Metaphor: the intertwinement of thought

and language
D.F.M. Strauss
Department of Philosophy
University of the Free State
BLOEMFONTEIN
E-mail: [email protected]

Abstract
Metaphor: the intertwinement of thought and language
The analysis of this article aims at reflecting on the nature of
metaphoricity within the context of thought and language –
inspired by the contributions of Elaine Botha in this regard
commencing about three decades ago. This paved the way for
those who were working within the tradition of reformational
philosophy to take a new look at the nature of metaphor.
Since thinking and talking are concrete activities in principle
functioning in all aspects of reality, they cannot as such provide
criteria to decide on the order relationship between the logical-
analytical and lingual aspects of reality. It turns out that, without
a proper view of the differences between concept and word, an
account of the nature of metaphor remains inconsistent. Uni-
versal traits, logical objectification, and the conceptual unknow-
ability of what is individual, surfaces in the article. The founda-
tional role of spatial relationships appears to be linked to imag-
ing and imagining, informing the proposal to differentiate be-
tween modally and entitary directed knowing.
The linguistic turn in particular inspired a renewed interest in
language and the central place of metaphorical language use.
After considering the connections between analogy and meta-
phor a new approach to the distinction between modal analo-
gies and metaphors is proposed – one that is geared towards
the interconnections between the different dimensions of our
experiential world. The last part is dedicated to Lakoff and
Johnson (1999) who have developed a peculiar view of the
“embodied mind”, “conceptual metaphor”, and “cross-domain
mappings”, while the article concludes with an argument about

Koers 76(1) 2011:11-31 11

Metaphor: the intertwinement of thought and language

the limits of substitution and take into account expanded con-

ditions.
Opsomming
Metafoor: die verstrengeling van gedagtes en taal
Die ontleding van hierdie artikel is gerig op die aard van meta-
fore binne die konteks van taal en denke – geïnspireer deur die
bydraes van Elaine Botha oor hierdie problematiek wat sowat
dertig jaar gelede begin het. Dit het die weg gebaan vir diegene
wat binne die reformatories-wysgerige tradisie werk om met
nuwe oë na metafore te kyk.
Aangesien denke en taalgebruik konkrete aktiwiteite is wat in
beginsel in alle werklikheidsaspekte funksioneer, kan dit as
sodanig nie maatstawwe bied met behulp waarvan die orde-
relasie tussen die logies-analitiese en die linguale aspekte van
die werklikheid bepaal kan word nie. Dit blyk dat rekenskap van
die aard van ’n metafoor nie op ’n konsistente wyse gegee kan
word, sonder ’n verantwoorde siening van die verskille tussen ’n
begrip en ’n woord nie. Universele trekke, logiese objektivering,
begripskennis, en die begripsmatige onkenbaarheid van wat
individueel is, verskyn ook in die artikel. Die funderende rol van
ruimteverhoudings skyn verbind te wees aan beeldvorming en
verbeelding, wat die voorstel ondersteun om tussen modaal- en
entiteitsgerigte kennis te differensieer.
Dit was egter die sogenaamde linguistieke ommeswaai wat her-
nude belangstelling in taal en in die sentrale plek van meta-
foriese taalgebruik geïnspireer het. Deur te let op die verband
tussen analogie en metafoor word ’n nuwe benadering tot die
onderskeid tussen modale analogieë en metafore voorgestel –
een wat gerig is op die verbindingslyne tussen die verskillende
dimensies van die werklikheid. Die laaste gedeelte word gewy
aan Lakoff en Johnson (1999) wat ’n besondere siening van die
“embodied mind”, “conceptual metaphor”, en “cross-domain
mappings” ontwikkel het, terwyl afgesluit word met ’n argument
insake die grense van substitusie en die verdiskontering van
meer uitgebreide kondisies.

1. Introduction
Within the circles of reformational philosophy, Elaine Botha certainly
deserves credit for being the first one who thoroughly entered the
field of philosophical reflection on the nature of metaphor – culminat-

12 Koers 76(1) 2011:11-31

 D.F.M. Strauss

ing in her recent book published by Peter Lang.1 A fundamental as-

pect of this theme will be investigated in this article – metaphoricity
in the light of the interconnections between thought and language as
well as the order relation between them. In line with the theoretical
approach of Elaine Botha, this article proceeds from the biblically in-
formed ideal of achieving a non-reductionist understanding of reality,
in the context of discerning both the uniqueness of thought and lan-
guage and attempting to avoid reducing either of them to the other.
The attempt to reduce what is truly unique to something else invari-
ably leads to the deification of something or some aspect within cre-
ation, normally accompanied by imperialistic “all”-claims such as,
“everything is number”, “everything is matter”, “everything is ratio-
nal”, or “everything is interpretation”. The distortions thus created in-
evitably result in insoluble antinomies. A Christian approach to scho-
larship, directed by the central biblical motive of creation, fall and
redemption and guided by the theoretical idea that God subjected all
of creation to his law-Word, which delimits and determines the co-
hering diversity we experience within reality, in principle safeguards
those in the grip of this ultimate commitment and theoretical orien-
tation from absolutising anything within creation, although in fact we
often do not escape from misunderstanding the diversity within crea-
tion. The method applied will largely be modal analysis – accounting
for the uniqueness and mutual coherence between aspects and their
relation to entities. It will turn out that a new understanding of meta-
phoricity flows from the analogical mapping between aspects, be-
tween entities, between aspects and entities and between entities
and aspects. One may call the approach explored in this article tran-
scendental-empirical. Its aim is to observe our experience of thought
and language by asking what underlying (transcendental) conditions
make this experience possible.

