She is the author
of The Legacy of Parmenides: Eleatic Monism and Luter Presocratic Thought
(Princeton University Press, 1998; reprinted with an additional introduction by
Parmenides Publishing in 2004); of Anaxagoras of Clazomenae: Fragments;
Text and Translation with /Votes and Essays for the Phoenix Presocratics Series
of the University of Toronto Press (2007); and is co-editor (with Daniel W. Graham) of The Oxford Handbook of Presocratic Philosophy (Oxford University
Press, 2008) and has published work on early Greek philosophers including
Xenophanes, Heraclitus, Parmenides, Empedocles, Democritus, Anaxagoras,
and Gorgias.
Ana Laura Edelhoff studied philosophy, Greek, and Latin at the Freie Universität
Berlin and the University of Oxford. She is specialising in ancient philosophy
and contemporary ethics. Currently she is finishing her Ph.D. in philosophy at
the Humboldt University Berlin. From October 2016 until September 2017 she
vi
vii
RATIONAL IMPRESSIONS AND STOIC PHILOSOPHY
11.
RATIONAL IMPRESSIONS
AND THE STOIC
PHILOSOPHY OF MIND
Vanessa de Harven
Indeed, in order to approach the Stoics on their own terms, one must recognize
how they cut across their predecessors' ways of thought. In this case. the dit%rences between Plato, Aristotle and the Stoics on the philosophy of mind must be
couched as a debate over the scope of physics — namely, whether there is a place
for soul in physics. They each give different answers to the question. How much
can you sin' about the soul independent of body? For Plato (and Descartes). the
answer is (almost) everything, for Aristotle the answer is some, and for the Stoics
insofar as the Stoic soul is itself a body, there is no sense in which one
it's none.
can speak of soul without thereby speaking about body: and the topic of soul is
squarely in the domain of physics, in stark contrast to Plato. On the other hand,
insofar as the Stoics also see soul as a psychic entity entirely mixed with and yet
separable from a distinct corporeal entity (the body). the Stoics are more like
Plato than Aristotle. On this dualist understanding, the Stoic answer to how much
can be said about soul without body is Plato's, (almost) everything; in contrast to
Aristotle. Stoic soul is not what unifies an animal's body, but rather what gives it
At the heart of Stoic philosophy of mind is the rational impression (logike phantasia). As it happens, the Stoics think that the mind resides in the heart, but this
is not what makes the Stoic account interesting; seating the mind in the heart is
a commonplace for the time. What sets the Stoics apart is their focus on mental
phenomena of soul (psyche) over Aristotle's physiological mechanisms, and on
semantic content over Plato's desiderative psychology — all of which begins with
the rational impression. Much scholarly attention has been directed to the cataleptic impression central to Stoic epistemology, and to the downstream functions of
assent and impulse at the heart of Stoic moral psychology. The rational impression
itself, however, and thus the Stoic philosophy of mind proper, has been relatively
neglected.' This chapter seeks to elucidate the distinctive nature of the rational
impression on its own terms, asking precisely what it means for the Stoics to define
logike phantasia as an impression whose content is expressible in language.'
First some brief background on Stoic theory. The Stoics are well known for
their robust corporealism: they say that only bodies exist, or are, and they cast a
large swath of reality as corporeal. For example, all qualities or properties encompassed by Plato's Forms, including even the virtues, and the soul itself are considered bodies. They reason that insofar as the soul and body interact, and all
interaction must be corporeal, the soul must itself be a body.' Virtue is also a
body, namely the corporeal soul disposed a certain way, or in some state, like a
well-worn leather glove with its own patina, shape, and suppleness.4 The Stoics
are also well known for saying that, while only bodies exist, not everything that
is Something (ti, the Stoics' highest ontological genus) exists. Alongside their
innovatively robust category of corporeals (somata) that exist or
are (einai, on),
the Stoics recognize a class of incorporeal entities that subsist (huphistanai, have
hupostasis); these include, canonically, place, void, time, and the lekta,
or sayables, roughly the meanings of our words.' Thus the Stoics are not brute corporealists, but sophisticated physicalists grappling with Plato's beard.6
214
sensation and desire.'
1. Soul
I will begin by saying more about the soul considered as a body. Soul is pneuma,
a portion of fiery breath in a certain tension (toniken kinesin) "moving simultaneously inwards and outwards, the outward movement producing quantities and
qualities, and the inward one unity and substance".8 The cosmos is pervaded by
pneuma as the immanent divine guiding principle, and each body is what it is in
virtue of the particular state of rarity and tension of the portion of that pneuma
that constitutes it. The Stoic scala na!urae is a function of increasing complexity
and unity due to the state of a body's pneuma.`' A stone, for example. is a solid
object that holds together in virtue of the tenor (hexis) of its pneuma acting on
its matter. Plants are alive and are said to have physique (phials) because their
pneuma is more rarefied and in a greater state of tension than mere hexi.s; plants
are therefore more complex entities than stones in that the tension of their pneuma
is an internal principle of motion and rest rather than of mere unity. Finally, the
pneuma in animals is yet more rarefied and complex, and is called psyche because
the internal principle of motion and rest includes the capacity for impression and
impulse. Thus, from the standpoint of Stoic physics, soul is a body insofar as it is
pneuma in a certain state of tension.1 '
The soul's corporeality can also be considered from a metaphysical standpoint,
in terms of the so-called Stoic categories: substrate (hupokeimenon), qualified
individual (poion), disposed individual (pas echon), and relatively disposed individual (pros ti pas echon)." According to the Stoics, a complete analysis of any
body makes reference to all four of these metaphysical aspects. Applying this
fourfold analysis to soul, we can see that (l) soul qua substrate is pneu/na; (2)
each particular soul is a poion because its pneuma is in a state of tension such as
to constitute an individual soul with certain qualities, characteristics, and abilities;
215
VANI:SSA DL HARVI:N
(3) the 195.s. echon is that individual soul in a certain state, literally being in a
CeY _
tain way — this is how the Stoics corporealize the virtues that flummoxed their
materialist predecessors, and how they corporealize our thoughts. Virtue for the
Stoics is a stable state of character by which you see the world aright: it is a pos
echon because it is a habituated state of the poion. impressions, though they are
temporary and fleeting, fall into this category as well because an impression just is
the soul undergoing a pathos (i.e., being affected in some way and thus in a certain
state). (4) Finally, every soul also lies in a determinate relation to its immediate
surroundings and, ultimately, to the cosmos as a whole, so it is relatively disposed.
Now, the Stoics recognized two senses of the term soul, or psuche:
(a) the
corporeal entity as a whole that is mixed through and through with body. sustaining the composite animal, and (b) the commanding faculty (hégemonikon,
and
sometimes kurieuon), which is the highest part of soul.'= We have been considering soul in the first sense; the second sense refers to the hegemonikon, the part by
which an animal is aware of and engaged with itself and its surroundings — the
locus of impressions. We turn now to this second sense of soul and thus to Stoic
psychology and the philosophy of mind proper. The Stoics famously liken the
soul to an octopus, with the commanding faculty located in the region of the
heart, and seven other parts growing out from it and stretching into the body. Five
of these parts are the senses; for example, sight is pneuma that extends from the
commanding faculty to the eyes. The other two are the reproductive faculty, or
seed (sperma), extending from the commanding faculty to the genitals; and, in a
deeply innovative move whose implications are at the heart of this chapter, voice
(phone), extending from the commanding faculty of soul to the windpipe and
tongue." Again, what is characteristic of soul, in contrast to the mere phials of
plants, is that it has impression (phantasia) and impulse (horme); these psychic
functions make an animal aware of the world, and able to interact with it. Phantasia and horme are in effect the input and output faculties of the
hégemonikon,
a stimulus-response mechanism presupposing a single subject or self that thinks
and acts.14
2. Phantasia, generically
We turn now to the Stoic account of phantasia considered generically, as the
animal's input function.15 Starting from Aëtius' testimony about Chrysippus, in
the following passage, 1 will argue that phantasiai are states of direct, reflexive
awareness of the world.
