Download the tension between aristotle_s theories

1
THE TENSION BETWEEN ARISTOTLE'S THEORIES AND USES OF
METAPHOR
[Artículo publicado en Studies in History and Philosophy
of Science, vol. 28, nº 1, pp. 123-139, 1997]
Alfredo Marcos
Departamento de Filosofía, Universidad de Valladolid,
Plaza de la Universidad, 1,
47002-VALLADOLID, Spain
0. Introduction
Today the philosophy of science commonly acknowledges
the cognitive and scientific role of metaphor1. Moreover,
metaphor is beginning to be treated linguistically as a
cognitive phenomenon2, while a flourishing tradition exists
in
rhetorical
studies
advocating
the cognitive and
communicative relevance of metaphor and other tropes in
different discursive contexts3.
There
is
still,
however,
an
on-going
debate
concerning the role of tropes in scientific discourse and
their compatibility with a realistic theory of science. In
this regard, Aristotle's work is one of the most
interesting case-studies in history from which some light
might be thrown on this debate, as there is an unavoidable
three-way tension between his methodological claims,
rhetorical and literary theories and scientific practice in
his texts. In his biological works, for example, there are
many
instances
of
explanatory
resources other than
definition followed by deduction, while metaphor, simile,
analogy and model are used throughout. Yet Aristotle, as a
natural philosopher, never abandons a realistic point of
view, while it was he who developed the first explicit
theory of metaphor. Despite the ubiquitous presence of
metaphorical expressions in his biological texts, in some
(e.g., APo4) he denies the cognitive functions of metaphor.
We are therefore justified in demanding a new, more
consistent, interpretation of the relevant Aristotelian
passages, and an explanation of their relevance to some
contemporary philosophical problems.
The approach will be as follows: firstly, I shall
comment on the texts where criticism is made of the use of
metaphorical expressions in science (section 1). The second
section deals with the actual use of metaphors, similes,
analogies and models in Aristotelian biology, examples
being used to bring out their extent and function. In
section three I shall attempt to counter this assumed
opposition by means of a new interpretation of the
Aristotelian theory of metaphor and APo methodology, for
which a delimitation of metaphor, simile, analogy and model
is
required,
as
is
a discussion of their mutual
relationships (3.1). We shall then be in a position to
suggest a reading which reduces the perennial tension
between different Aristotelian treatises (3.2). At that
point, a comment on the role of APo in the economy of
scientific explanation would seem to be in order (section
4). I shall end with a conclusive summary (5.).
1. The Assumed Refusal of Metaphor
It would be easy to establish, by means of examples,
the practical significance of metaphorical images in
natural history and metaphysics. Difficulties arise,
however, from the theoretical and methodological points of
view, owing to the puzzling nature of Aristotle's explicit
statements about the cognitive value of metaphor5.
When Empedocles took the salty ocean to be the
Earth's sweat, Aristotle judged this figure as unfit for
the understanding of the nature of things and suitable only
as ornament6. He also criticized other comparisons made by
Empedocles7, by the other presocratic philosophers8 and by
Plato9.
We can easily accommodate these texts by considering
that the criticism affects only
particular instances of
metaphor by virtue of their lack of appropriateness or
accuracy, due to their obscureness or emptiness and their
merely ornamental character, but not for being metaphors,
which is to say that Empedocles, for instance, is the
target of Aristotle's criticism, not for using metaphors,
but for using bad ones10. So, all that is important here is
that criticism is levelled at the quality of the figures,
not at their metaphorical nature.
In
APo,
however,
we
may
find
more
explicit
condemnation of metaphor in science11:
"We may add that if dialectical disputation must
not employ metaphors, clearly metaphors and
metaphorical
expressions
are
precluded
in
definition: otherwise dialectic would involve
metaphors"12.
On the basis of these texts, subsequent tradition
abolished the cognitive dimension of the Aristotelian
theory of metaphor13, although it might be a historical
mistake to blame Aristotle himself for such a loss on the
grounds of these passages alone. Aristotle was the first
philosopher to assert the importance of the role of
metaphor in the growth of knowledge and language14. In my
opinion, the Aristotelian theory of metaphor should be
placed near the current cognitive theories of metaphor
(such as the formular view, intentionalism, intuitionism,
contextualism and interactionism) and in opposition to such
non-cognitive theories as emotivism15. Let us then comment
on another set of Aristotelian passages where extensive use
is made of tropes, and their cognitive value is asserted.
2. The Actual Use of Metaphors, Similes, Analogies and
Models in Aristotelian Biology
Metaphor, simile, analogy and model appear throughout
Aristotle's biological treatises16, the following occurring,
3
for example in PA: the blood vessels and heart are compared
to vases17, the flow of blood is like water in an irrigation
channel18, the belly resembles a manger from which the body
takes its nourishment19, the warm area of the heart is like
the acropolis for a polis, or the fireplace for a home20.
The term "concoction" (pepsis), the centre of Aristotle's
"thermodynamic" physiology, is a metaphorical one21. Some of
these figures, as P. Louis22 points out, are taken from
Plato's Timaeus, while the origin of some others is even
older23, for example, male and female in generation are
compared respectively to the Sun and Earth24, a figure going
back at least as far as Hesiod. He also compares the male
principle in generation, as an efficient cause, to a
carpenter or a potter25. For Aristotle, even the development
and motion of a living being could be seen as the motion of
marvellous automata26.
Aristotle often uses elements taken from everyday
life and work, especially from fishing and sailing, such
activities obviously being familiar to the Greek people, so
a quadruped's legs are seen as the supports of a ship in
dry dock27, the back legs of a grasshopper are like the
rudder of a boat28, and a lobster's tail is like an oar29.
The elephant's trunk is likened to a diver's breathing
tube30, and the neck and beak of certain long-legged birds
are seen as a fishing-rod with line and hook31.
It is worth noting that all of them refer to the
functions of organs or tissues and they seek to explain
these functions by means of an analogy with artificial
objects and their well-known goals32.