Contemplating the relationship between thought and language is

perhaps as old as philosophy itself. Various positions were assumed
ever since Lao Tse said that “a thought once uttered is a lie”. Early
Greek philosophy soon explored an alternative option, paticularly in
the claim of Parmenides that “thought and being are the same”

1 This excellent work by Botha (2007) explores a theme indirectly related to our
current discussion on metaphor in the context of thought and language –
although what is later introduced in our discussion, regarding the conditioning
role of interdimensional connections for different kinds of metaphor, may prompt
Botha to consider the expansion of the scope of metaphors and the strict
distinction between modal analogies and metaphors.

Koers 76(1) 2011:11-31 13

Metaphor: the intertwinement of thought and language

(Diels & Krantz, B.3).2 In his characterisation of being the em-

ployment of spatial properties are prominent, namely continuity and
the whole-parts relation. The B Fragments 2 and 3 of Parmenides
hold that being “... was not and will never be because it is connected
in the present as an indivisible whole, unified, coherent” (Diels-
Krantz, B.8.3-6).3 The subsequent development of philosophy,
particularly during the Middle Ages, further explored this idea.
In retrospect, this turned out to be the starting point of the sub-
sequent metaphysics of being with its inherently rationalistic under-
tones – elevating human thought to be capable of encompassing
anything, even reaching to the rank of what is divine, with God
envisaged as the ipsum esse, the Highest Being. Since creatures
were supposed to participate in this highest being, an account was
needed to explain the relationship between God and creatures.
Such an account was found in the theory of the analogy of being
(analogia entis). All creatures are in God according to their highest
being. Thomas Aquinas (1964) holds that we can know God through
his creatures, because in an eminent way God bears all the per-
fections of things within Himself. Within God they are one, but we
know God by means of these perfections as they flow from Him into
(the multiplicity of) creatures (procedentibus in creaturas ab ipso –
Thomas Aquinas, 1964:1.13.3).

This legacy opened the way to an appreciation of the problem of

similarities and differences, which is related to the notion of analogy.
When Aristotle discusses metaphor, he distinguishes four cate-
gories, mentioning naming on grounds of analogy fourth:
Metaphor consists in giving the thing a name that belongs to
something else; the transference being either from genus to
species, or from species to genus, or from species to species,
or on grounds of analogy. (Aristotle, 2001:1476 – Poetics,
1457:b.8-10.)

2 Parmenides only adds language, as an utterance, after thought and being has
been identified: “Thinking and the thought that it is are the same; for you will not
find thought apart from what is, in relation to which it is uttered.” (Diels & Krantz,
B.8.34-36.) The legacy of Parmenides is mainly observed in that the meta-
physics of being is actually a space metaphysics, supplemented by the idea of
an analogy of being. See the text below.

3 In following certain insights of Anaxagoras we find that Aristotle realised that

“everything continuous is divisible into divisible parts which are infinitely
divisible” (Aristotle, 2001:317 – Physica, 231:b.15 ff.).

14 Koers 76(1) 2011:11-31

 D.F.M. Strauss

The underlying question is how such a metaphorical name-giving is

possible? It appears that apart from relations between named
things, no metaphorical use of words or language is possible. How-
ever, such an understanding of these relations is embedded in hu-
man awareness of space, because spatial relationships enable the
discerning and distinguishing of spatial figures. Just think of lines,
squares, rectangles, triangles and spheres. Although we always re-
late spatial figures to concrete configurations of things and proces-
ses in our non-scientific experience, we simultaneously have an im-
plicit knowledge of the underlying spatial conditions making possible
the concrete shapes and forms of our everyday life.
Noticing similar patterns opens up the possibility for figurative
speech or language use, which is, therefore, also dependent upon
the foundational importance of spatial relationships.4 Eibl-Eibesfeldt
(2004) speaks of the spatial intelligence of human beings which, for
him, highlights the ability to “grasp” spatial relationships in a cen-
tered way. He holds that our thinking is spatial, combined with the
ability to translate invisible relationships into conceptual represen-
tations (Eibl-Eibesfelct, 2004:747).
The reason why spatiality is important for thought is that as soon as
the nature of a concept is contemplated we think of definitions which
delimit or demarcate what is grasped from what is not grasped. In
fact Plato already claimed in his dialogue, Parmenides, that thinking
being entails that at once non-being is also thought of. This insight
actually highlights the reciprocity of identification and distinction
which constitute the two most fundamental “legs” of logical thinking.
Identifying something relates to an awareness of the distinctness of
whatever is identified and the latter implicitly refers to the discrete
meaning of number – every number is distinct from every other num-
ber. Within the domain of logical analysis, distinction represents the
ever-present counterpart of identification, because in order to identi-
fy, one has to distinguish and vice versa. This mutual relationship
crucially depends upon the nature of concepts, for the latter is fitted
within the logical subject-object relation. A concept unifies a
multiplicity of logically objectified (universal) features. The necessary
presence of universal traits that are, via logical objectification,
brought together in the unity of a concept entails that concepts are

4 While acknowledging the importance of the whole-parts structure of physical

entities for basic level categories, Lakoff and Johnson (1999:28 ff.) do not
realise that the whole-parts relation has its original modal seat within the aspect
of space (cf. Strauss, 2009:87, 181, 236, 302 ff., 353 ff.).

Koers 76(1) 2011:11-31 15

Metaphor: the intertwinement of thought and language

blind to what is individual. This provides a strong argument for distin-

guishing between conceptual knowledge and concept-transcending
knowledge. Interestingly, whereas concepts are blind towards what
is individual,5 both the senses and human language appears to
transcend this limitation. Already during the Middle Ages it was ac-
knowledged that sensory perception can “tract down” what is indivi-
dual, while language is known for its ability to “point at” in a truly
deictic sense. Does this mean that language is foundational to logi-
cal analysis, or is it rather the other way around?