A. (1) (a) On the one hand, impression (phantasia)
is an affection
(pathos) coming about in the soul, revealing (endeiknumenon)
itself
and what has made it (to pepoiekos); for example, whenever through
sight (dia opseós) we observe (theórómen) what is white
(to leukon),
RATIONAL. IMPRESSIONS AND STOIC PHILOSOPHY
touto to pathos) that we are able to say that something white stands
behind the motion in us (hupokeitai kinoun hemas); and likewise, too,
through touch and smell. (c) The impression is so-called from light:
for just as light reveals (deiknusi) itself and the other things in its
compass (periechomena), also impression reveals itself and what has
made it. (2) On the other hand, an impressor (phantastnn) is what
produces an impression: for example, what is white, and what is cold,
and everything able to move (ho ti an dunétai kinein) the soul, this is
an impressor.
(Aëtius. 4.12.1-5 (39B))''
The faculty of phantasia is defined in Ala as affection revealing itself and what
has made it, then illustrated by the paradigm case of seeing. Speaking generally.
the work of phantasia is to receive information from the world, be impressed by it
and produce particular impressions, or states of awareness in reaction to impressors that move the soul. Speaking more technically, phantasia is a state of awareness, in contrast to the raw sensory data before it reaches the mind. The language
reflects this distinction, between the
of observing something white through sight
(through
sight), and the awareness that
senses considered as arms of the octopus
will make use of this distinci
hégemonikon
(observing)."
takes place only at the
to refer to what takes place
sensation
and
sensing
tion in my analysis with the terms
in the arms of the octopus and the sense organs (what is through sight and through
seeing), perceiving and perception to refer to the impression that takes place in the
hegemonikon (when we observe), and sense-perception for sensory impressions
specifically, as opposed to non-sensory impressions "obtained through thought
(dianoia), like those of the incorporeals and of other things acquired by reason"
(DL 7.51 (39A4).' s The difference between sensation and perception lies in the
animal's awareness of the impressor's information: the motion that carries raw
sensory data from the organ to the hegemonikon is not something the animal is
aware of; the imprint it makes on the hégemonikon, however, must be a case of
awareness — given the psychic nature of the hégemonikon, its affections can't fail
to be cases of awareness. This is the force of defining phantasia in A I a as pathos
revealing itself and what has made it; revealing is awareness. A I b then confirms
that phantasia gives us cognitive access to the impressor that moved it, enabling
us to say things about it. And A lc elucidates the reflexive dimension. I will take
each in turn.
We can get a little clearer on our cognitive access to the impressor by looking
at the mechanics of perception. The importance of impression understood as a
corporeal pathos must not be underestimated; herein lies the transfer of information from impressor to soul.
an affection is what has been engendered in the soul through seeing (dia horaseós). (b) And it is according to this impression
(<kata>
B. They [the Stoics] say there are eight parts of soul: the five senses, the
principles of procreation, the vocal faculty (phonetikon), and the reasoning faculty (logistikon) [i.e., the hégemonikon]. Seeing is when the light
216
217
VANESSA DE IIARVEN
RATIONAL IMPRESSIONS AND STOIC PHILOSOPHY
between sight and what stands behind it (tou hupokeimenou) [the visual
impressor, cf. A2] is stretched in the form ofa cone.... The conical portion of air comes to be at the point of the eye, while the base is at what
is seen; thus the thing seen is reported through the stretched air, like a
walking stick.
second is what made the alteration, and this is what is visible"? What sense can
make of this self-awareness? To begin with, we can say with A. A. Long that
we
having an impression involves awareness of oneself as the locus of that impresphantasia is a natural extension of an animal's
sion.=' This reflexive dimension of
self-perception present from birth? Hierocles' Elements of Ethics is an important
(DL 7.157 (53N-))
source of information about self-perception in the Stoic school, and here is how
he puts it:
The comparison to a walking stick illustrates that air is no less direct and corporeal a conduit to the hégemonikon than a walking stick; both are conveying
their information by touching. So even though the object of sight does not itself
touch the eye, seeing remains direct in that the object moves the eyes via the
medium of air, and the eyes convey that very motion to the hégemonikon. The air
and the walking stick are conduits but not intermediaries; and so too the portions
of pneuma extending between hégemonikon and sense organs are conduits, but
not intermediaries. Crucially, neither is any particular phantasia an intermediary
between the world and the person perceiving it, because the phantasia is nothing
but a temporary state of the commanding faculty receiving it.
So, when an impressor strikes the senses it makes an impact that imparts
information about itself to the sense organs, and when that motion reaches the
hégemonikon, the animal is aware of the impressor. The details of impact and
transfer of information are hazy at best. However, we do know the Stoics embrace
an analogy with wax being stamped and impressed to capture all the idiosyncrasies of a signet ring.19 Crucially, the Stoics are not comparing the hégemonikon
itself with wax, but rather comparing the wax's taking on of all the ring's idiosyncrasies with the hégemonikon's taking on all of its impressors' idiosyncrasies.
20
Thus the mechanism by which the wax and soul take on their itnpressors' qualities
is not the explicit point of comparison, and their being similarly sensitive to their
impressors does not entail that the hégemonikon is itself like wax in any further
respect, nor that impressions are pictorial or imagistic in any literal way. Although
the details of the mechanism elude us, it is clear that there is a direct transfer of
information when the impressor strikes the sense organs, creating a motion in
the soul that carries the information imprinted on it to the hégemonikon, where it
becomes a pathos revealing what has made it.-'
Now we will turn to the reflexive dimension of phantasia revealing itself as
well its impressor, illustrated in A I c in analogy with light revealing itself and
what is in its compass. What is salient in the analogy is that light and phantasia
both serve to reveal things, themselves and their objects. In the case of light,
revealing is obviously to be understood as illumination, making things visible,
so at face value the Stoics have said that light makes itself visible alongside the
objects it illuminates. In the case of impression, revealing is to be understood as
making aware, giving cognitive access to; so the analogy taken on its own terms
dictates just that phantasia makes an animal aware of itself alongside its impressor.'-z Sextus Empiricus even makes it explicit that there are two things being
grasped in phantasia: "one is the alteration itself, this is the phantasia;
and the
218
C. (1) Since an animal is a composite (suntheton) of body and soul, and
(2) both of these are tangible (thikta) and impressible (proshléta) and of
course subject to resistance (prosereisei), and also (3) blended through
and through (di'holiön kekratai), and (4) one of them is a sensory faculty
(dunamis aisthaikon) which itself undergoes movement in the way we
have indicated, it is evident that an animal perceives itself continuously.
For (5) by stretching out and relaxing, the soul strikes against (proshallei) all the body's parts, since it is blended with them all, and (6) in striking against the body it receives a striking in response. For the body, just
like the soul, offers resistance (antibatikon); and the affection (pathos)
that results (apoteleitai) is a joint pressure (sunereistikon) and resistance (antereistikon) in common (homou). (7) From the outermost parts,
inclining in, it [sc. the pathos] travels ... to the commanding faculty
(hégemonikon), with the result that there is an apprehension (antilc:psin)
of all the body's parts as well as the soul's. This is equivalent to the animal perceiving (aisthanesthai) itself.