The treatise De Anima is built upon a broad set of
similes and metaphors, all used to explain the most
difficult doctrinal points33: the unity of body and soul is
conceived as the unity of a circle and its tangent at a
point34, and as the unity of a wax tablet and the image
stamped upon it35. Concerning the body's instrumental
relationship to the soul, Aristotle says that the body is
to the soul as the eye is to sight, as the axe is to
cutting36. The active nous is like light37, while the passive
nous is compared to a blank board38. Other comparisons are
called upon for vegetative and sensitive faculties, memory,
imagination and will, the principle of movement, and
practical understanding. In conclusion, without metaphor,
there would be no De Anima at all.
Even notions so central to Aristotelian theories as
nature, soul or act are explained by means of analogies,
similes or metaphors: nature is thought of as a potter39, as
a house builder40 or a painter41, and so on. Regarding the
concept of act, so important throughout Aristotle's works,
the author states that only by analogy shall we be able to
grasp its meaning42.
The use of models is also noticeable as an
explanatory resource in Aristotelian biology, for example,
the human body is taken as a model for the study of other
living things43. Generally speaking, the conformation of a
domain following the structure of a phenomenon already
understood, i.e., the use of a familiar template, is an
important step in the explanation of anything belonging to
it. This procedure implies a flow of meaning from the
familiar to the new and strange. A flux of affections also
occurs, for, at least in Aristotelian biology, as well as a
deep knowledge, there is a very clear and profound esteem
for living beings44.
3. The Aristotelian Theory of Metaphor
3.1.
Differences and Relationships
Simile, Analogy and Model
between
Metaphor,
In Aristotle's opinion, metaphor and simile are very
similar:
"The
simile
(eikòn)
also
is
a
metaphor
(metaphorá); the difference is but slight. When
the poet says of Achilles that he
Leapt on the foe as a lion,
this is a simile; when he says of him 'The lion
leapt', it is a metaphor."45
Simile, or comparison, establishes a relationship
between two explicit terms by means of an explicit
grammatical connection, such as "like", while in metaphor
this grammatical connection is omitted, as often is one of
the two terms:
"[Simile] is a metaphor, differing from it only in
the way it is put; and just because it is longer
it is less attractive. Besides, it does not say
outright that 'this' is 'that', and therefore the
hearer is less interested in the idea."46
Despite their similarity, there are relevant æsthetic
and functional differences between these figures. Metaphor
poses a question, it surprises us, it triggers off a
heuristic process, it forces an interpretative task onto
us. Simile, on the other hand, does part of the work for
us, as we need a shorter interpretative run, the effort
probably yielding a lesser intellectual reward than the
metaphor. Aristotle consequently prefers metaphor to
simile, which he considers as a developed metaphor.
It should, by the way, be noted that this development
is not always automatic, for we compare certain aspects of
two objects, so a decision is required as to which aspects,
among all those possible, are relevant.
We should add to this that if we use a metaphor to
obscure our discourse, then it will lack any justification
in scientific texts. It should not be a means of expressing
obscurely what can be said plainly, but rather a way of
expressing difficult matters as clearly as possible, a
manner of stretching language into new areas of reality,
which is, furthermore, why Aristotle said that "[m]etaphor,
5
moreover, gives style clearness
distinction as nothing else can"47.
(saphès),
charm,
and
Analogy would seem to be another step in the
development of metaphor, so "the evening of the life..." is
a metaphor, "the old age is like the evening" is a simile
and "As old age is to life, so evening is to day" is an
analogy.
Aristotle discovered various ways of transferring
names,
which
enabled
him
to draft a taxonomy of
metaphorical expressions, whereby it is possible to use the
name of the kind (génos) to name the species (eîdos), the
name of species to refer to the kind, the name of one
species may be applied to another or, finally, the name may
be transferred "on grounds of analogy"48. We commonly
associate the last mentioned with metaphor in a narrow
sense, the others being more like our synecdoche or
metonymy49.
Aristotle holds that there is proportional analogy
whenever "the second (B) is to the first (A) as the fourth
(D) to the third (C)"50. Originally proportion referred
mainly to quantitative ideas, but qualitative ones were
included early on: "Thus a cup (B) is in relation
to
Dionysus (A) what a shield (D) is to Ares (C) [...] As old
age (D) is to life (C), so is evening (B) to day (A)"51.
With regard to the cognitive relevance of figures and
their explanatory power, Aristotle wrote that "[o]f the
four kinds of metaphor the most taking is the proportional
kind (kat' analogían)"52, a doctrine which does not mean,
however, that a paraphrase from metaphor to analogy could
always be easily attained only by making the concealed
analogies explicit. We would stress once more that
understanding
new
metaphors
often
requires
an
interpretative effort. This heuristic task yields the
poietic discovery of new analogic relationships. Every good
metaphor is followed by what might be called a heuristic
inertia. The development of metaphor into analogies, then,
requires the concurrence of all the general intelligence.
Aristotelian
texts
have
been
very
useful
for
understanding the mutual differences and relations of
metaphor, simile and analogy and, to this end, I have kept
close to them. We cannot dismiss the fact, however, that
Aristotle used some models in his biological treatises,
using for example the human body as a model to explain
other living things, as we have seen above, or artifacts in
the explanation of natural things. Nevertheless, he does
not theorize about models, which means that his texts are
of no help to the understanding of the role of models and
their relationships with other explanatory resources. We
therefore have to look elsewhere to throw light on models,
even at the risk of temporarily abandoning the historical
context we are dealing with. In my opinion, some of the
ideas put forward by Lakoff and Indukhya may be extremely
useful in the comprehension of the functions of models and
their relationships with other figures53.