2. The order relation between the logical and the lingual

Let us start by mentioning a few succinct remarks found in the
fifteenth edition of Encyclopedia Britannica (1975) on metaphor. It is,
first of all, described as a “figure of speech” based upon “an implicit
comparison of two unlike entities” (Encyclopedia Britannica, 1975:
831). The specification “unlike” precludes entities of the same type,
but does not deny that similar (kinds of) entities are still distinct.6
This article proceeds by saying that a “metaphor makes a qualitative
leap from a reasonable, perhaps prosaic, comparison to an identi-
fication or fusion of two objects”, that is, “to make one new entity
partaking of the characteristics of both”. While the words “like” or
“as” marks the explicit comparison present in a simile, the word “is”
designates the subtle identification present in a metaphor. But do we
really get a new entity?7 It is, therefore, not surprising that this article
(Encyclopedia Britannica, 1975) records that many “critics regard
the making of metaphors as a system of thought antedating or
bypassing logic”.

Of course the decisive question is: What kind of “identification” takes

place in metaphorical language use? If it is understood in a strictly
logical sense, the inevitable conclusion would be that something

5 This was already realised by Aristotle (2001:799 – Metaph. 1036a.8-9) and as

recently as 1952 De Vleeschauwer still emphasises the fact that “knowledge of
what is individual is simply impossible” – something about which philosophy,
according to him, has had clarity since its inception (De Vleeschauwer,
1952:213).

6 Suppose we consider the category of “flying entities”, then one may encounter a
metaphor in which reference is made to the “wing” of the aeroplane. It seems as
if we have a higher level similarity and a lower level difference – “flying”
encompasses both natural and artificial flying entities.

7 When a metaphor is understood as a form of predication, the problem of

“illogical” identification disappears (cf. Köller, 1975).

16 Koers 76(1) 2011:11-31

 D.F.M. Strauss

illogical is involved. Black (1979:21) discerns something similar in a

metaphorical statement:
So perhaps the ‘mystery’ is simply that, taken as literal, a meta-
phorical statement appears to be perversely asserting some-
thing to be what it is plainly known not to be.

What Black calls “taken as literal” could be rephrased by pointing out

that asserting the impossible appears to be prominent when meta-
phorical language is understood in a strictly logical sense. This sug-
gests that the logical mode of identifying and distinguishing is in-
timately connected to metaphors even if it cannot be equated with it.
How are we to disentangle the thought and lingual elements co-
conditioning metaphorical language use?

It does not help to argue that thinking and speaking function in all
aspects of reality, because this insight does not elucidate the order
relation between the logical and the lingual. However, we are getting
closer to an understanding of this problem when we consider the
fact that the logical sense of children appears to develop more ra-
pidly than their linguistic abilities and competence. Consider the
following striking example. A little girl, who first notices a pigeon and
learns its name, can abstract “concretely”, for instance when she
shortly thereafter refers to a shrike as a pigeon. The child actually
designates the concept “bird” with the name (verbal sign) “pigeon”.
This is only possible, because from the concrete sensorially per-
ceived image of a pigeon, the girl has lifted out certain bird-cha-
racteristics, e.g. a beak, wings, feathers, while simultaneously relin-
quishing the specific characteristics that distinguish a pigeon from a
shrike.

This kind of abstraction is part of our everyday life, since ordinary

people are continually classifying (identifying) all sorts of entities by
placing them within certain categories. Otherwise, how would one be
able to identify a particular horse as a horse (i.e. belonging to the
category of horses), or a particular car as a car? Without general
concepts, such as cars and horses (in which the detail of particular
cars and horses are relinquished), this would be impossible.

This example shows that within the intellectual development of

human beings, logical concept formation precedes matching lingual
abilities. Viewed from the perspective of the distinctness of, and co-

Koers 76(1) 2011:11-31 17

Metaphor: the intertwinement of thought and language

herence between modal aspects, language use is built upon the

basis of logical skills.8

All language, including metaphorical language use, is based upon

the ability of lingual identification and lingual distinction, presuppos-
ing the original logical-analytical meaning of identification and dis-
tinction. An important difference between these two aspects is ob-
served when a concept and a word are compared. Consider the
concept of a circle and of a square. Clearly, as Kant already argued,
it is illogical to confuse these two – in a logical sense there is no
“square circle”.9 Yet, in a lingual context, we may focus on merely
one element of the meaning nuance of the word “circle”, namely “an
enclosed space”. Then the word can easily be combined in the well-
known metaphor of a “boxing ring”. While the concept of a square
and the concept of a circle bring with them every element analyti-
cally implied by them, our lingual abilities, embracing all the typical
semantic features (such as synonymity, ambiguity, metaphoricity,
etc.), may isolate particular meaning-nuances as is the case in the
lingual expression “boxing ring”. If these two words brought with
them everything entailed by their corresponding concepts, it would
simply have led to contradiction. This example shows that what has
been correctly logically objectified, within the context of language-
use, thereby manifesting the foundational position of concepts, may
transcend the restrictions of logicality in metaphorical language-use.

In a different context, Dooyeweerd (1938) advances another argu-

ment for positioning the cultural-historical aspect (and the lingual
aspect) after the logical-analytical aspect. He refers to instances in
which the process of meaning disclosure manifests itself within the
cultural-historical and post-cultural-historical aspects, without affect-
ing a deepening of non-theoretical thought to the level of the sys-
tematic mastery of a given cognitive domain. Because formative

8 Within the Afrikaans language, a quite interesting example of this foundational

relationship is found. The double negation in the Afrikaans language generates
a logic peculiar to the language itself. It is found that relatively young children (3-
5 years), who display a clear sense of logical consistency and logical
soundness, answer questions phrased in terms of the double negation with
“yes”, where older children and adults, who have matured lingually to such an
extent that they are “at home” with the (apparently “illogical”) double negation of
Afrikaans, would say “no”. In Afrikaans one may ask: “Is jy nie honger nie?”
(“Aren’t you hungry?”) A young child will answer yes whereas more mature
language users would say no.

9 See his Prolegomena zu einer jeden künftigen Metaphysik die als Wissenschaft
wird auftreten können (Kant, 1969 [1783]:341; § 52b).