(Hierocles. 4.38-53 (53B5-9))
Self-perception is defined as an affection in the soul, an inherently psychic pathos
of the hégemonikon, resulting from the reciprocal pressure of body and soul. The
reciprocal nature of this pathos is what makes the impression reveal both itself
and the impressor, two things. As Hierocles explains, (I) the body and soul are
in contact; and (2) being corporeal, they can touch as agents, be impressed as
patients, and in so doing offer resistance to each other. Thus (6) the soul strikes
the body and thereby receives a blow in response, and the result is a single but
joint event or activity shared by agent and patient. This much follows just from
body and soul being in contact, as described in (1) and (2). The conclusion (7),
that self-perception is apprehension of all the parts of body and soul, requires
additional premises. Premises (3) and (5) establish blending as the kind of contact in question, which yields the all the conclusion; if soul and body were not in
contact by total blending, there would be awareness of only the parts that make
contact (as in an ordinary case of perception, where the impressor is an external
object of perception). And Hierocles confirms this reading (at 4.4-11) when he
tells us that blending is responsible for the joint affect (stanpatheia) being total
for both body and soul (but not for its being joint). Premise (4) then secures that
the joint pathos is a case of apprehension (i.e., awareness); because the patient is
219
VANESSA DE iIARVEN
a sensory faculty' (psychic by nature), its pathos is eo ipso a case of awareness.
More from Hierocles:
D. For in general the apprehension (antilepsis) of something external is
not completed (sunteleitai) apart from perception of itself: for in common with (meta) perception of what is white. it bears saying that we also
perceive ourselves being whitened (leukaino,nenôn) ... with the result
that since in all cases straight from birth the animal perceives somethinyg,
and perception of something else is naturally conjoined (sumpephuken)
with perception of itself, it is clear that from the start animals perceive
themselves.
(Hierocles, 6.3-10; cf. 6.17-22)
This passage makes clear that an impression requires for its completion a perception of the animal itself as undergoing something — it is not an impression
without a reflexive component. And because self-perception in the sense of continuous self-awareness is itself a state of the rational soul (i.e., the patient of the
striking), that self-perception is conjoined with the incoming information from the
impressor. Just as the psychic nature of the hégemonikon means its affections are
cases of awareness, so too the hégemonikon's state of continuous self-perception
makes its impressions reflexive. The preposition meta indicates the closeness of
this relationship, echoing the force of sumpephuken, literally grown together,
and
sunereistikon in C6. So, while the impression is itself a second object of awareness alongside its impressor, the reflexive story is rather more nuanced than this.
In the logical analysis of impression, there is only one impressor; but in the physical and psychological analysis, the hégemonikon is aware of two things jointly:
the impressor in relation to the self that is perceiving it. The impressor is the agent
and hence the proper impressor, but the pathos is a joint product (sumpatheia) of
agent and patient: impressor and soul together. The reflexive role of impression
revealing itself is thus not to be confused with the role of direct impressor.
Nevertheless, the self (i.e., the animal's constitution or articulation) can serve
as an impressor in its own right; in that case, the reflexive element remains in play
and the story is no different. The animal perceives itself as impressor, such as by
flapping its wings and focusing on how its wings or legs work, and in so doing
has an impression that is a joint product of impressor (self) and impression (also
self).27 It is aware of the self in relation to itself: this is my constitution, these are
my parts and their functions ... this is my impression. Thus there are two senses
of self-perception in play for the Stoics. First, the continuous joint pathos that is
a contributing cause to every impression; this is the reflexive element of phantasia. Second, self-perception with the self (the animal's present constitution and
articulation) in the role of impressor, conjoined with the reflexive element. Note
that neither sense of self-perception entails that the world is revealed through
phantasia as an intermediate entity. The animal is directly aware of two things in
220
RATIONAL IMPRESSIONS AND STOIC PHILOSOPHY
relation to each other. not one via the other. Thus phanmsia is best characterized
as direct. reflexive awareness of the world.
3. Phantasia logike, the rational impression
Thus far I have avoided using the term content, usually speaking in terms of
information that is conveyed, but it should be clear from my analysis that I take
impressions to be quite content-ful. Indeed, insofar as an impression is a state of
a wareness, it is characterized precisely by its content — what is impressed. As we
turn to the nature of the rational impression specifically, content will be front and
center. The question now will be: in what does the content of the uniquely human
rational impression consist? The short answer is that rational impressions have
content expressible in words or language. What makes our utterances language. as
opposed to mere vocal sound. is that they are significant (semcltika and what is
signified by the speaker and grasped by the rational hearer is a lekton.'-ç As we will
see, lekta are inextricable from rational impressions. The question then will be:
what is the relation between rational impressions and lekta? We will begin with
some passages describing the rational impression unique to humans.
E. Further, among impressions, there are those that are rational (logikai) and those that are irrational (alogoi); and rational are those of the
rational animals. while irrational are those of the irrational. Thus rational
impressions are called thoughts (noeseis), while the irrational ones don't
happen to have a name. And there are those that are expert and those that
are inexpert; at any rate (goun), a statue (eikiin) is viewed one way by an
expert and another way by a non-expert.
(DL 7.51 (39A6-7); cf. Galen, Del med (SVF 2.89))
F. For the impression arises first (proegeitai), and then thought, which
has the power of speaking out (eith'hé dianoia eklalétiké huparchousa),
expresses (ekpherei) in language (logo(i)) what it undergoes by (ho parchef hupo) the impression.
(DL 7.49 (39A2/33D))
G. They [sc. the Stoics] say that the lekton is what subsists (10 huphistamenon) according to (kata) a rational impression (logiké phantasia), and
a rational impression is one in which the content of the impression (to
phantasthen) is expressible (esti parastesai) in language (logo(i)).
(SE, M. 8.70 (33C); cf. DL 7.63 (33F))
This sequence of passages shows that what is characteristic of humans is the
rationality of their impressions, and that impressions are rational when they are
thoughts whose contents can be expressed in words or language. One might even
221
VANESSA DE HARVI:ti
frame these passages as a syllogism: rational impressions are thoughts (E): though
t
is linguistic and semantic (F): therefore lekta (the Stoics' linguistic and semantic
entities) subsist according to thought. the rational impression (G). Now, this much
is uncontroversial, but only because it does not yet take a stand on what it means
for the lekton to subsist according to, or kaut, the rational impression
(cf. Alb)
Everyone agrees that lekta are the expressible content of rational impressions, but
there is an important disagreement as to whether the !ekta give otherwise semantically empty impressions their propositional content, or whether lekta
owe their
propositional content to the rational impressions according to which they subsist.
We can sloganize the debate by asking whether rational impressions are conduits
or causes of lekta. 29
I avoided operating in terms of content because the term is laden with interpretive baggage, which it is now time to lay bare. The most basic presupposition
about mental content is that it is propositional. This by itself says little, since
everything hangs on what it means to be propositional; so let's accept it and see
what comes out with different senses of propositional. One sense is practically
axiomatic in the literature: to be propositional is to have content that is all and
only from lekta construed as mind-independent entities (i.e., propositional content comes from thought grasping propositions).30 It is a corollary of this view
that sense-perception reports only bare sensory qualities (e.g., colors, shapes
and sounds), which are not part of the propositional content, on the model of the
wooden horse in Plato's Theaetetus (184-186). if all content comes from
lekta,
whatever else there is (including the sensory information from the wooden horse).
it doesn't count as content. It is also a corollary of the view, that sense-perceptions
are objects or inputs for reason construed as something distinct; the mind that
thinks is aware of phantasiai as its objects. Since reason is what distinguishes
humans from animals, it must be some part of soul in addition to the faculty of
impression they have in common.