In his "Cognitive Semantics", Lakoff states that we
elaborate abstract models by means of schematization,
categorization, metaphor and metonymy. He believes that
these processes all require the use of a projective
imagination based on our structured experience (through
bodily, social and cultural factors) and our innate
sensorimotor faculties and activities, and notes some
general schemes based on our experience, which, owing to
their experiential basis,
bring us meaning in a direct
way: the container,
part-whole,
link, source-path-goal,
up-down and linear order schemata.
When faced with the unfamiliar, we react - perhaps
without full awareness - by trying to reduce it to the
basic experiential schemes that we know how to handle.
Indeed, it is an essential factor of our everyday
understanding and communication54 and, of course, of
scientific enquiry. We can thus grasp the explicative power
of the very structure of biological treatises insofar as it
fits some of these schemes. PA and HA were written in a
top-downwards order, based on the opposition between
internal and external organs, on the container model, in
such a way that the very arrangement of data has an
explanatory effect. This is how we begin to assimilate the
nature of living things.
Other models have an even more important role
regarding the explanatory economy of biological treatises,
for Aristotle adds a profuse philosophical reflection to
their direct understanding as based on experience. Such is
the case of the source-path-goal schema, which has an
evident relationship with the philosophical topic of telos,
and
of
the
part-whole
schema,
which
concerns
the
hylemorphic structure attributed by Aristotle to living
beings, as well as
the careful study of their parts
(meros)55. It should come as no surprise, then, that
Aristotle should find these models so useful and powerful
in explaining living things.
It should be noted that these schemes, when applied,
are not entirely determined by their bases, but rather are
submitted to criticism. Thus, for example, the final cause
in biology does not involve awareness of the purpose, or
deliberation, despite being drawn from our everyday
intentional experience.
Indurkhya56 suggests a gradation ranging from pure
metaphor, through simile and analogy, to models, for, in
his view, models are already outside the domain of
metaphors
because
their
interpretation
is
entirely
conventional.
Following
on
from
Indurkhya,
we
can
appreciate other gradations, whereby metaphor is less
symmetrical than simile, simile less than analogy, and
analogy less than model, besides which metaphor is more
emotionally charged than the other figures.
7
Indurkhya uses the notion of metaphor in an
intermediate way between the narrow and broad senses. In
the narrow sense, metaphor "refers to a specific way of
using the words and phrases of a language", while in the
broad
sense
"it
is
applied
to
the
process
of
conceptualization itself, leading to the aphorism 'all
thought is metaphorical'"57.
We are interested in a very broad meaning of
metaphor, including models and even any general concept,
for, in Aristotle, metaphor is equal to transfer, which is
involved by pure metaphor, as it is by simile, analogy,
model and even general concepts, despite their gradual
differences,
which
are
secondary
in
epistemological
discussion to common traits. In particular, it is possible
to detect a genetic relationship between those figures
requiring a more metaphorical reading and the more
conventional ones, that is, there are expressions born as
pure metaphors which later become plain conventions. On the
other hand, even the most established convention requires
interpretation to some extent, a metaphor still lingers
even in the most literal concept. With regard to this idea,
Aristotle wrote: "Metaphor consists in giving the thing a
name that belongs to something else"58, and "[i]nevitably,
then, the same formulæ, and a single name, have a number of
meanings"59. The only way a common name can reach this
plural signification is by becoming transferable itself.
3.2. Metaphorical Figures as Creative Discoveries60
In his Rhetorics, Aristotle states:
"[...] we all naturally find it agreeable to get
hold of new ideas easily: words express ideas and
therefore those words are the most agreeable that
enable us to get hold of new ideas. Now strange
words simply puzzle us; ordinary words convey
only what we know already; it is from metaphor
that we can best get hold of something fresh (he
dè metaphorá poieî toûto málista). When the poet
calls old age 'a withered stalk', he conveys a
new idea, a new fact, to us by means of the
general notion (dià toû génous) of 'lost bloom',
which is common to both things. The similes of
the poets do the same, and therefore, if they are
good similes, give an effect of brilliance"61.
Let us comment on some salient features of this
passage. After this text, no doubt could remain about the
cognitive
purport
of
metaphor
and simile, although
Aristotle does stress that in order to be cognitive, they
must fulfil certain requirements, that is, they must be
proper.
Secondly,
we
are
informed
that
teaching
is
accomplished by means of the kind (dià toû génous), when an
objective similarity hits one in the eye. The kind is but a
means of gaining knowledge - it is not the final purpose of
knowledge. Showing that two entities are similar in some
way, that they belong to the same kind, enables us to
transfer our knowledge of the more familiar one to the
other, thus affording us a better understanding of the new
or inexperienced. This transfer must, however, be subject
to the filter of critical scrutiny to avoid improper uses.
Thirdly, Aristotle unites the æsthetic and cognitive
aspects of an expression. In his Rhetorics62, he also
asserts that learning and admiring are sources of
pleasure63.
What does Aristotle mean by a proper metaphor or
comparison? We may recall here the passage from Poetics64
defining four types of metaphorical expression bearing in
mind that he goes on to say that "of the four kinds of
metaphor the most taking is the proportional kind". It is
therefore clear that an image is proper insofar as it is
based upon an objective proportional analogy and expresses
a real similarity allowing us the information transfer from
one pole to the other.
What, then, became of the creative aspect of
metaphor? Did it turn out to be a mere discovery? Is this
kind of knowledge not simply a mirror of nature?
The concept of creative, or poietic, discovery is
used by Haley65 as an intermediate between the traditional
and interactionist views of metaphor. According to the
former, true metaphor is just a discovery of underlying
similarities, where the cognitive subject has a rather
passive
function
it
is
a
mirror
of
nature.
Interactionism, on the other hand, proclaims metaphoric
creativity, with a subject that creates a web of
connections,
organizing
reality
in
an
active
way.