18 Koers 76(1) 2011:11-31

 D.F.M. Strauss

control (mastery) reveals the unclear meaning of the historical as-

pect, and since scholarly reflection requires this deepened meaning
of analysis, it must be clear that the rise of truly scientific thought is
dependent upon the disclosure of the logical-analytical mode, and
the anticipation of the meaning of the historical modality as an
aspect coming after the logical aspect in the order of cosmic time. It
is, therefore, also striking that the historicistic mode of thought
accepts science as a “cultural factor” – to the exclusion of non-scien-
tific thought (Dooyeweerd, 1938:33; 61, footnote 49 & 50).

Regarding the order of aspects the cultural-historical aspect is pre-

supposed within the structure of the sign mode.10 If this view is
accepted, an argument showing that the logical aspect precedes the
historical is sufficient to conclude that it also precedes the sign
mode, on the basis of transitivity: if B < C and if A < B, then A < C.

3. Logical thinking and imagining

The close connection between thinking, spatiality and imaging may
open up another avenue to understand the relationship between
thought and language – one that is focused on logical conceptua-
lisation and imagining.11

Mäckler (2000:30) mentions the following definition of art by Croce:

“Art is intuition, intuition is individuality and individuality does not
repeat itself.”12 Human knowing indeed appears to be co-
conditioned by the two fundamental dimensions of reality, namely
knowledge of modal aspects and knowledge of entities. The former
is known through functional relations and the latter through imaging
that takes the shape of imagining in our uniquely human acquaint-
ance with the world. These two legs of knowing – modally directed
and entitary directed – imply each other and open the way to ac-
count for our knowledge of universality and individuality. Compare
Croce’s following conceptions of these forms of knowledge:

10 Verburg (1951:31) speaks of the “formative-instrumental substrate within lan-

guage”.

11 Since all our non-theoretical concepts are actually conceptual representations, it

is clear that spatial forms are crucial in non-theoretical thinking, showing that
spatiality is a foundational to thinking and imaging, i.e. to conceptual representa-
tions.

12 “Kunst ist Intuition, Intuition ist Individualität, und Individualität wiederholt sich
nicht.”

Koers 76(1) 2011:11-31 19

Metaphor: the intertwinement of thought and language

He states that knowledge has two forms: it is either intuitive

knowledge or logical knowledge; knowledge obtained through
the imagination or knowledge obtained through the intellect;
knowledge of the individual or knowledge of the universal; of
individual things or of the relations between them: it is, in fact,
productive either of images or of concepts. (Croce, 1953:1;
italics – DFMS.)

Surely imaginativity, as the manifestation of a specific directedness

of human knowing towards the dimension of (individual) entities,
extends across this entire dimension and cannot be restricted to
aesthetic imaginativity alone – as suggested by Seerveld (1968:45;
1979:284; 1980:132; 2001:175). In addition, the flexibility of human
understanding allows for a cross-utilisation between the two dimen-
sions of human experience, since the modal aspects serve as points
of entry to an understanding of entities. The nature of the modal as-
pects, on the other hand, can only be explained with the aid of
metaphors which are the result of imaginatively relating different
kinds of entities through predication. Yet our knowledge of concrete
entities embraces both their universality (belonging to some or ano-
ther type), and their individuality. Therefore, the difference between
what is universal and individual does not coincide with the difference
between the logical and the lingual.

Although neo-Darwinists claim that animals and humans are similar,

because animals not only use tools, but make them as well,
archeologists emphasise the human formative imagination which is
capable of inventing something different from what is presented to
the senses (cf. Narr, 1976). This view is complementary to Kant,
who defines the Einbildungskraft (imagination) as the capacity to
have a representation of an object without its presence to the sen-
ses (Kant, 1969 [1787]-B:151). This enables human beings to have
a historical awareness: memory (historical past) and expectations or
planning (historical future) – while animals are said to live in the
now.
In reaction to the conceptual rationalism of the eighteenth century
and the historicism of the nineteenth century, the linguistic turn
started to explore an alternative option. Participating in this linguistic
turn, Heidegger and Gadamer realised that language itself may be
emphasised to escape from the relativism of historicism. Van Nie-
kerk (1993:39) acknowledges this step when he points out that, ac-
cording to Gadamer, the “world” should be recognised as a creation
of language. Heidegger also realised that a new universal was
needed. In Being and time, he focuses on “there-being” as a “being-

20 Koers 76(1) 2011:11-31

 D.F.M. Strauss

in-the-world”, but he still concentrates on historical being (geschicht-

liches Dasein). However, Gadamer (1991) points out that Heidegger
did not want to once more introduce something essential or divine
with his notion of Sein (Being). Much rather, his purpose was to in-
troduce something like an event that opens the space in which
hermeneutics could become (without a final foundation) a new
universal (zum neuen Universale wird). This space is the dimension
of language.13

Dooyeweerd (1938) switches from the idea of organic coherence to

that of meaning coherence, and Seerveld explores a new under-
standing of symbolical objectification in the form of ambiguity and
allusivity. The title of Croce’s 1920 work is quite significant: Aesthe-
tic as science of expression and general linguistic (cf. Croce, 1953 –
“expression” is indeed a “general linguistic” term). Also compare the
terms used by Zuidervaart (1995): the aesthetic qualifying function is
designated as “interpretable expressions” (purely semantic-herme-
neutical categories). Even in 2001, when Seerveld once more ar-
gues in favour of “allusivity”, he remarks that it “is more sound for
doing justice to the symbolic character of Western as well as non-
Western craft and art” (Seerveld, 2001:163).

Croce, in his preface to Aesthetic (Naples, December 1901), writes:

If language is the first spiritual manifestation, and if the aes-
thetic form is language itself, taken in all its true scientific
extension, it is hopeless to try to understand clearly the later
and more complicated phases of the life of the spirit, when their
first and simplest moment is ill known, mutilated and disfigured.
(Croce, 1953:xxvii.)

