3. Theoretical framework
3.3. Metaphor and meaning
As the previous analysis indicates, an important aspect of metaphor is the discourse between the literal and the metaphorical. Moreover, it seems obvious that both of these modes of language carry meaning, but there are notable arguments against the existence of metaphorical meaning. The most prominent of these comes from Davidson (1967; 1979), whose argument against metaphorical meaning has its roots in the numerous attempts to tame natural languages through truth-conditional semantics—in the same manner that Tarski’s2 semantic theory of truth is applied to formal languages, such as programming languages and formal, i.e.
mathematical logic (Davidson 1967: 312-5). Trying to link natural and formal systems has proven to be problematic, because, in contrast to natural languages, the latter “are artefacts designed for one or another purpose and are good or bad to the extent that they serve these purposes” (Chomsky 2000: 64).
Contrary to all of the aforementioned analysis, Davidson’s (1979: 29-30) controversial thesis is that “metaphors mean what the words, in their most literal interpretation, mean, and nothing more.” First, Davidson (1979: 30) claims that metaphors cannot be paraphrased, because there is nothing to paraphrase; if adequate paraphrases do not exist and paraphrases are equivalent to what is actually said, the inability to produce a corresponding paraphrase means that there is nothing beyond the literal to say. Secondly, while the role of metaphor in literature, science, philosophy, and law is recognized, it is void of unique meaning and limited to the simplistic role of a conduit of the ordinary meaning with the function of presenting an element of novel surprise (Davidson 1979: 31, 36). In addition, he states that metaphor is a signifier of likeness
2 Alfred Tarski was a noted logician who studied the concept of truth in formalized languages.
between its components which equates it with a corresponding simile (Davidson 1979: 31, 38).
Thus, it would only point to something evident and trite which exists and is apparent without the metaphor and which would maintain only its literal meaning even after being incorporated into a metaphor. The argument also disregards the notions that, according to Richards (1936:
94), new meanings are born as a result of two distinct thoughts interacting, and that metaphors bear specific cognitive content, i.e. meaning (Davidson 1979: 44). Finally, Davidson (1979: 39-41) argues that in terms of truth, metaphors contain nothing else but patent falsehoods or absurd truths: “the sentences in which metaphors occur are true or false in a normal, literal way, for if the words in them don’t have special meaning, sentences don’t have special truth.”
The first argument concerning paraphrasing has been rendered invalid by a plethora of linguists and metaphor experts. Black (1962: 37-9, 46; 1979: 189) states that metaphor is neither a substitute for a formal comparison nor a cryptic literal substitute with a literal meaning; a paraphrase inevitably says too much, too explicitly, missing the elegant and concise, yet rhetorically rich nature of a metaphor. Trying to duplicate a metaphor through a paraphrase is like explaining a joke—the effect of the original linguistic form is clearly superior in terms of rhetorical impact. The contrast between examples (9) and (10) displays the inadequacy of clumsy attempts at literal paraphrasing.
(9) Among the pack of computer visionaries, Steve Jobs was the alpha male.
(10) Steve jobs was feared and respected as a powerful leader by other inferior computer science innovators, which is very reminiscent of the behavior and hierarchy found in a pack of wild animals, such as lions.
In addition, as stated previously, metaphor has the important task of filling lexical gaps through catachresis in cases where the literal vocabulary is insufficient (Black 1962: 32). This function is especially important in academia and science. An astute example of this is the word quark,
which first appeared in James Joyce’s (1939: 383) Finnegans Wake to describe a cry of a gull, but since then it has been institutionalized in the terminology of particle physics. Some metaphors used in this manner have transcended the literal meanings; when working on his theories of relativity, Einstein used the terms mass and force purely metaphorically (Hesse 1993: 64).
The uniqueness of a metaphor’s meaning and form is further supported by the findings related to the conceptual theories of metaphor; linguistic form is conceptualized in spatial terms, meaning that metaphors are conceived as they appear in a specific spatial structure of a sentence (Bolinger 1977; Lakoff and Johnson 1980: 127). Thus, substitutes, comparisons, literal paraphrases, and similes cannot capture the exact meaning of metaphor. Furthermore, the previously stated advancements in the neural theory of metaphor show that the argument against cognitive metaphorical content has been proven to be inherently wrong.
Because Davidson is fixated with individual words and their perceived literal meanings, he fails to recognize that the interaction aspect of the tension theory emphasizes the intercourse of thoughts which are not bound to specific words with absolute literal meanings (Richards 1936:
94). Davies (1984: 298) notes that Davidson is wrong in saying that words are the wrong currency for a metaphorical account of an experience; it is not the words themselves which are inadequate—but the literal assertions attached to them. The erroneous assumption about compulsory likeness, or similarity underlying the metaphor structure has been debunked by both the linguistic and conceptual theories, which acknowledge that distant resemblance and seemingly paradoxical dissimilarity are, in fact, qualities that belong to powerful metaphors (Richards 1936: 125-7; Cuadrado and Durán 2013: 10). This distance between the components necessitates a metaphorical meaning which transcends the simple sum of the original literal meanings.
Davidson’s final argument about the truth values attached to literal and metaphorical meanings can be addressed by taking a game-theoretical approach to metaphor semantics, but first, a closer examination and understanding of the theory of truth-conditional semantics is needed.