Nevertheless, this view fails to provide a clear account of
the constraints affecting the creation, interpretation and
evaluation of figures. Indurkhya is also aware of this
shortcoming and seeks to solve it. In my opinion, however,
finding a solution to this problem depends on the
acknowledgement of
the objective pole, that is, real
similarities that one can either discover or fail to
discover. Yet nothing in the expression itself allows for
mechanical decoding, for a metaphor works or not according
to the interpreter, to his background, and his creativity
in building conjectures. It also depends on the world
itself, on the potential (but real) similarities between
entities dwelling in it. What then, could possibly
constitute a creative discovery?
We
shall
see.
Similarities
uncovered
by
true
metaphorical expressions are real. There are objective
constraints existing as possibilities in entities - any two
entities either have or do not have the potential to be
seen as similar in some respect by a given cognitive
subject. We cannot, however, rest on any special intuitive
faculty for similarities. The potential for objects to be
seen as similar cannot be actualized or communicated
without an active subject66. In the first place, we need to
9
invent conjectures or hypotheses and set them up against
the facts. In this way, we are able to descry new
resemblances between objects. On the other hand we can also
try to communicate them by means of a metaphorical
expression, that is, by building new language or stretching
the semantic range of existing language. To construe a
metaphor, however, the receiver needs to display the same
creative attitude as we have before nature. It is in this
sense that metaphor is just as much a discovery as a
creation. It may rightly be called, then, a creative, or
poietic, discovery.
The expression "creative discovery" is not explicitly
mentioned in Aristotle's works, though I would not consider
it anachronistic to say that its meaning may be inferred
from several passages, for example:
"Metaphors must be drawn, as has been said
already, from things that are related to the
original thing, and yet not obviously related just as in philosophy also an acute mind will
perceive resemblances even in things far apart"67.
Therefore,
"the greatest thing by far is to be a master of
metaphor. It is the one thing that cannot be
learnt from others; and it is also a sign of
genius [...]"68.
Spotting resemblances for the first time requires the
invention of new points of view, of new interpretative
hypotheses, of new and fallible conjectures. Similarity is
not that which is at the same time in two different places
or substances, but that which can be abstracted from both
by a cognitive agent. Consequently, similarity is not a
direct or ontic relationship between two or more objects,
as all dynamic actions are, but one established by means of
a subject69. In spite of the objective character of
potential similarities, there are no actual ones unless
they are established by a cognitive subject.
We very often find that a good metaphor, because of
its creative nature, seems unpredictable yet, owing to its
character of objective discovery,
it appear obvious to
nearly everybody once enunciated. Thus, Aristotle said that
metaphor gave greater clarity than anything else could70 and
makes us see71. Metaphor, Aristotle states, brings our
senses face to face with reality: "I mean using expressions
that represent things as in a state of activity (ósa
energoûnta semaínei)"72.
Nevertheless, the fact remains that passages in
favour of a cognitive view of metaphor are found in
contexts far removed from the methodology of science, while
it is precisely in APo that we find the most explicit
refusal. On this point, some remarks will be necessary.
4. APo and Aristotelian Science
To begin with, we
poetics with rhetorics, for
aimed purely at convincing,
paradigmatically creative
Aristotle clearly states
poetry74.
should not carelessly equate
the latter deals with discourse
while the former has to do with
discourse73. In his Poetics,
the philosophical nature of
Secondly, it should be remembered at this point that
the taxonomy of metaphor is coined in terms of "species",
"kind", "proportion", all of them belonging to the very
core of philosophy, biology and even Aristotelian science
in general75.
In the third place, there is some evidence to show
that Aristotle does not confer such great importance on the
methodological rules of the APo while doing empirical
science, so he never took it as a rigid set of constraints
for scientific practice. We have already seen that in DA he
uses more comparisons than definitions whenever he finds a
problematic node, the same being the case for the other
biological treatises, where functional aspects of organisms
are systematically explained by means of metaphors, similes
and analogies, even the very structure of treatises being
thought out on the basis of models. In addition, however,
in his Metaphysics, Aristotle states that the important
concept of act lacks a proper definition76, while individual
substances lack both definition and demonstration77, as well
as essence78. We are informed in GA that we cannot expect a
demonstration of principles, so they require "another
method (álle gnôsis)"79.
Let us now examine two less-known passages, where
Aristotle rules out explanations precisely on the grounds
of their logical character, that is, because, being too
general, they are closer to the logos than to the
particular object of research:
"Perhaps an abstract proof (apódeixis logikè)
might appear to be more plausible than those
already given; I call it abstract because the
more general it is the further is it removed from
the special principles involved. [...] For all
theories not based on the special principles
involved are empty; they only appear to be
connected with the facts without being so really.
[...] that which is empty may seem to be
something, but is really nothing."80
"And we must grasp this not only generally in
theory, but also by reference to individuals in
the world of sense, for with these in view we
seek general theories, and with these we believe
that general theories ought to harmonize"81.
The logical apparatus of definition and demonstration
does not work properly unless a connection is provided
between theoretical terms and our experience of concrete
reality. The judgement about truth of principles used as
premises in deduction, the ascription of reference to the
terms, the knowledge of causal connections concealed behind
11
logical ones, all remain outside the logical apparatus of
APo.
This set of methodological rules acquires its full
meaning once the above operations have been carried out
and, even then, the outcome of deductive machinery is
subject to common-sense scrutiny and personal experience.
We must now address the problem of determining
whether or not APo still plays a role in the economy of
scientific explanation.
Aristotle often uses terms in a non-univocal sense,
subjecting them to semantic stretch82, whereby they acquire
new meanings. He starts with a focal familiar meaning and
applies the terms analogically. Important notions are said
in "many ways" (pollakôs legômenon), so we are justified in
suspecting that the ideal of strict univocity of APo soon
became unfit for actual science in Aristotle's view83.
Indeed, univocity could only be reached
either by
increasing the complexity of language or by decreasing the
complexity of the world. The first solution would make
language a useless point-by-point reproduction of the world
- one object, one word, like a map drawn to a scale of
1:184, while in the second hypothesis, the world would
include but few entities, one for each concept, that is,
the real world would be like Plato's ideal world.