This picture first took hold with William and Martha Kneale, Jan Lukasiewicz,
and Benson Mates who brought much insight and attention to Stoic logic in the
mid-20th century. The philosophical currency of propositions is established there
by reference to Frege, Carnap, Quine, Church, and others.31 The lekton
is thereby
taken to be an independent semantic entity that gives our subjective and private
thoughts their objective semantic content. Call this the de-psychologizing orthodoxy. Michael Frede, who did much to establish this picture as the orthodoxy,
takes the view that impressions are the way that a lekton is perceived, but of themselves have no propositional content.32 This assumption that all semantic content is from lekta, plays out in an ongoing debate over the status of non-rational
RATIONAL IMPRESSIONS ANI) STOIC PHILOSOPHY
orthodox view is that if all mental content is propositional and conceptual, then it
looks like animals and children (since they are classified as irrational) cannot have
mental contents, only the bare sensory qualities of the wooden horse: but these are
not sufficient to explain complex animal behavior. Thus one solution is to say that
content can be propositional without being conceptual. in which case children and
animals can have articulable mental contents sufficient to explain behavior without having to grasp that content.'' This is an unfortunate choice: either animals
and children have mental content without being able to grasp it, or they have no
mental content at all. My suggestion is that this is a false dichotomy resting on the
mistaken axiom and corollaries of the de-psychologizing orthodoxy. So, l propose
to suspend the axiom that semantic content only comes from lekta and see where
it leads when we take our inquiry to the texts and read the Stoics on their own
ternis. We will begin with Sextus.
H. They say that the human does not differ from irrational animals in
respect of uttered speech (prophoriko(i) logo(i)) (for crows and parrots and jays utter articulate sounds), but in respect of internal speech
(endiatheto(i)), nor merely by the simple impression (for these too
have impressions) but by the transitive (metabatikêê(i)) and synthetic
(sunthetiké(i)) impression. Wherefore (dipper), having the conception of
implication (akolouthias ennoian echön), straightaway one also grasps
the thought of a sign, through implication; for, that is, the sign itself is
such: "if this, then that." Therefore the fact that there are signs also (to
kai sémeion huparchein) follows from (hepetai) the nature and constitution (kataskeuê(i)) of the human being.
(SE, M. 8.275-6 (53T); cf. M. 8.285, PH 1.65)
Sextus reports that humans differ from animals in two ways: internal speech
and rational impressions. First, it is not the actual proffering of speech that
makes us different from animals, but endiathetos logos, which we can think
of as the ability to make statements to oneself." This gap between impression
and utterance where internal discourse takes place is what gives us the uniquely
human control over assent and impulse, in contrast to blue jays that utter articulate cries as an automatic response to their impressions.3G Origen confirms this
impressions in animals and children. This debate is thought to be instructive on
an Aristotelian-style assumption that since animals and humans have impression
in common, their impressions must be the same (on the wooden horse model,
corollary one) and reason must be something further receiving those impressions
(corollary two). So if we can get clear on what goes in animals and children,
we'll know about the added ingredient of reason as well.33 The problem facing the
picture when he reports that the rational animal "has reason that passes judgment on impressions (logon ... ton krinonta tas phantasias), rejecting some of
these and accepting others, in order that the animal may be guided according to
them [se. those they accept]"." Passing judgment on impressions does not signal
that the mind takes impressions as its objects on a wooden horse model, but rather
that there is a gap between perception and assent where reflection and deliberation take place. This is self-perception in the second sense described previously,
not merely reflexive but with the self in the role of impressor. indeed, examination of our impressions is the most important kind of self-perception because it
is for the sake of evaluating the way we see the world, analyzing our seeings
222
223
VANESSA DE IIARVEN
before owning them with our assent. In this gap we have the opportunity to evaluate and examine our impressions before passing judgment. choosing the lektonNe
assent to and becoming responsible for the resulting impulse and action. Epictetus
calls it parakolouthia and even personifies the process for rhetorical effect: "Wit
for me a bit. phantasia; let me see who you are and what are about. let me test
you"." Given this gap between impression and assent, there must be somethin(,
about the rational impression that makes it available for this kind of examination
prior to assenting to a lekton.
And this is precisely the other thing differentiating us from animals: because
impressions are rational they can be examined and evaluated. Sextus
explains
that our impressions are transitive, literally moving from one place to another
(metabatike), and constructive or compositional (synthetike).
We can capture this
point by saying they are discursive: being metabatike signals that rational impressions are inferential, and being .synthetike signals that they are compositional they are seeings as (as I will put it).39 And this is to say that rational impressions
are conceptual and thus propositional. To see something as F is to predicate a
concept of it; and since Stoic concepts are analyzed as conditionals, seeing as is
inherently inferential.40 The rational impression, being discursive, thus looks quite
content-fir/, with precisely the kind of content whose implications can be evaluated and selected for assent or rejection. Here we see how the rational impression is at the heart of Stoic moral psychology, as the foundation of oikeisos
and
prohairesis, and the lynchpin of their compatibilist ethics, as the one thing in the
cosmos over which we have control. So, to summarize, the content of the rational
impression is inherently inferential, conceptual, semantic, and linguistic — that is,
propositional.41 Thus, with the benefit of passage H, we see an alternative sense of
propositional emerging, which does not make reference to propositions grasped
as independent objects of thought.°'- We see an alternative axiom emerging as
well: the content of the rational impression is in the impression. That is why they
are called thoughts (E).
With the help of Cicero's Acad. 2.21, we can also see an alternative to the first
corollary, the wooden horse model of perception. This passage has been of interest in the animal minds debate, as a sort of loophole to ascribe enough content to
animals to account for their behavior without giving them full-fledged rationality.
But with the axiom that mental content comes all and only from lekta suspended,
there is no reason to seek this loophole. Denying lekia to animals need not deny
RA"iIONAL IMPRESSIONS AND STOIC PHILOSOPHY
(animo iam), not the senses. Next (deinceps). "That is a horse," "That
is a dog." The rest of the series follows next. connecting greater things
that are as if to encompass a complete comprehension of things: "If it is
a human, it is a mortal, rational animal." From this class [sc. of things
perceived by the mind] conceptions (notitiae) of things are imprinted
(imprin naltur) on us, without which there can be no understanding (intellegi) nor discussion (disputari) of anything (quicguam).
(Cicero. Acad. 2.21 (39C))
Those things perceived by the senses themselves are colors, tastes, and sounds
et al. that get communicated to the soul by the sense organs: this is the raw sensory data. These colors, tastes, and sounds are equally characteristic of (i.e., still
present when those things are registered by the senses in a certain respect, such
as predicatively, as white, sweet, or melodious). The idea here is that the sensory
content delivered by the arms of the octopus persists and in that respect is equally
characteristic of the content of the impression when it strikes the hegemonikon
and is seen as F. The force of saying these are comprehended by the mind now
is to signal the awareness that takes place once the sensory impact reaches the
hegetnonikon from the arms of the octopus — the mind, but not the walking stick
will be aware of the impressor as F. So much is confirmed by Sextus:
For what grasps the truth in those things underlying (hupokeimenois)
[sc. the impressors, A1b] must not only be moved in a white manner
(leukantikós) or a sweet manner (glukantikós), but also be led (achthenai)
to an impression of such a thing that "This is white" and "This is sweet."
(M. 7.344, cf. 7.293)
Sextus goes on to confirm that this is no longer the job of sense because sense
grasps only color, flavor and sound, but cannot grasp "This is white." Thus I take
the idea that the mind grasps what the senses offer in a certain respect to indicate
that rational impressions are inherently conceptual, echoing Sextus' description of
is sweet," "That is melodious," "This is fragrant," "This is bitter." We
have comprehension (tenemus conprehensa) of these by the mind now
them as synthetic and inferential. I do not find the wooden horse model in this passage, taking impressions to report only bare sensory qualities to thought as something distinct. Rather, taking the testimony that rational impressions are thoughts
(E) at face value, I find the alternative corollary that impressions are inherently
conceptual states of mind, which I have sloganized as: all seeing is seeing as.
An alternative to the second corollary that mind, or thought is something distinct that takes impressions as its inputs, is also discernible in Cicero's testimony.
This passage is in fact describing the Stoics' developmental account of rationality, according to which reason is constituted by the concepts we acquire through
experience.43 For the Stoics, humans have logos, or reason, from birth; it is the
faculty or capacity for rationality, which develops gradually through our interactions with the world. The Stoics are akin to modern empiricists in this regard,
starting us with a tabula rasa "fit for writing upon," and equating our rationality
224
225
them any mental content whatsoever, nor does it give any reason to think that
sense-impressions merely provide raw sensory data.