One should not be surprised that Rookmaaker’s first reaction to

Seerveld’s Ph.D. thesis (1958) was that in his aesthetics he argues
the aesthetical aspect away (cf. Birtwistle, 1996:342).

• The development of the meaning of the aesthetic aspect in

Seerveld’s thinking
Seerveld starts from the notion of the “coherent symbolical objecti-
fication of meaning” (Seerveld, 1968:45). He then moves via ambi-
guity and allusivity (cf. Seerveld, 1979:284 ff.) to imaginitivty. This is
a revision of his well-known definition: “Art is the symbolical objecti-
fication of certain meaning aspects of a thing [better: ‘meaning-

13 Dieser Raum ist die Dimension der Sprache. (Gadamer, 1991:172.)

Koers 76(1) 2011:11-31 21

Metaphor: the intertwinement of thought and language

realities’ – to accept a corrective comment from N. van Til], subject

to the law of allusiveness” (Seerveld, 1979:290; cf. Seerveld, 1980:
132, note 12). What Seerveld does not realise is that the term
nuancefulness indeed analogically reflects what he wants to avoid,
namely a coherence with the numerical and spatial modes in those
aspects to which he wants to restrict the meaning of beauty/
harmony, because he aims at avoiding beauty and harmony in his
designation of the core meaning of the aesthetic aspect. However,
since the terms nuancefulness and allusivity are synonymous to
many-sidedness, it does not require much reflection to realise that
the opposite of what was aimed for is achieved: the term many ori-
ginally appears in the numerical mode while the element of sided-
ness refers to spatial configurations or sides. In addition to hidden
nuances (many-sidedness), the term allusivity echoes the meaning
of lingual ambiguities, thus showing that Seerveld’s views appear to
conflate the lingual and the aesthetic aspects. This reminds one
immediately of the title of Croce’s 1920 work, Aesthetic: as science
of expression and general linguistic (cf. Croce, 1953). It is also note-
worthy that Zuidervaart (1995:54) employs purely semantic-herme-
neutical categories when advancing the idea that the aesthetic
qualifying function of art works ought to be designated as “inter-
pretable expressions”. He designates “fit” as a technical norm, ana-
logically derived from the spatial whole-parts relation.14

There are more arguments supporting the foundational position of

the logical-analytical aspect of reality in relation to the sign mode.
Language presupposes choice and the effect of this indispensible
element of choice is that it always requires interpretation. Any lingual
expression is characterised by these hallmarks. Moreover, interpre-
tation exceeds any specific language, because whenever something
is translated into another language further interpretation is needed.
However, within a language the ever-present reality of metaphors al-
so underscores the inherent feature required of interpretation. Meta-
phors entail an enriched element of suggestiveness precisely be-
cause they lack the univocality of clear-cut concepts.

Concepts, by contrast, although they can be named or designated,

are not lingual by nature. For this reason they cannot be translated.
Only the words designating a concept can be translated into a

14 Even in 2001, when Seerveld once more argues in favour of “allusivity”, he

remarks that it “is more sound for doing justice to the symbolic character of
Western as well as non-Western craft and art” (Seerveld, 2001:163).

22 Koers 76(1) 2011:11-31

 D.F.M. Strauss

different language. A concept or an argument (inference) is

“grasped” or “understood”. It depends upon immediate insight.
Whereas language is formed presupposing the immediate functional
foundation of the cultural-historical aspect, concepts are aquired on
the basis of intuitive insight one either obtains this insight or one
does not.

Furthermore, the well-known expression that we form a concept is

employed in analogy to the cultural-historical subject-object relation.
In terms of what we shall argue below, this expression actually
represents a certain kind of metaphor.

Since concepts unite, through logical objectification, a multiplicity of

universal features, they are blind to the individual. Language has
access to the designation of universal concepts, but it can also point
to what is individual, known as the deictic function of language. The
same capacity is inherent in our (human) perception. Consider the
construction of an “identity-kid” in criminal investigations. “Seeing” is
of course also an ability of animals, although they lack the possibility
of (logically) identifying and (lingually) pointing at the criminal.

4. Analogy and metaphor

We have seen that the connection between metaphor and analogy
had already been made by Aristotle. Yet a more precise account is
required to clarify what the underlying conditions in this respect are.
One way to approach this problem is to further explore the nature of
analysis. We have seen that logical analysis rests upon the nature of
identification and distinction. The latter, in turn, presuppose simila-
rities and differences. Discerning differences presupposes similari-
ties and observing similarities presupposes differences. Therefore,
two things can never be said to be absolutely different, because the
assertion of this assumed difference is based on a crucial similairity
– both are similar in “being things”.

The most striking feature of an analogy is that it “short-circuits” the

relationship between what is similar and what is different, because it
succeeds in making either of the two subservient to the other. Con-
sider a simple metaphor, such as the elbow of the finger. Bending is
typical of an arm and of a finger. Therefore, this metaphor inter-
connects what is similar and what is different in such a way that
either of the two is manifest in the other: either the difference is
shown in the similarity or the other way around. In the case of the
elbow of the finger the difference is shown in what is similar. The
bending (i.e. what is similar) is different in the case of a finger and of

Koers 76(1) 2011:11-31 23

Metaphor: the intertwinement of thought and language

an arm – which means that this difference is shown in the moment

of similarity. Alternatively, in the moment of difference (a finger and
an arm) what is similar manifests itself – for both a finger and an arm
can bend.