Nevertheless, in Aristotle's view, language is inexact
because of the very nature of things - there are more
entities than terms85. I consider this passage ontologically
revealing, because the problem arises only when the
ontological load is born more by concrete substances (as in
Aristotle) than by ideal forms. In such cases, name
transfer seems unavoidable.
Furthermore,
methodological
terms
like
"demonstration" or "definition" are also subject to
semantic stretch, so they do not mean exactly the same in
different research contexts86.
Nevertheless, the methodology of APo
still retains
a relevant role (paradoxically) as a rhetorical device. It
is an important part of the process of legitimizing new
wisdom, opposed to traditional wisdom, which rested on
story-telling, poetry and myth. The new wisdom, on the
other hand, sought for accuracy and stability in the
meaning of words, univocity as an ideal, and deductive
inference:
"The polemic against metaphor and myth - as G.E.R.
Lloyd writes - is thus part of the campaign waged
by philosophy and science against poetry and
religion"87.
This demarcation is not only sociological, however.
Nor is it only a dispute between groups aiming at social
control or intellectual prestige. There is more to it than
that. The methodological prescriptions of APo may never
have been strictly followed, metaphorical resources against prohibition - may actually have been used in
scientific practice and methodological constraints may have
been softened in various treatises, but something did
change: once Aristotle had established the desideratum of
stability and univocity in the meaning of terms, together
with the ideals of exactitude, definition and deductive
control of inference, the philosopher and the scientist
were affected by them in a way that the poet or storyteller never had been before. Scientists and philosophers
must submit their metaphors, similes, analogies and models
to criticism or empirical control, and must follow the
heuristic inertia of images and test their implications.
They must set them up against reality. Once the will of
truth, the ideal of precision, the necessity of exploring
implication as well as a self-critical mind were accepted
as
values,
the
particular
formalist
devices
and
methodological constraints of APo became secondary to
actual scientific practice.
All these values are acknowledged and pursued in
Aristotle's works. His De Caelo explicitly states:
"To give a satisfactory decision as to the truth
it is necessary to be rather an arbitrator than a
party to the dispute"88.
5. Conclusion
Metaphor is not a superfluous ornament nor is it
mechanically reducible, no universal rules existing for its
translation into literal language. Understanding metaphor
does not, however, depend on any special intuitive faculty.
I have suggested reading Aristotelian texts with what may
be termed an interpretative view of metaphor, for
understanding a metaphor involves an interpretative task
very similar to understanding the world. Every good
metaphor has its own heuristic inertia. We need to create
conjectures and challenge their functionality, a process
which depends on the interpreter's creativity, on his
ability to recognize their limits and on his previous
experience and knowledge. No automatic rules, no special
faculties.
Moreover, the very notion of literal meaning is
problematic. Aristotle refers to conventional (usual)
rather than to literal meaning in is study on different
kinds of names89.
In short, metaphoric meaning is not
opposed to literal meaning, although it does gradually
differ from conventional meaning90. An expression conceived
as a metaphor could be developed into a simile or analogy,
indeed it might even become conventional, while, conversely
any conventional expression could be hiding a lethargic
metaphor.
It
may
therefore
be useful to look on
metaphorical expressions as living things: during their
lifetime; they can remain detached from any cognitive
discourse or become integrated as conventional language
into science or philosophy. Another consideration should be
added with regard to the importance of the topic we are
dealing with: both the increase in our knowledge and the
enlargement of our linguistic resources depend on our
13
ability to grasp new similarities and express them in a
metaphorical
way.
Today's
conventional
language
and
knowledge
were
metaphorical
yesterday,
and
even
conventional language requires interpretation each time it
is
used
pragmatically.
As
Nietzsche
reminded
us91,
convention is no more than lethargic metaphor, so each
concept has to be understood in its context, from a
concrete subject's experience, and referred to the world by
him. Only in this way does language leave the domain of the
general to enter the terrain of concrete substances.
A good metaphor gives us more than emotional content
- it can transmit to us genuine information about the
world. The radical distinction between the genesis of
emotion and objective information transfer is artificial,
for metaphor explains by bringing in the unusual, new or
unknown to what is familiar or already experienced. Owing
to the fact that our experience is not emotionally neutral,
however, the flow of information runs together with
emotional connotations. Aristotle, for instance, in his
biological treatises, displays as much knowledge as esteem
regarding living things92.
Metaphorical
expressions,
apart
from
their
communicative or emotional virtues, can be either true or
false93 - a good metaphor uncovers objective potential
similarities existing between entities in the world, a
discovery which is, however, a creative one.
Between the claim that biological works (and natural
history
treatises
in
general)
strictly
follow
the
methodological constraints of APo, and the assumption that
it has nothing to do with them, we would suggest an
intermediate
position:
expressive
resources
such
as
metaphors rejected in APo
are often used in scientific
practice, although the way in which they are used is
determined by desiderata of precision, clear meaning of
terms, empirical tests and deductive control of inferences.
Acknowledgements
This text is part of a research project developed at
Valladolid University. A short version was read at the
Tenth International Congress of Logic, Methodology and
Philosophy of Science, Florence, 1995. I thank Professor
G.E.R. Lloyd very much for his insightful commentaries
during my stay at Cambridge University. I am also indebted
to an anonymous referee.
1
See, for instance, E. Bustos, 'Las metáforas científicas y
el realismo semántico', Arbor 542 (1991), 69-82; G.