1. Those characteristics that belong to those things we say are perceived
(percipi) by the senses (sensibus) are equally characteristic of that further
set of things said to be perceived not by the senses themselves (ipsis) but
by them in a certain respect (quodam modo), e.g. "That is white," "This
VANESSA DE HARVEN
with the concepts we write on it. The initial development of rationality consists
in our basic concept acquisition, first of preconceptions that arise naturally from
the world, and then conceptions, which are a function of study. art and convention:* Thus, in passage I, we begin to get rationalized with basic sensory concepts like white, sweet, melodious, fragrant, and bitter: next we move to kinds
like horse and dog. Then we move from basic predication to connecting our
concepts together and appreciating their inferential import, as in: if somethin
g
is a human, then it is a mortal, rational animal. The human soul is considered
rational once it has acquired a complete stock of concepts from experience, and
a proper grasp of their inferential import. Only then is a person considered to
have voice; that is, to be a reliable language user who understands the meanings
of her words and is responsible for her actions. Only then do our impressions
become thoughts.
It is important to appreciate the full force of this point, that impression are
thoughts. I said earlier that the psychic nature of the hégemonikon (the kind of
poion that it is) makes the impressor's imprint result in a case of awareness,
a pathos that reveals itself and its cause. Further, that the state of continuous
self-perception (the poion pas echon) makes every impression reflexive. Now,
by that same token, we can appreciate that the rationalized state of the human
hégemonikon makes the pathos propositional in addition. The habituated state of
the hégemonikon (pas echon) determines how the incoming information is conceived (i.e., how the patient is affected). And the way the patient is affected is
what is impressed (to phantasthen); that is, the content of the rational impression
that is expressible in words. So, in answer to the first question posed in this section, in what does the content of the rational impression consist, the answer is this:
incoming information from the impressor, as it is conceived. This is not two steps
or two components, but one. Another way to put my point is to say that I reject the
distinction between character and content; the content of an impression just is the
incoming information, the way it is conceived.
This idea that the state or disposition of a person's soul determines the content
of the rational impression is confirmed by the Stoic account of expertise. Earlier, in passage E, rational impressions were subdivided into expert and inexpert.
Let's take this at face value and ask what it means for a rational impression to
be expert. We know that the Stoics explicitly categorize expertise (e.g., having
the art of medicine or, ideally, the art of life) as a tenor (hexis) of soul, and so a
pas echon. Further, that "expertise is a system [developed] out of (ek)
cognitions
(katalepsean) in joint training (sunggegumnasmenan) relative to some goal useful in life".'6 To be an expert, then, is to have developed one's rationality with
care and attention to some goal. Just as basic human rationality is constituted
by the development of our preconceptions and conceptions from our interactions with the world, so expertise is just a further habituation or training of the
hégemonikon with a certain end in mind, say carpentry, medicine, or virtue (the
art of life). As the preposition ek signals, expertise is not a new ingredient or
part of soul, but soul itself disposed in a certain way —pneuma that is maximally
226
RATIONAL IMPRESSIONS AND STOIC PHILOSOPHY
sensitive to the world's maximal intelligibility and detail, just as the comparison
with wax dictates.
Julia Annas finds it puzzling that the impression itself is characterized as expert
or inexpert and suggests an interpretive dilemma." On the first model, the expert
and the non-expert have the same phantasiai with the same content at step one,
but assent to different lekta at step two because they only take in or accept part
of the information contained in the impression. On this picture, the information
contained in the phantasia is raw sensory data (on the wooden horse model) and
step two marks the introduction of conceptual or propositional content. On the
second model, the expert and inexpert have different phantasiai with different
content. but there is no distinction to be made between perception and assent.
only a distinction between the striking of the sense organ (step zero) and what
happens at the hégemonikon (step one = step two). Annas, stumped. finds no reason to prefer either horn of the dilemma. But notice that the dichotomy rests on
the orthodox assumption that the lekton assented to contains the content of the
phantasia. If phantasiai only have content derivatively in virtue of grasping an
external, independent lekton, experts and inexperts can only differ in these two
ways: either they have the same raw, non-conceptual data (same phantasiai) and
assent subsequently to different lekta, or they have different phantcrsiai with different conceptual content and eo ipso assent to different lekta. With the orthodoxy
suspended, however, there is no reason to accept Annas' dichotomy.
Rather, a natural alternative emerges that takes the expert and inexpert to have
different phantasiai with different content at step one and sees assent (step two)
as a distinct phase of cognition on the other side of internal discourse. The world
in all its specificity makes a causal impact (tupasis) on our sensory apparatus
(step zero); this much is the same in everyone, expert and inexpert alike (assuming comparable eyesight, e.g., neither needs glasses). Then, in analogy with the
walking stick, the tupasis imprinted with the impressor's information travels up
the arms of the octopus to the hégemonikon, where it becomes a phantasia. This is
step one, and it is not the same in the expert and inexpert, because their souls are
differently habituated and so they receive the information differently. An expert
soul is sensitive to more of the incoming information, capable of seeing more than
the novice. When an arborist looks at a certain tree, she sees a silver birch at a particular stage of its life, in some determinate state of health, etc. I see a tree. When
an artist looks at a painting, she sees scale, composition, brush strokes, media, and
technique, reading the creative process off the canvas. I see a painting. We see the
sane painting, and we have the same incoming sensory data, but we do not see
it the same way. As Cicero puts it just before telling us that things are perceived
by the senses in a certain respect (Acad. 2.20): "How many things painters see in
shadows and in the foreground which we do not see!" The content of an expert's
impression (the expert's seeing as) is different from the non-expert's because the
expert soul is habituated by experience and art, a collection of inferential conceptions and maximal sensitivity to detail. There is thus a wealth of content ill the
expert's rational impression, which cannot be captured by any one simple lekton.
227
VANESSA DE HARVEN
RATiONAI. IMPRESSIONS AND STOIC PHILOSOPHY
This is why impressions are neither objects of thought for a separate mind (corollary 2) nor reporters of mere sensory data on a wooden horse model (corollary i ).
but thoughts about their impressor — thick with conceptual and inferential content
resulting from the impressor making an impact on the hégemonikon.4
4. Phantasia logike, source of lekta
We now turn to the second question raised earlier in the chapter, what is the relation between the rational impression and the lekton? What does the preposition
kata convey when we are told that the lekton subsists according to
the rational
clearly come to be around the heart, both voice and [public] discourse
issuing from the heart through the windpipe.
(Galen, PHP 2.5.16-20)54
Here is a breakdown of the passage: (16) the commanding faculty is where
meanings and discourse (logos) come to be. Why? First, because internal discourse has its source in thought, and voice has its source in internal discourse:
therefore (simply put) voice comes from the hegemonikon (where thought takes
place)." (18) Second, because the source of discourse must he (i.e., on principle)
impression (A 1 b, F)? The order of explanation in passages F and G already
suggests that lekta depend on the rational impression: F says, first we have an
the place of reasoning and thinking, and where words are imbued with meaning.
Diogenes of Babylon confirms: "discourse is sent forth having been imprinted
(ensese.masmenon) and in a way stamped (hoion ektetupdmenon) by conceptions
impression, then we express its content (by uttering lekta):49 G says, lekta
subsist
according to the rational impression, because the rational impression has the content to give them their subsistence.50 More concretely, in passage H after telling
in the mind" (Galen, PHP 2.5.12). Thus we see that the mind-dependence of lekta
is attested by a wide variety of sources: a hostile skeptic, neutral doxographer.
fussy Platonizing physician, and now in a grammatical context by Ammonius,
us that humans have internal discourse and rational impressions, Sextus goes on
to infer (and reiterate at 8.285) that the existence of conditionals follows from the
human nature and constitution. Given that conditionals are composed of lekta,
this conclusion amounts to an explicit claim that lekta depend on rational impres-
head of the neoplatonist school in Alexandria:
sions as prior semantically (since they have conceptual content of themselves)
and ontologically (the fact that there are conditionals follows from the rational
impression).