Initially, after Elaine Botha started to introduce her serious interest in

the theme of metaphoricity into our philosophical discussions, my
first reflections experimented with the idea that we have to diffe-
rentiate between two kinds of analogies: analogies between modal
aspects, and analogies between entities. My first suggestion at the
time (1981-1982 – including a discussion at an international con-
ference of the Association for Calvinistic Philosophy in The Nether-
lands) – was to call analogies between aspects modal analogies and
those between entities metaphors. While metaphors are replaceable
by totally different metaphors, modal analogies (retrocipations and
anticipations) can solely be “synonymised,” that is, one can only
provide synonymous terms without actually replacing what is at
stake. The term life is synonymous with vital or vivid. As soon as
these terms are replaced by terms derived from a different mode the
original meaning of a modal analogy turns into an illigitimate re-
ductionism.
Subsequent to my reading of Lakoff and Johnson (1999) on their
understanding of conceptual metaphor I realised that there are more
options to be considered – at least in terms of the dimensions of
reality. If it is meaningful to distinguish between modal aspects and
entities one should move beyond intermodal and interentity simila-
rities and differences, because two further options surface: the simi-
larities and differences between aspects and entities and between
entities and modalities.

5. Cross-domain mappings
Lakoff and Johnson (1999) started to explore a scheme analogous
to what is found in the mathematical function concept, manifested in
their idea of source domains and target domains. As their starting
point, they look at the scope and richness of our subjective mental
life, where “subjective judgements” are made about “such abstract
things as importance, similarity, difficulty, and morality” and where
we meet “subjective experiences of desire, affection, intimacy, and
achievement”.

They are of the opinion that our conceptualisation of these ex-

periences are derived from other domains of experience, which are
sensorimotor domains. They give special attention to the cognitive
24 Koers 76(1) 2011:11-31
 D.F.M. Strauss

mechanism operative in such conceptualisations, designated as

conceptual metaphor. In another work Lakoff and Núñez (2000)
define conceptual metaphor as “a cognitive mechanism for allowing
us to reason about one kind of thing as if it were another. This
means that metaphor is not simply a linguistic phenomenon, a mere
figure of speech. Rather, it is a cognitive meganism that belongs to
the realm of thought” (Lakoff & Núñez, 2000:6). Lakoff and Johnson
(1999:45) hold, therefore, that conceptual metaphor “is pervasive in
both thought and language”. They develop their views in terms of
the embodiment of human existence. However, Botha (2006:28)
correctly points out:
Lakoff and Johnson’s anchoring of meaning in the bodily
existence is a significant step away from the Cartesian and
objectivist position, but falls short because of its location of
meaning in the subjective and materialistic dimensions of
reality.

The “embodied” nature of human subjectivity should rather be un-

derstood in terms of the subject-functions every human being has
within all aspects of reality, and not merely related to the material,
organic, and cognitive dimensions. Furthermore, it is not human
subjectivity that gives rise to the various aspects of reality, because
ultimately our humanity is conditioned by these aspects which make
possible all our concepts and the rich variety of metaphors we can
imaginatively invent.

Nonetheless, we have to ask the following question: is the idea of a

conceptual metaphor sound? We have argued that a concept is a
logical-analytical configuration and therefore not lingual by nature.
Surely Lakoff and Johnson (1999) developed their view of “concep-
tual metaphor” carefully and systematically. It particularly applies,
among other features, the idea of cross-domain mappings between
source domains and target domains. These authors say that it is “a
grounded, inference-preserving, cross-domain mapping” (Lakoff &
Núñez, 2000:6).

The strong element in this approach is the recognition of the in-

numerable images generated by cross-domain mappings that are
employed by metaphors in all possible contexts of human endea-
vour. By and large, these cross-domain mappings concern simi-
larities and differences between different (conceptualised) entitary
(including event) domains, but also between modal domains and
entitary domains (or vice versa). In all these cases, metaphors are
instances of a distinct type of analogy in the sense defined by us.

Koers 76(1) 2011:11-31 25

Metaphor: the intertwinement of thought and language

Although language is thought of as being constituted by arbitrary

signs,15 the possible ranges of semantic domains are not arbitrary.
Brandt (2000:16) states: “Basic semantic domains are neither lan-
guage-dependent nor culture-dependent, but languages, cultures,
and individuals may fill them differently to some extent.” The whole
of experience may arbitrarily be divided into “comparable segments”,
it is “also possible to identify genuine parts of it that remain stable
under cultural variation” (Brandt, 2000:16). Unfortunately the do-
mains Brandt has in mind are rather entitary-like, for he identifies “a
physical domain”, “a social domain”, “a mental domain”, and “a
speech-act domain” (Brandt, 2000:17). From an intermodal per-
spective one may first of all discern functional semantic domains,
that is to say, instances where the semantic domain of a word
embraces modal terms belonging to a specific aspect (or modal
analogies of them). Thereafter one may contemplate the addition of
semantic domains encompassing certain types of entities such as
physical or social entities.

6. The limits of substitution

Since the choice of source and target domains is relatively arbitrary,
it is understandable why we had to note that metaphors could be
replaced by other ones unrelated to the initial ones for we have seen
that only in the case of purely functional, modal analogies every
attempt at such an exchange is unsuccessful. For example, different
ways of capturing the spatial analogy within the structure of the
social aspect are: social distance; social proximity (“next-to-each-
other”); social super- and sub-ordination; social position; social
wholes and parts, and so on.
All these expressions are in an important way connotatively sy-
nonymous insofar as they (analogically) reflect some or other struc-
tural feature of the spatial aspect. This ability to “synonymise” modal
analogies is absent in the case of analogies between entities (or
entities and modal properties or modal domains) as designated by
metaphors. One may replace the metaphor “the nose of the car” by
referring to the “bonnet of the car”. Whereas we do have denotative
synonymity in this case, connotative synonymity is absent.

15 De Saussure (1966:67) says that the “bond between the signifier and the
signified is arbitrary”.

26 Koers 76(1) 2011:11-31

 D.F.M. Strauss

7. Expanded conditions
If we specify that the term metaphor does not apply to modal ana-
logies, we still have to contemplate what can be included in the
realm of metaphors. In fact, we have to expand our argument that
aspectual analogies (similarities and differences between aspects of
reality) ought to be distinguished from metaphors, because there are
actually more possibilities (cf. Strauss, 2009:155 ff.).