Corradi, The Metaphoric Process (London: Routledge, 1995);
F. Fernández Buey, La ilusión del método (Barcelona:
Crítica, 1992), pp. 152-179; M. Hesse, Models and Analogies
in Science (Nôtre Dame: Nôtre Dame University Press, 1965);
M. Hesse, 'The Cognitive Claims of Metaphor', Journal of
Speculatice Philosphy (1988), 1-16.; M. Hesse, 'Theories,
Family-Resemblance and Analogy', in Helman (ed.),
Analogical Reasoning (Dordrecht: Kluwer, 1988), pp. 317340; M. Hesse, 'Models, Metaphors and Truth', in Z. Radman
(ed.), From a Metaphoric Point of View. A Multidisciplinary
Approach To the Cognitive Content of Metaphor (Berlin: de
Gruyter, 1995), pp. 351-372; J. Hintikka (ed.), Aspects of
Metaphor (Dordrecht: Kluwer, 1994); G. Holton, 'Metaphors
in Science and Education', in Z. Radman (ed.), op. cit.,
pp. 259-288; K. Knorr Cetina, 'Metaphors in the Scientific
Laboratory: Why are they there and what do they do?', in Z.
Radman (ed.), op. cit., pp. 329-349; S. Maasen, P.
Weingart, E. Mendelsohn (eds.), Biology as Society, Society
as Biology: Metaphors (Dordrecht: Kluwer, 1995); A. Marcos,
'Biología, realismo y metáfora', Agora (forthcoming); J.
Martin Soskice and R. Harré, 'Metaphor in Science', in Z.
Radman (ed.), op. cit., pp. 289-307; E. Montuschi, 'What is
Wrong with Talking of Metaphors in Science', in Z. Radman
(ed.), op. cit., pp. 309-327; A. Ortony (ed.), Metaphor and
Thought (Cambridge: Cambridge University Press, 1979); E.C.
Way, Knowledge, Representation and Metaphor (Dordrecht:
Kluwer, 1991); and
F. Suppe, The Semantic Conception of
Theories and Scienfic Realism (Urbana: Illinois University
Press, 1989). This author thinks that the language of
15
science is not literal, but
that science does aim at
objective truth (see op. cit. pg. 23).
2
From the field of linguistics see B. Indukhya, Metaphor
and Cognition (Dordrecht: Kluwer, 1992); M. Johnson, The
Body in the Mind (Chicago: University of Chicago Press,
1987); G. Lakoff, Women, Fire and Dangerous Things: What
Categories reveal about the mind (Chicago: University of
Chicago Press, 1987); G. Lakoff and M. Johnson, Metaphors
We Live By (Chicago: University of Chicago Press, 1980);
E.R. Mac Cormac, A Cognitive Theory of Metaphor (Cambridge,
Mass.: M.I.T. Press, 1985).
3
See, for instance, R. Bathes, Le degré zéro de l'écriture
(Paris: du Seuil, 1953); G. Genette, Figures III (Paris: du
Seuil, 1972); Groupe µ, Rhétorique générale (Paris:
Laruosse, 1970); P. Ricoeur, La Métaphore vive (Paris: du
Seuil, 1975); H. White, Tropics of Discourse: Essay in
Cultural Criticism (Baltimore: Johns Hopkins University
Press, 1978); H. White, The Content of the Form (Baltimore:
Johns Hopkins University Press, 1987); and specially on
rhetoric and science see B. Roland, 'Science versus
literature', in M. Lane (ed.), Introduction to
Structuralism (New York: Basic Books, 1970), pp. 410-416;
O. Gal, 'Tropes and Topics in Scientific Discourse:
Galileo's De Motu', Science in Context 7 (1994), pp. 25-52;
A. Gross, The Rhetoric of Science (Cambridge, MA: Harvard
University Press, 1990); D. Locke, Science as Writing (New
Haven: Yale University Press, 1992).
4
From now on, I will quote Aristotelian works using the
following abbreviations: HA:Historia Animalium; PA:De
Partibus Animalium; GA:De Generatione Animalium; DA:De
Anima; MA:De Motu Animalium; PN:Parva Naturalia; D Cael:De
Caelo; Meta:Metaphysica; Mete:Meteorologica; APo:Analitica
Posteriora; EN:Ethica Nichomachea; EE:Ethica Eudemia;
Pol:Politica; SE:Sophistichi Elenchi; Rhet:Rhetorica;
Poet:Poetica; Top:Topica. I take aristotelian texts in
their english translation from W.D. Ross and J.A. Smith
(eds.), The Works of Aristotle Translated into English
(Oxford: Clarendon Press, 1908-1952).
5
However, some authors have reduced the cognitive value of
metaphorical language in Aristotelian science on
theoretical as well as on practical grounds. So, for
instance, Pierre Louis downgrades the scope of metaphors in
biology acknowledging only a secondary role in explanation
(P. Louis, 'Introduction', in Aristote, Les parties des
animaux (Paris: Les Belles Lettres, 1956), pp. V-XL, see
pp. XXIX and XXX). He exhibits a position very similar to
the post-positivism of his age (see Bustos, op. cit., note
1, pp. 69-72).
M. Vegetti, in his turn, stresses the
rhetoric function of metaphorical figures in Aristotelian
biology (M. Vegetti, 'Quand la science parle à vide:
procédés dialectiques et métaphoriques chez Aristote', in
V. Coorebyter, Rhétoriques de la science (Paris: P.U.F.,
1994), pp. 7-32, see pp. 18 and ff.).
6
Mete 357a 24 and ff..
7
For example in GA 777a 7 and ff..
17
8
Top 127a 17 and ff.; GA 747a 34 and ff.; PN 347b 9 and
ff.; PA 652b 7.
9
Meta 991a 20, 1079b 24; Pol 1264b 4 and ff., 1265b 18 and
ff..
10
G.E.R. Lloyd, The Revolution of Wisdom: Studies in the
Claims and Practice of Ancient Greek Science (Berkeley:
University of California Press, 1987), see pp. 183-187.
11
We can find another unqualified refusal of metaphoric
figures in Top: "for a metaphorical expression is always
obscure" (139b 32 and ff.).
12
APo 97b 37-39.
13
See, as an explanation of this process, P. Ricoeur, op.
cit., note 3, pp. 63-86; see, specially, pp. 66-67.