We can add to this now the explicit testimony that lekta arise out of (ek),51
that is, from and in consequence (parhuphistamenon)52 of the rational impression. Further, Sextus tells us at M. 8.80 that for the Stoics speaking
(/egein)
is uttering voice that signifies the subject matter in mind (to ten tou nooumenou pragmatos semantike propheresthai phiinen). Likewise, Diogenes reports
that "language is semantic voice (phOne sémantike) sent forth from thought
(ekpempomene apo dianoias)".53 Galen also lays bare the Stoic commitment to
the dependence of lekta on thought, quoting Zeno, Diogenes of Babylon, and
Chrysippus on this pòint: mind is the source of language. Following is the quote
he attributes to Chrysippus, telling us explicitly, that words receive their meaning from thought.
J. [16] It is reasonable (eulogon) that that in which (eis ho) the meanings
at that point (hai en touto(i) semasiai) come to be and out of which discourse (logos) comes to be, is the ruling (kurieuon) part of soul. For the
source (pege) of [internal] discourse (logos) is none other than thought
(dianoias), and the source of voice (phones) none other than [internal]
discourse, and [i.e.,] (kai) on the whole, simply, the source of voice is
none other than the ruling part of soul.... [ 18] For on the whole, whence
K. The Stoics reply [to the peripatetic] that the nominative case itself has
fallen (pepteken) from thought (apo noematos), which is in the soul. For
when we wish to reveal (delesai) the thought of Socrates that we have in
ourselves, we utter (propherometha) the name Socrates [i.e., Socrates in
the nominative case]. Therefore, just as a pen is said both to have fallen
and to have its fall upright if it is released from above and sticks upright.
so we claim that the nominative case [literally 'the direct case] falls
from thought (apo tes ennoias), but is upright because it is the archetype
of meaningful utterance (tés kata ten ekphonesin prophoras).
(Ammonius, InAr De int. 43,9-15 (33K))"
Jonathan Barnes has objected to taking this (and a plethora of other texts connecting lekta with thoughts) as evidence of Stoic semantic theory. He urges that passage K expresses no more than the commonplace that producing someone's name
is a good way of telling someone who you are thinking about, and cautions that
[e]ven if Ammonius in this passage is referring to the Old Stoa (which
is not clear), and even if he is reporting Stoic views with fidelity (which
cannot be taken for granted), nevertheless his report has nothing to do
with any theory of meaning. In general, we should not read philosophical
theories into platitudes.
(Barnes 1993, p.54)
(hothen) discourse (logos) issues must be where (ekeise) reasoning (dialogismos), thinkings (dinanoeseis) and preparations (meletas) of linguistic expressions (réseiOn) come to be, just as i said. [19] And these things
As a founder of the de-psychologizing orthodoxy, Barnes' dogma should be
taken with a grain of salt. It is doubtful that Ammonius spends his time rehearsing
platitudes rather than theory. Further, his caution about platitudes cuts both ways,
since it is equally important that we not allow philosophical theories to upend
228
229
VANESSA Of
iIARVi:N
RATIONAL IMPRESSIONS AND STOiC PHILOSOPHY
platitudes, in this case to upend the wealth of clear textual evidence attesting to
the mind-dependence of lekia.
whose content can be expressed in words is to get to the very heart of the human
mind: we are above all the language animal. and lekta are the currency of agency.
For example. Barnes writes off Galen's testimony as a physiological comment
about the windpipe, which says nothing about words receiving their meaning
from thought; then he adds, even if we do have to read this evidence as establishing that thinking causes speaking, it is obviously false that speech is caused b\
thought because we sometimes speak unreflectively (i.e.. without thinking).
But
the fact that we sometimes speak unreflectively is hardly a counterexample to the
general claim that human voice is made semantic by thought. And Barnes' dismissal of Sextus' testimony at 8.80 as telling us that voice is significant because
the pragma it has in mind is just a lekton is circular: presupposing that rational
knowledge and rationality. 55
impressions have semantic content only because they grasp independent
lekta.
The scarcity of our textual evidence makes every passage precious, so Barnes'
summary dismissal of a dozen different passages connecting lekta with thoughts
is not to be taken lightly. Tossing so much evidence aside is a steep price to pay
for any interpretation.57 However, my point is not just that the cost is too high, but
that the expense is artificial. With the axiom suspended, there is no need to dismiss
all this textual evidence, or to fall on the horns of false dichotomies.
5. Conclusion
Taken on its own terms, the Stoic rational impression reveals a rich philosophy
of mind driven by content. The rational impression is a state of the rational mind,
the human hegemonikon rationalized by experience. It gives us direct cognitive
access to the world (as the image of the walking stick suggests), conjoined with
self-perception. Being an affection of rational soul, it is inherently propositional
and thick with content for internal discourse. We can now see how many ways
rational impressions are at the heart of Stoic philosophy of mind. First, to have a
mind just is to have rational impressions; Stoic psychology is entirely monistic
and cognitive through and through, paving the way to an austere Socratic intellectualism. Second, to be subject to internal discourse is to be able to engage in
the most important kind of self-perception, parakolottthesis: the ability to put the
impressor in the role of impression to see what it tells us about ourselves. Third,
for rational impressions to be subject to internal discourse is also to have control over our assents, actions and state of character, and thus to have both moral
responsibility and hope for moral improvement; here we find the famous doctrine of oikeiosis. Further, the fact that rational impressions are direct and reflexive states of awareness has implications for Stoic epistemology; it will be clear
already to the initiated that my account of rational impressions will psychologize
the cataleptic impression, in contrast to externalist readings that are part and parcel of the orthodoxy I have been describing. Finally, because rational impressions are thick with propositional content, they are the source of lekta: causes
not conduits. Thus we can see how the Stoic philosophy of mind reaches across
ethics, epistemology and into logic. To define logike phantasia as an impression
230
Notes
With the exception of Barnouw (2002)
Sextus Empiricus (SE)..1/. 8.70 (33C). passage G later; parenthetical citations like 33C
refer to Long & Sedley (1987) by chapter (33) and order therein (C).
3 Nemesius 78,7-79.2 (45C). 81.6-10 (45D).
4 SE. .1/. 11.22-26 (60G): Stobaeus 2.70.21-71.4 (601). 2.73.1-13 (60J): Simplicius.
In Ir Cat. 212,12-213.1 (28N).
5 Alexander. In Ar Top. 301.19-25 (2713): SE, M. 1.17 (27C): Galen. Alert. med.
10.155.1-8 (27G). I will often transliterate Greek terms to avoid awkward English,
what is
here lekton (singular) and lekta (plural): alternate translations of lekton include
said. what is meant. thing said. articulable. meaning.
6 As Quine (1948) terms the problem anon-being: cf. de Harven (2015) and Vogt (2009).
7 Annas (1992. pp. 50-56) and Long (1999, p. 564. 1982. pp. 34-36 & 44. contra BonholTer and Pohlcnz).
8 Nemesius 70.6-71,4 (47J).
9 Annas (1992. pp. 51, 50-56).
10 Nemesius 291.1-6 (530); Philo, Leg. alleg. 1.30 (53P). 2.22-3 (47P): I lierocles 1.533. 4.38-53 (5313): Galen, Intr. 14.726,7-11 (47N).
11 I follow Long & Sedley in describing these as metaphysical aspects under which an
object can be considered. although I am not entirely satisfied with the term: cf. Mena
(1999), Rist (1971). Sedley (1982), and Sorabji (1988).