Metaphors may explore analogies between different entities (E-E:

“the nose of the car”), between entities and functional aspects (E-A:
such as the “web of belief”) and between aspects and entities (A-E:
for example when evolutionary biology speaks of the origin of “life”
instead of the genesis of living entities). Another A-E example is
when we speak of the “social glue” of society.

The distinct cosmic dimension of time, embracing all aspects and

entities, opens even further metaphorical possibilities. Take for ex-
ample the biotic time order of birth, growth, maturation, ageing and
dying. They can be related to the different phases of the day –
namely when early morning is designated as the birth of the day. If
the direction is reversed, physical time and biotic time enable us to
refer to the evening of life. The passage of time, in turn, may
metaphorically be related to a flying object – for example when we
remark that time is flying.
In other words, the metaphorical use of words strectches over all
possible kinds of relationships between the various dimensions of
reality, such as that of time, modal aspects and entities and pro-
cesses. Yet, since a metaphor has its original seat within the sign
mode and not within the logical-analytical aspect, the idea of
conceptual metaphor is problematic – in spite of the claims of Lakoff
and Johnson (1999) to the contrary. However, appreciating meta-
phors from the perspective of the sign mode does not deny the in-
termodal coherence between the sign mode and the logical-ana-
lytical mode. In fact, metaphoricity presupposes the conceptual-logi-
cal dimension (foundational aspect), but ought to be distinguished
from it. If the sign mode and the logical-analytical mode were not
different, the (lingual) identification present in metaphors would have
been contradictory (such as in the case of the illogical concept of a
square circle).

Koers 76(1) 2011:11-31 27

Metaphor: the intertwinement of thought and language

8. Conceptual metaphor
Strictly speaking, the expression “conceptual metaphor”, although
described in an intelligible way, conflates the sign mode with its
foundational logical-analytical aspect. Concepts are not words and
for that reason they cannot be metaphors. Furthermore, in the ab-
sence of an articulated theory of modal functions, the nature of
intermodal, interaspectual or interfunctional connections are distor-
ted by the theory of conceptual metaphor. Ontic interconnections are
presented as if they were, in fact, interconnections between con-
ceptual domains. It also explains why the discussions of conceptual
metaphor avoid an analysis of the ontic meaning of an aspect.

For example, instead of discussing the indefinability of love, as core

meaning of the ethical aspect, Lakoff and Johnson (1999) embark
on the investigation of the “concept of love” by asking whether or not
it is “independent of the metaphors for love”. The modal analogies
within the structure of the ethical aspect are not metaphors and
should, therefore, not be confused with genuine metaphors for love.
When Lakoff and Johnson (1999) explain the metaphors for love,
they clearly do not realise that they mix modal analogies and meta-
phors. They say: “Love is conventionally conceptualised, for exam-
ple, in terms of a journey, physical force, illness, magic, madness,
union, closeness, nurturance, giving of oneself, complementary
parts of a single object and heat.” (Lakoff & Johnson, 1999:71.) The
bi-unity of a couple in love embodies the numerical analogy within
the modal structure of the ethical or moral aspect: it is a “loving
union”. The moral closeness or proximity of people loving each other
highlights the spatial analogy within the structure of the moral aspect
being close by or far apart are analogies of the meaning of spatial
distance within moral relationships as reflected in language by terms
such as kinship (blood relationship), marriage parters, bosom friend,
and so on.

Let us consider an example of confusing intermodal (interaspectual)

connections with conceptual domains in the sense of Lakoff and
Johnson (1999). They refer to the rising level of water being poured
into a glass which, according to them, enables the cross-domain
mapping between quantity and verticality.16

16 They (Lakoff & Johnson, 1999:47-48) write:

This correspondence between quantity and verticality arises from a
correlation in our normal everyday experiences, like pouring more
water into the glass and seeing the level go up. Early in development,
Johnson hypothesises, such correlations are ‘conflations’ in which

28 Koers 76(1) 2011:11-31

 D.F.M. Strauss

When one enters into an analysis of what is given in reality (in a pre-
conceptualised ontic sense), then one realises that spatial pheno-
mena – such as vertical lines or verticality and horisontality – are
founded in the meaning of number. Within space we can discern
dimentionality as an order of extension in one, two, or three dimen-
sions. Without the foundational quantitative meaning of one, two and
three, the entire notion of spatial dimensions collapses into nothing-
ness. Likewise, the spatial awareness of magnitude also factually
presupposes number, because when we speak of different dimen-
sions it is also possible to speak of length (one-dimensional exten-
sion), surface (two-dimensional extension), volume (three-dimen-
sional extension), and so on. Therefore, verticality is embedded in
dimensionality and the latter also collapses into nothingness when
separated from its coherence with the quantitative meaning of the
numbers employed in designating different kinds of magnitudes.

The structural meaning of the spatial aspect is intrinsically con-

nected to the (foundational) quantitative meaning of number. Con-
sequently, one should not confuse “conceptual domains” with “ontic
functions”, because the latter ultimately co-conditions both our (in-
tegral) experience and our concept formation. This confusion may
also tempt us to deny ontic interconnections between modal aspects
on the basis of the supposed disconnectedness of “conceptual
domains”. Lakoff and Núñez (2000:324) are even drawn into a
position where their emphasis on “conceptual metaphor” convinces
them that continuity and discreteness are opposites instead of
belonging to mutually cohering but distinct ontic functions of reality.

9. Concluding remark
Metaphors indeed transcend the logical-analytical mode – without
being able to cut their ties with the conceptual basis found in the
logical aspect. Without the foundational role of (analytical) concept
formation (thinking), the entire distinction between a source domain
and a target domain, as well as the distinction between literal and
metaphorical language, becomes meaningless. Precisely because a
metaphor is not a concept, it can employ words metaphorically

quantity and verticality are not seen as separate, and associations

between them are formed. After the conflation period, according to
Grady, the associations between More and Up and between Less and
Down constitute a cross-domain mapping between the sensorimotor
concept of verticality (the source domain) and the subjective judgment
of quantity. Conventional linguistic metaphors like ‘Prices fell’ are
secondary manifestations of the primary cross-domain mapping.