14
E. Bustos, 'La teoría aristotélica sobre la metáfora', in
D. Sánchez Meca y J. Domínguez Caparrós (eds.), Historia de
la relación filosofía-literatura en sus textos, Suplementos
de Anthropos, vol. 32 (Barcelona: Anthropos, 1992), pp. 1721, see pp.
15
19-21.
For a clear characterization of these theories see I.
Scheffler, Beyond the Letter (London: Routledge and Kegan
Paul, 1979).
16
See G.E.R. Lloyd, Polarity and Analogy (Cambridge:
Cambridge University Press, 1966); E. Montuschi, Le
Metafore Scientifiche (Milano: Franco Angeli, 1993); E.
Martino, Aristóteles, el alma y la comparación, (Madrid:
Gredos, 1975); M. Vegetti, op. cit., note 5, pp. 7-32; G.A.
Luccheta, Scienza e Retorica in Aristotele (Bologna: Il
Mulino, 1990); S. Gastaldi, 'Teoria e Funzioni della
Metafore in Aristotele' in Dimostrazione, Argumentazione
dialetica e argomentazione retorica nel pensiero antico
(Génova: 1993); G. S. Bordoni, Linguaggio e Realtá in
Aristotele (Bari: Laterza, 1994).
17
PA 650a 32, 650b 8, 665b 12, 666a 18, 667a 26; cf. Timaeus
73d.
18
PA 668a 13, 27, 35; cf. Timaeus 77cd, 79a.
19
PA 650a 19; cf. Timaeus 70e.
20
PA 670a 25-26; cf. Timaeus 70a. Mario Vegetti analyzes the
functions of some of these similes. For example, a broad
analysis of the heard-acropolis-fireplace and body-polishome comparisons can be seen in Vegetti, op. cit. note 5,
pp. 19 and ff..
21
"Originally used of the ripening of fruit, then of cooking
and digestion it came to be applied to a wide range of
physiological processes", G.E.R. Lloyd, op.cit, note 10,
pp. 204-205.
22
P. Louis, op. cit., note 5, see pp. XXIX and XXX.
23
See, in this sense, Luccheta, op. cit., note 16.
24
GA 716a 15 and ff..
25
GA 730b 5 and ff..
26
See GA 743b 11 and ff.; MA 701b 2 and ff..
27
PA, 655a 11, 23, 25.
28
PA 683b 1 and HA 535b 12.
29
PA 684a 3.
30
PA 659a 2.
31
PA 693a 23.
19
32
In fact, we can find more intersting comparisons
throughout Aristotelian treatises. See, for instance, in PA
653a 3 and ff., 653a 21 and ff., 654a 7 and ff.. We could
also quote, mainly in virtue of its evocative force, the
resemblance supposed to exist between plants and little
babes sleeping all the time (EE, 1216a 3-9 and EN, 1176a
34-35).
33
See Martino, op. cit., note 16.
34
DA 403a 11-16.
35
DA 412b 6-9.
36
DA 412b 9 and ff..
37
DA 430a 10 and ff..
38
DA 429b 29 and ff..
39
PA 654b 29 and ff.; GA 730b 27 and ff..
40
PA 668a 16.
41
GA 743b 20 and ff.; 764b 30 and ff..
42
Meta 1048a 35 and ff..
43
HA, I, 6, 491a 19-23.
44
See PA, I, 5, 644b 22-645a 36.
45
Rhet, 1406b 20 and ff..
46
Rhet, 1410b 17-19.
47
Rhet, 1405a 8.
48
Poet, 1457b 6-9.
49
See Bustos, op. cit., note 14, pp. 19-20.
50
Poet, 1457b 16 and ff..
51
Poet, 1457b 20-25; see also EN 1131a 29-30.
52
Rhet, 1410b 36 and f.; see also Montuschi, op. cit., note
16, p. 31.
53
G. Lakkof, 'Cognitive Semantics', in U. Eco et al. (eds.),
Meaning and Mental Representation (Bloomington: Indiana
University Press, 1988), 119-154; B. Indukhya, op. cit.,
note 2.
54
See M. Johnson, The Body in the Mind (Chicago: University
of Chicago Press, 1987); G. Lakoff, Women, Fire and
Dangerous Things: What Categories reveal about the mind
(Chicago: University of Chicago Press, 1987).
55
See G.E.R. Lloyd, 'Aristotle's Zoology and his
Metaphysics. The status quaestionis. A Critical Review of
some Recent Theories', in D. Devereux y P. Pellegrin
(eds.), Biologie, Logique et Métaphysique chez Aristote,
(Paris: Éditions du C.N.R.S., 1990), pp. 7-35; P.
Pellegrin, 'Taxinomie, moriologie, division: réponses à
G.E.R.Lloyd', in D. Devereux y P. Pellegrin (eds.),
Biologie, Logique et Métaphysique chez Aristote, (Paris:
Éditions du C.N.R.S., 1990), pp. 37-47.
56
Indurkhya, op. cit., note 2, pp. 26 and ff..
57
Indurkhya, op. cit., note 2, p. 13.
58
Poet, 1457b 7.
59
SE 165a 13 and f..
60
Let me sketch out a remark here, mainly in the light of an
acute observation made by an anonymous referee: I am not
trying even to suggest that metaphorical figures belong to
the "context of discovery", while APo is the correct
methodology in relation to the "context of justification".
In APo, and generally in traditional methodology (there are
traces even in Mill or in the first Carnap), justification
21
comes hand in hand with some assumed method of discovery,
the "right" one for each author. I think, of course, that
no particular method of discovery is sufficient to
guarantee justification. My concern here is to explain how
some metaphors are the expression of a creative discovery,
and how they could help, with a new point of view or new
heuristic trends, to obtain new discoveries. But we can
only obtain a provisional and pragmatical justification by
following the suggestions of metaphor and contrasting them
with experience. This is not, of course, the tenor of APo.
61
Rhet 1410b 10-19.