12 SE. M. 7.232, 7.234 (53F): Calcidius 220 (530): Aimas (1992. p. 63): Long (1982. p. 239).
13 Ai:tius 4.21.1-4 (53H); Calcidius 220 (53G): DL 7.157 (passage 13): Long (1982.
p. 51): Sedley (1993, pp. 330-331).
14 For the relevant notion of sell see Annas (1992. pp. 58-59. & 64). Inwood (1984,
pp. 162-164), and Long (1991, p. 107 & 1999), all of which I take to be compatible
with cautions from Gill (2006).
15 This vexed term has been translated many different ways. as appearance. image. representation, presentation. impression, and others. Although I will (Alen transliterate
phantasia (singular) and phaniasiai (plural). I have chosen impression as the most phenomenologically neutral translation. which also captures the literalness of' the impact
of the world on soul and aptly conjures the empiricism of the modern era.
16 This is part of a fourfold analysis comparing phantasia and its objects. impressors.
with imagination and its objects, apparitions.
17 Galen, P1-/P 2.5.35-36 (de Lacy); Aëtius 4.23.1 (53M): i-fierocles 4.38-53 (53135-9):
cf. Long (1982, pp. 47-48 & 95-97) and Annas (1992. pp. 62-63 & 85) for comparison
to the brain and central nervous system.
18 The account I give will proceed in terms of sensory impressions and often just in terms
of vision, which arc the paradigm cases of impression, but everything 1 say is meant to
be applicable to non-sensory impressions as well, with the soul acting as corporeal agent
to generate the pathos by being impressed in relation to (epi) things with a nature like
the incorporeals (what is intelligible. vs. sensible). as reported in SE, M. 8.409 (27E).
19 SE, M. 7.251 (40E3).
20 As Chrysippus' resistance to overextending the analogy confirms, D1.. 7.50 (39A3).
21 Cf. Inwood (1984, pp. 161-164) and Sedley (1993, pp. 330-331) for confirmation that
no gap is possible between body and soul.
231
VANESSA DE tlARVEN
22 Note that the causal direction is reversed in these two cases: light makes itself-and the
things it illuminates visible. but phantasia does not make the world intelligible: this
difference does not undermine. but rather reinforces what is salient in the analogy.
23 SE. M. 7.162.
24 I.ong (1982, p. 47).
25 Seneca, Ep. 121.6-15 (5713). I iierocles 1.34-9. 51-7. 2.1-9 (57C):
cf. Long (1982&
1991. p. 107) on sell-perception. and (1993. p. 96) fòr a comparison of self perception
to proprioception.
26 Thus Reed (2002) cannot be right in relegating the impression to a bodily precondition
for awareness and thought. something we are aware of only indirectly via the impressor.
27 Hicrocles 6.50-7.9, where the self is even described as an aisthéton. an object of
sense-perception.
28 SE. M. 8.11-12 (33B): DL 7.57 (33A).
29 I do not intend the word cause in the technical Stoic sense. which requires all causes to
he capable of action and passion: by cause 1 mean source. or grounding body.
30 Cf. Brittain (2002) as the notable exception to Annas
(1992). Barnes (1993. 1999),
Caston (1999. unpublished). Frede (1987). Inwood (1985). Lesses (1998). Mates
(1961), Sedley (1993), and Sorabji (1990).
31 See especially Mates (1961, pp. 19-26).
32 Cf. Frede (1987, p. 156).
33 Cf. Brittain (2002, pp. 256-257) and Lesses (1998. pp. 1-3) for perspicuous summaries of the issues.
34 As Caston (unpublished) and Sorabji (1990) suggest; note here the seeds of epistemic
externalism, which is a related orthodoxy established by Annas (1990), Frede
Striker (1974); cf. Nawar (2014) and Perin (2005) for recent defenses of internal(1987),
ism.
35 As Long (1971, p. 88) puts it: cf. Gourinat (2013) for additional texts and valuable
cautions against assimilating Stoic endiatltetos logos to the Platonic notion more akin
to internal dialogue than declamation (diexodos).
36 This point may be put either in terms of denying assent to animals and introducing
it only with humans, or granting animals assent but calling it voluntary only with
humans; cf. Brittain (2002, p. 257, n. 10).
37 Origen, Princ. 3.1.3 (53A5); cf. Calcidius 220 (530).
38 Epictetus, Diss. 2.18.24-26.
39 As Bury (1935) translates, they are transitive and constructive: cf. also Long (1971.
p. 87 and n. 54). For synthesis as combination, see DL 7.53; SE, M. 8.60; Cicero. Fin.
3.33-34 (60D).
40 SE, M. 11.8-11 (301); this paraphrasing move is how the Stoics eliminate concepts
from the ontology (cf. Caston (1999)).
41 I will now use these terms interchangeably - with the axiom suspended, priority among
them is not material.
42 I take this sense to be akin to Brittain (2002) and Lesses (1998).
43 Galen, PHP 5.2.49, 5.3.1 (53V).
44 Aëtius 4.11.1-4 (39E); cf. Long (1982, p. 51) for whom logos
is not one faculty among
others, but the mode of the whole soul's operation.
45 i am in agreement with Sandbach
(1930, contra Bonhoffer), that prolèpseis are not
innate ideas; cf. Henry Dyson for a more recent defense of such Platonizing.
46 Olympiodorus, In Plat. Gorg. 12.1 (42A). Cf.
also Stobaeus 2.73,1-13 (60J); Cicero,
Acad. 2.22 (42B); Plutarch, Virt. mor. 440E-441D (61 B).
47 Annas 1992, pp. 82-83.
48 Cf. Long (1991, pp. 109-110) who speaks of rational impressions as thick with content and potential judgments; Caston (unpublished) has recently called into question
the thin notion of content and passive model of mind, but he remains committed to
232
RATIONAL IMPRESSIONS AND STOIC PHiLOSOPiIY
the orthodox axiom and thus differs only in requiring more propositions to imbue the
impression with its full content
49 This priority is not merely temporal: immediate context (not quoted) shows that this
order explains the priority of rational impression to assent, knowledge and reasoning.
50 Others in the current minority that favors mind-dependence include Imbert (1978).
Long (1971. 1991. 1999. 2006). and Nuchelma ns (1973).
51 DL 7.43: Barnes (1993. p. 56) does cop to the "lameness' of dismissing this piece of testimony as generally confused in order to work around the obvious language of dependence.
12 SE. Al. 8.11-12 (33B): Sryianus In Ar: meta. 105.25-30: Lloyd (1985) has argued
nicely that parlurphistamenon. which is also used to describe the relation of' the incorporeal. place. to its occupying body. signals strong dependence. a parasitic relation
between the lekton and rational impression.
DL 7.56: cf. 7.55 (331-1+).
54 Reading en touto(i) adverbially in the sense Drat that point in time place (i,S.I IV). corresponding to enséntainesthai in 20, confirmed by ekeithen (thence) in 20, and hothen
in 18: attributive position signals that they become semantic at that very point: I am
grateful to David Crane for discussion of this passage.
55 1 disambiguate logos as internal or external (i.e.. public) discourse based on context:
the salient contrast is between discourse arising in thought. on the one hand. and voice
on the other: since voice. which is clearly public, has its source in discourse - it must
be some other kind, namely internal discourse: therefore. on the assumption that being
a source is transitive, the source of voice is the commanding faculty. where thoughts
happen and meanings come to be.
56 Translation Long & Sedley with modifications: for argument that ptöseis should be
considered elements of lekta, alongside predicates. see M. Frede (1987): cf. Long
(1971) for the opposite position.
57 And the nonchalance of Barnes' approach hardly makes the cost easier to bear: aller
summarily dismissing five different texts unfriendly to his view, he offers no more
defense than the following imaginary exchange in a footnote: "'So we should discard
most of the texts which inform us about Stoic sayables?' -Yes. - 'Surely that is not
methodologically defensible?' -Yes; it is."
