16 December 2010
As dead as a doornail?
About the scientific usefulness of the metaphor
There are no unsuccessful metaphors, just as there are no unfunny jokes (Davidson, 1978, 31).
‘Knowing is nothing but working with one’s favorite metaphors’, Friedrich Nietzsche (1979) sated in one of his notebooks. To a certain extent Donald Davidson (1978), philosopher of language, agrees, acknowledging that language is permeated with metaphors. Examination of their linguistic foundations, however, leads him to the conclusion that ‘to make a metaphor is to murder it’ (Davidson, 34). A consequence Hans Dooremalen, philosopher of mind, rather pursues, wishing metaphors in the scientific practice to be as dead as a doornail. The question in the current paper is whether such a metaphor free science is desirable, and whether it, supposed to be possible, would be eligible regarding the development of knowledge.
In his article Davidson (1978) is on the hunt to shoot the idea of metaphors’ double meaning. Stating that ‘the metaphor belongs exclusively to the domain of use’ (Davidson, 33), he argues against the common view that a metaphor, besides its literal meaning, has a second, figurative meaning. According to Davidson the additional portion of meaning assigned to metaphors is essentially nothing more than ‘ordinary meaning’ (Davidson, 40). Dismissing that mystical extraness of the metaphor, however, does not imply necessity of its ultimate rejection. It is an apparent fact that metaphors are principally untrue, but does that mean that they do not have a true function in our scientific language? What if, driven by the fact that the metaphor is a priori untrue, we abnegated the metaphor in science, would the remaining science then continue to be in the traditional scientific realm at all?
Charles Peirce (1877), scientist and semiotic philosopher, argues that the ambiguity raised by the metaphor provokes annoying doubt. This ‘irritation of doubt’ is philosophers’ and scientists’ main motivator to discover ‘the truth’, by getting rid of untrue statements and all possible doubt. ‘The sole object of inquiry is the settlement of opinion’, Peirce states. Yet how accurate is this scientific image?
Science is aimed at gaining knowledge about the world (in which one could question the scope of ‘knowing’ the world; see appendix), by conducting experiments based on hypotheses. According to Popper (1972) this aim demands doubting scientists, who should peal of all layers of untrue statements to eventually possibly ever become near ‘the truth’. Doing so, theories might be constructed that are able to survive several crucial tests trough time (Popper, 30). According to McMullin (1984), such theories are ‘expected to predict novel phenomena, that is, phenomena that were not part of the set to be explained’ (McMullin, 30). Theories thus are developed to get an insight in yet unknown properties of the world. Just this realm of unknowingness, the ‘explanatory gap’ (Levine, 1983; in: Chalmers, 1996, 47), is the main interest in science.
According to Davidson (1978), it is the metaphor that ‘makes us attend to some likeness, often a novel or a surprising likeness, between two or more things.’ (Davidson, 33). Thus the metaphor might point to a side of the world were the good stuff is happening, the place to be for the scientist. By its obvious falseness, the metaphor might open our eyes for the possible trueness of its opposition. Thus metaphors ‘provide a kind of lens […] through which we view the relevant phenomena’ (Davidson, 45). So to answer the first question in the current paper: metaphorical homicide would yield the loss of this convenient device to detect entities and relations in the world and to generate hypotheses and theories.
This would not only hurt the process of science, it also leads to the second question if the remaining metaphor free science would be science at all. Considering that explanatory gaps, containing yet unknown entities screaming to be inquired, are the primary candy store of objects of study, it would be paradoxical to say that science starts by finding the literal meaning of these objects, which would lead to naming the entities in a purely literal doubtless way. This would say that knowing the thing is required to know the word, which would be circular: to know the word requires to know the thing, but to inquire the thing it’s required to name, define and demarcate it. Demanding scientists to exclusively use vocabulary that is explicitly literally true, would be to ask scientists to know the extension of the objects of study in advance, that is, a priori. Besides the current philosophical reservation against such an idea of knowledge, it would also imply that science is to come to an end. If we are restricted by the words we truly know to be literal to guide our scientific practice, there would be a time in science in which everything is being said with the words we have in our scope. Science then would hit on the finity of vocabulary, without being able to develop furthermore.
To conclude I would say that I do think that metaphors form a problem in science, that is, metaphors containing the risk to be taken literally. Metaphors are literally untrue out of principle, and obviously such an untrue statement should not be taken as true or, for that matter, as scientific tool for explanation. However, to state that by this feature metaphors should be filtered out of science all together, in my opinion ignores its practical usefulness. As Aristotle already said, ‘the point of a metaphor is precisely that it is not taken literally’ (Klamer & Leonard, 1994, 22). I think metaphors function as a warning, producing a linguistic contradiction thát obvious that the irritation of doubt is raised, which forms one of the most persistent motivators in science. It challenges scientists to keep getting intrigued, to keep looking around and to stay wondering about the great unknown world. By dismissing the metaphor out of science, a huge internal motor of science would be dead. So to murder the metaphor would rather make science as dead as a doornail.
Now what about “truth”. During class “truth” was defined as the correspondence between a sentence and a state of the world, the correspondence between the utterance and the facts. In science the ultimate goal is to discover the “truth”, all the efforts we make are directed at peeling off the layers of untrueness to ultimately discover “the truth”. But what about this word, what is it referring to? Stated as I just did regarding the goal of science, it is suggested that “truth” is a state of the world, a property of our surroundings that is more or less able to be discovered. However, in the definition of truth as given above, it is suggested that “truth” applies to “correspondence”, to a relation between entities in the world, that is, between states of the world and utterances. And then we have this discussion about metaphors, of which we say that they are untrue by virtue of their being a metaphor, that is, by stating a relation that is not to be found in the real world. But in this case, the trueness is applied to the sentence, to the inherent inability of the sentence to ever correspond or apply to the real world.
I am wondering though, if these application of the term “truth” refer to the same kind of “truth”. Is trueness of a sentence the same as the “truth” we more or less expect to discover in a far away future? Is this “truth” we expect to be inherent to the world, the true state of things we might someday grasp in science, the same kind of “truth” as the “truth” we find in sentences, which we might define through logic devices like truth tables? And what about the situation in which a sentence are true but the corresponding state of the world is not, like when you utter a true sentence about a lie? Where then do we apply this term “truth”, in language, world, or relations? Is “truth” equal to “reality”? And is “truth” equal to “trueness”?
Suppose we say that these applications of the word “truth” are not the same, then the discussion about the dismissal of metaphors based on their untrueness which does not truly correspond to our scientific aim to discover the “truth”, would be metaphorical itself (that is, it would be to equate entities that are not equal).
But suppose we say that these “truths” are the same, something curious happens: the metaphor being a part of the world is true (by virtue of their existence in language, by our use and by the name we gave them), therefore, the existence of metaphors is “truth”. Thus, the use of metaphors is “truth”, hence, despite the linguistic untrueness of the sentence, the dismissal of this linguistic device would be to deny a part of the “truth” we aim to discover in science.
Either way, dismissal of the metaphor out of science based on its trueness is problematic regarding “the truth” we are desperately trying to discover in science.