62
Rhet 1371b 4 and f..
63
S. Mas Torres, 'Platón y Aristóteles: sobre filosofía y
poesía', in D. Sánchez Meca y J. Domínguez Caparrós (eds.),
Historia de la relación filosofía-literatura en sus textos,
Suplementos de Anthropos, nº 32 (Barcelona: Anthropos,
1992), pp. 5-10, see p. 8.
64
Poet, 1457b 6 and ff..
65
I take the expression "creative discovery" from M.C.
Haley, The Semiosis of Poetic Metaphor (Bloomington:
Indiana University Press, 1988), where a Peircian theory of
poetic metaphor is explained.
66
Even our natural ability to catch surface similarities has
phylogenetically evolved by means of creative activity and
corrections, as authors like Popper or Quine have pointed
out. See, for instance, K. Popper, A World of Propensities
(Bristol: Thoemmes, 1990); W.V. Quine, 'Natural Kinds', in
W.V. Quine, Ontological Relativity and Other Essays (New
York: Columbia University Press, 1969), pp. 114-138.
67
Rhet 1412a 10 and ff..
68
Poet 1459a 5 and ff.. See also PN (464b 5 and ff.), where
Aristotle wrote a beautiful passage on resemblance in
dreams in the same purport as the previously quoted ones
(it even contains a metaphor full of suggestions).
69
In this sense, Scaltsas affirms that "similarity between
substances cannot consist in the presence of a distinct
(abstract) component in different substances. Rather, it
consists in the derivation of the same distinct entity out
of different substances", in Th. Scaltsas, Substances &
Universals in Aristotle's Metaphysics (Ithaca: Cornell
University Press, 1994), pp. 197-8.
70
See Rhet 1405a 8 and ff..
71
See Rhet 1411a 25 and ff..
72
Rhet 1411b 24-26. The author is stressing the sensitive
aspects of understanding in this passage. Others exist in
the same direction, for instance, those that establish the
cognitive relevance of images: Aristotle affirms that we
take delight in our senses, "and above all others the sense
of sight" (Meta 980a 21 and f.), and that never does the
soul think without an image (DA 431a 14-17; PN 450a 1 and
f.). Understanding is compared to the soul's sight (EN
1096b 29), and, specially, active understanding to the
light (DA 430a 14-17). With regard to wise and prudent
persons (phrónimos) we can read: "for because experience
has given them an eye they see upright" (EN 1143b 11-13).
23
On cognitive functions of imagination for Aristotle, see
also M.V. Wedin, Mind and Imagination in Aristotle (London:
Yale University Press, 1988); on perception, D.K.W. Modrak,
Aristotle: The Power of Perception, (Chicago: University of
Chicago Press, 1987).
73
See P. Ricoeur, op, cit., note 3, pp. 13-61 and 310-311.
74
Poet 1451a 37.
75
According to E. Montuschi, metaphor "institutes a sort of
symbiosis between the logical and categorial apparatus of
philosophy, on the one hand, and, on the other hand,
the
transgressing effects that analogy systematically pursues",
Montuschi, op. cit., note 16, p. 30.
76
Meta 1048a 35 - 1048b 1. See also P. Ricoeur, op. cit.,
note 3, pp. 325-399.
77
Meta 1039b 28-30.
78
Meta 997a 31-33.
79
GA 742b 32 and f..
80
GA 747b 27 - 748a 10. M. Vegetti (op. cit, note 5, p. 8)
has recently commented on this passage in a different
sense. This author believes that "apódeixis logikè" is
criticized because of its dialectic character. However, in
my opinion, this text has nothing to do with the dialectics
vs. science distinction. Logical demonstration or
explanation is criticized precisely for its generality, for
its lack of empiric charge, for its emptiness, i.e., for an
excess of abstraction (as the aforementioned text on MA
confirms). But all that relates it more to strict episteme
than to dialectics. In addition, the commonly acknowledged
opinions (éndoxa) are not
necessarily very abstract or
general; there exist also reputed views on concrete
subjects.
81
MA 698a 11-14.
82
See Lloyd, op. cit., note 10, p. 198.
83
See Lloyd, op. cit., note 10, p. 191.
84
Not to mention the insoluble metalinguistic issue, because
words, of course, are objects.
85
See SE 165a 6-14 (see above in note 59).
86
See G.E.R. Lloyd, 'The Theories and Practices of
Demonstration in Aristotle', in J.J. Clearly and D.C.
Shartin (eds.), The Boston Area Colloquium in Ancient
Philosophy, vol. VI. (Lauham, Maryland: University Press of
America, 1990), pp. 371-401.
87
Lloyd, op. cit., note 10, p. 210.
88
De Cael 279b 9-12. See also P. Aubenque, La Prudence chez
Aristote (Paris: P.U.F., 1963 [reprint 1993]), p. 39.
89
See Poet cap. 21.
90
See E. Bustos, 'La polémica del significado literal', in
E. Bustos, J. Echeverría, E. Pérez sedeño, E. Sánchez and
M.I. Balmaseda (eds.), Actas del I Congreso de la Sociedad
de Lógica, Metodología y Filosofía de la Ciencia en España
(Madrid: U.N.E.D., 1993), pp. 160-163.
91
See F. Nietzsche, 'Ueber Wahrheit und Lüge im
aussermoralischen Sinne', in G. Colli and M. Montinari,
Sämtliche Werke. Kritische Studienausgabe (München: de
Gruyter, 1980), vol. 1, pp. 873-890.
92
See PA I 5.
25
93
And this despite the fact that we never can reach a full
certainty about that, as Peirce stated. With regard to
Peircean fallibilism see C. S. Peirce, 'The Scientific
Attitude and Fallibilims' in J. Buchler (ed.),
Philosophical Writings of Peirce (New York: Dover
Publications, 1955), pp. 42-59.; on Popperian fallibilism
see, for instance, K. Popper, op. cit. note 66, pp. 29-51.