58 i am indebted to Victor Caston. Jean-Baptiste Gourinat. and A. A. Long for helpful
conversations about this material, as well as to Verity 1-larte and my fellow participants
in the JHP 2014 Master Class.
Bibliography
Annas, J. (1990) Stoic epistemology. In Everson. S. (ed.) Epistemology. Companions to
Ancient Thought, 1. Cambridge, MA: Cambridge University Press.
. (1992) Hellenistic Philosophy of Mind. Berkeley: University of California Press.
Barnes. J. (1978) Principles of stoic grammar. in Rist. J. M. (ed.) The Stoics. Berkeley:
University of California Press. pp. 27-75.
. (1993) Meaning, saying and thinking. In During. K., and Ebert. T. (eds.) DialeI.
ktiker und Stoiker: Zut Logik der S1oa und ihrer• 1'or•lriufer; Philosophie der :Antike.
Stuttgart: F. Steiner.
. (1999) Meaning. In Algra. K., Barnes, J., Mansfield, J.. and Schofield. M. (eds.)
Hellenistic Philosophy. Cambridge, MA: Cambridge University Press.
Barnouw. J. (2002) Propositional Perception: Phantasia, Predication, and Sign in Plato,
Aristotle, and the Stoics. Lanham, MD: University Press of America.
Brittain, C. (2002) Non-rational perception in the Stoics and Augustine. Oxford Studies in
Ancient Philosophy 22: 253-308.
233
VANESSA DE HARVLN
Bury. R. G. ( 1967) Sextus Empiricus: Against the logicians. Vol. 2. Cambridge. MA: Loch
Classical Library. I larvard University Press.
Caston. V. (Unpublished) The Metaphysics. of Stoic Representation. Presented at The
Metaphysics of the Stoics: Causes. Principles. and Mereology. Corpus Christi Colle` e
Oxford. June 23, 2015.
—
. (1999) Nothing and something: The stoics on concepts and universals. Oxford
Studies in Ancient Philosophy 17: 145-213.
RATIONAL IMPRESSIONS AND STOIC PHILOSOPHY
Lukasiewiez..1. (1951) Aristotle s Syllogistic from the Standpoint ofModern Logic. Oxford:
Clarendon Press.
elates. B. (1961) Stoic Logic. Berkeley: University of California Press.
ncient Philosophy 17:
Mean. S. (1999) "The stoic theory of categories. Oxford Studies in Ancient
715_247.
Nawar. T. (2014) The stoic account of apprehension. Philosopher's Imprint 14 (29): 1-21.
Nucheimans• G. (1973) Theories of the Proposition. Amsterdam: North-llolland Publish-
de Harven, V. (2015) f low nothing can be something: "The stoic theory of void. Ancient
Philosophy 35 (2): 1-25.
Dyson, H. (2009) Profepsis and Ennoia in the Early Stoa. Berlin: Walter de Gruyter.
Frede, M. (1987) Stoics and skeptics on clear and distinct impressions. In Essays in Ancient
ing Co.
Perin. C. (2005) Stoic epistemology and the limits of externalism. Ancient Philosophy 25
. (1994) The stoic notion of a - lekton•. In Everson. S. (ed.) Language. Companion
s
to Ancient Thought: 3. Cambridge, MA: Cambridge University Press.
sical Library. 1-Iarvard University Press.
Ramelli. 1.. and Constan. D. (2009) Hierocles the Stoic: Elements of'Ethics. Fragments and
Excerpts. Atlanta. GA. USA: Society of Biblical Literature.
Philosophy. Minneapolis: University of Minnesota Press.
Gill, C. (2006) Epictetus: A new subjective-individualist self? in The Structured Self in
Hellenistic and Roman Thought. Oxford: Oxford University Press.
Gourinat, 1. B. (2000) La dialectique des stoïciens. Paris: Vrin.
. (2013) Le discours intérieur de Fame dans la philosophie stoïcienne. Chóra:
Revue d'études anciennes et médiévales 11: 11-22.
Graeser, A. (1978) The stoic theory of meaning. In Rist, J. M. (ed.) The Stoics. Berkeley:
University of California Press.
I-licks, R. D. (1925) Diogenes Laertius•: Lives of Eminent Philosophers. Cambridge, MA:
Loeb Classical Library, Harvard University Press.
Imbert, C. (1978) Théorie de la représentation et doctrine logique dans le stoïcisme ancien.
in Brunschwig, J. (ed.) Les Stoïciens et leur logique. Paris: Vrin.
Inwood, B. (1984) Hicrocles: Theory and argument in the second century AD. Oxford
Studies in Ancient Philosophy: 151-184.
. (1985) Ethics and Human Action in Early Stoicism. Oxford: Oxford University
Press.
Knealc, W., and Kneale, M. (1962) The Development of Logic. Oxford: Oxford University
Press.
Lesses, G. (1998) Content, cause and stoic impressions. Phronesis 43 (I ): 1-25.
Lloyd, A. C. (1985) Parhypostasis in Proclus. in Boss. G., and Seel, G. (eds.) Proclus et
son influence, actes du colloque de Neuchâtel. Zürich: GMB Editions du Grand Midi.
(2): 383-401.
2: 21-38.
Quille. W. V. O. (1948) On what there is. The Review of Metaphysics
Cambridge.
MA: Loeb ClasRackham. i i. (1967) Cicero: De .Vautra Deor•um.:lcademic•a.
Reed. B. (2002) The stoics' account of the cognitive impression. Oxford Studies in Ancient
Philosophy 23: 147-180.
(ed.) Problems in Stoicism.
Rist. J. M. (1971) Categories and their uses. In Long. A. A.
London: The Athlone Press.
Classical
Sandbach. F. H. (1930) Ennoia and Prolepsis in the stoic theory of knowledge.
Quarterly 24: 44-51.
Schaffer. J. (2010) Monism: The priority attic whole. Philosophical Review 119 (I ): 31--76.
27: 255-275.
Sedley. D. (1982) The stoic criterion of identity. Phronesis
(1993) Chrysippus on pscyhophysical causation. In Brunschwig. J.. and Nussbaum.
M. (eds.) Passions and Perceptions. Cambridge. MA: Cambridge University Press.
Sorabji. R. (1988) Matter, Space and Motion. Ithaca: Cornell University Press.
(1990) Perceptual content in the stoics. Phronesis 35 (3): 307-314.
su
Striker, G. (1974) xptttjptov ttjç à)al0ciaç. Nachrichten der Akadem ie der I fïssenshotften
Göttingen, I. Phil-hist: Klasse 2: 48-100.
Vogt, K. (2009) Sons of the earth: Are the stoics metaphysical brutes? Phronesis 54 (2):
136-154.
Teubner.
von Arnim, H. (1905) Stoicormn i eterum Fragmenta. Stuttgart: BG
Long, A. A. (1971) Language and thought in stoicism. in Long, A. A. (ed.) Problems in
Stoicism. London: The Athlone Press.
. (1982) Soul and body in stoicism. Phronesis 27: 34-57.
. (1991) Representation and the self in stoicism. In Everson. S. (ed.) Psychology,
Companions to Ancient Thought: 2. Cambridge, MA: Cambridge University Press.
. (1993) Hierocles on 'oikeiosis' and self-perception. In Boudouris, K. (cd.) Hel-
lenistic Philosophy. Athens.
. (1999) Stoic psychology. In Algra, K., Barnes, J., Mansfeld, J., and Schofield, M.
(eds.) Hellenistic Philosophy. Cambridge, MA: Cambridge University Press.
. (2006) Stoic psychology and the elucidation of language. In From Epicurus to
Epictetus: Studies in Hellenistic and Roman Philosophy. Oxford: Oxford University
Press.
Long, A. A., and Sedley, D. N. (1987) The Hellenistic Philosophers. Cambridge, MA:
Cambridge University Press.
234
235