Koers 76(1) 2011:11-31 29

Metaphor: the intertwinement of thought and language

without violating the analytical scope of some or other concept to

which the word(s) under consideration may refer.

List of references
ARISTOTLE. 2001 [1941]. The basic works of Aristotle. Ed. by R. McKeon with
an introduction by C.D.C. Reeve. New York: The Modern Library.
BIRTWISTLE, G.M 1996. Filosofie van de kunst en de aesthetica. (In Van
Woudenberg, R., ed. Kennis en werkelijkheid. Amsterdam: Buijten &
Schipperhijn. p. 342-370.)
BLACK, M. 1979. More about metaphor. (In Ortony, A., ed. Metaphor and
thought. New York: Cambridge University Press. p. 1-24.)
BOTHA, M.E. 2006. Metaphor, embodiment and fiduciary beliefs in science. (In
Lategan, L.O.K. & Smith, J.H. Time and context relevant philosophy.
Festschrift dedicated to D.F.M. Strauss. Bloemfontein: Association for
Christian Higher Education. p. 17-36.)
BOTHA, M.E. 2007. Metaphor and its moorings, studies in the grounding of
metaphorical meaning. Bern: Peter Lang.
BRANDT, P.A. 2000. The architecture of semantic domains – a grounding
hypothesis in cognitive semiotics. Revista portuguesa der humanidades,
4(1 & 2):11-51.
CROCE, B. 1953. Aesthetic as science of expression and general linguistic. 3rd
ed. Trans. from the definitive Italian edition from 1920. New York:
Noonday.
DE SAUSSURE, F. 1966. Course in general linguistics. Ed. by C. Bally & A.
Sechehaye. London: McGraw-Hill.
DE VLEESCHAUWER, H.J. 1952. Handleiding by die studie van die logika en
die kennisleer. Pretoria: Moerau.
DIELS, H. & KRANZ, W. 1959-1960. Die Fragmente der Vorsokratiker. Vol. 1-3.
Berlin: Weidmannsche Verlagsbuchhandlung.
DOOYEWEERD, H. 1938. Recht en historie. Paper presented at the 23rd
Scientific Gathering of the Free University of Amsterdam, 13 July 1938.
EIBL-EIBESFELDT, I. 2004. Grundriß der vergleichenden Verhaltensforschung:
Ethologie. Hrsg. Vierkirchen-Pasenach: Buch Vertrieb Blank GmbH.
ENCYCLOPEDIA BRITANNICA. 1975. Vol. 15. London: Hemingway Benton.
GADAMER, H.-G. 1991. Die Hermeneutik und der Diltheyschule. Philoso-
phische Rundschau, 38(3):161-177.
KANT, I. 1969 [1783]. Prolegomena zu einer jeden künftigen Metaphysik die als
Wissenschaft wird auftreten können. Hamburg: Felix Meiner edition.
KÖLLER, W. 1975. Semiotik und Metapher. Stuttgart: Metzier.
LAKOFF, G. & JOHNSON, M. 1999. Philosophy in the flesh: the embodied mind
and its challenge to western thought. New York: Basic Books.
LAKOFF, G. & NÚÑEZ, R.E. 2000. Where mathematics comes from: how the
embodied mind brings mathematics into being. New York: Basic Books.
MÄCKLER, A., Hrsg. 2000. 1 460 Antworten auf die Frage: Was ist Kunst?
Köln: DuMont.
NARR, K.J. 1976. Cultural achievements of early man. (In Altner, G. & Hofer, H.
Die Sonderstellung des Menschen. Stuttgart: Fischer. p. 98-152.)
SEERVELD, C.G. 1968. A Christian critique of art and literature. Toronto: The
Association for Reformed Scientific Studies.

30 Koers 76(1) 2011:11-31

 D.F.M. Strauss

SEERVELD, C.G. 1979. Modal aesthetics: preliminary questions with and

opening hypothesis. (In Kraay, J. & Tol, A., eds. Hearing and doing:
philosophical essays dedicated to H.E. van Runner. Toronto: Wedge
Publishing Foundation. p. 263-294.)
SEERVELD, C.G. 1980. Rainbows for the fallen world: aesthetic life and
aesthetic task. Toronto: Tuppence.
SEERVELD, C.G. 2001. Christian aesthetic bread for the world. Philosophia
reformata, 66(2):155-177.
STRAUSS, D.F.M. 2009. Philosophy: discipline of the disciplines. Grand
Rapids: Paideia.
THOMAS AQUINAS. 1964. Summa theologica. Latin text and English
translation, introductions, notes, appendices, and glossaries. New York,
McGraw-Hill.
VAN NIEKERK, A. 1993. Rasionaliteit en relativisme: op soek na ’n
rasionaliteitsmodel vir die menswetenskappe. Red. deur J. Mouton.
Pretoria: HSRC. (Studies in research methodology.)
VERBURG, P.A. 1951. Enkele lijnen en feiten in de ontwikkeling der taaltheorie.
(In Zuidema, S.U. & Popma, K.J. Wetenschappelijke bijdragen door de
leerlingen van Dr. D.H. Th. Vollenhoven, aangeboden ter gelegenheid van
zijn 25-jarig hoogleraarschap aan de Vrije Universiteit. Franeker: Wever.
p. 13-32.)
ZUIDERVAART, L. 1995. Fantastic things: critical notes towards a social
ontology of the arts. Philosophia reformata, 60:37-54.

Key concepts:
analogies
concept
conceptual metaphor
language
metaphor
thought

Kernbegrippe:
analogieë
begrip
begripsmetafoor
denke
metafoor
taal

Koers 76(1) 2011:11-31 31

Metaphor: the intertwinement of thought and language

32 Koers 76(1) 2011:11-31