Rudimentary organs may be compared with the letters in a word, still retained in the spelling, but become useless in the pronunciation, but which serve as a clue in seeking for its derivation.
—Charles Darwin (1859)
When I query an informant about what a word means, I frame the question as a spoken or written linguistic text and am answered with a linguistic text. We communicate meanings as words. For many philosophers, questions of meaning are primarily questions about language, but this book has generalized the concept of meaning to interpretations of all kinds. Linguistics and philosophy of language are vast territories of erudition into which an ill-informed novice should venture with trepidation. Nevertheless, I have been persuaded that I need to at least sketch how I would relate my account of meaning to language.
An important reason to read a text is to understand the author’s intentions. And an important reason why an author might compose a text is to have her intentions understood. Languages are elaborate conventions, shared by communities of speakers, used for the composition and interpretation of linguistic texts. Conventional meanings evolve because authors and readers often both benefit from mutual understanding of authorial intentions, but the difference between an author’s intentions and how the author intends her text to be interpreted means that language can be used to misinform as well as inform.
I have defined meaning as the physical output of a process of interpretation and text as an interpretation intended to be interpreted. By these definitions, two kinds of text are central to language that I will call public and private texts. Public texts are the strings of spoken or written words that are outputs of language users and that are inputs perceived and interpreted by other language users. Each language user possesses a private text used in the composition and comprehension of public texts. The private text has an intricate material form. It is a text because (1) it is a physical interpretation of the language user’s life-experience in the context of innate a priori knowledge, and (2) it informs the composition and comprehension of public texts. The private text is informed by the evolutionary and developmental history of the language user, especially her lived experience of public texts. Mastery of a language is the context that allows decryption of texts written or spoken in words of that language. A child learns a language in much the same way as British naval intelligence (Room 40) broke German code 7500: by observation of many exemplars together with inspired guesses, tested for intelligible meaning in encounters with multiple texts, all in the context of innate and learned knowledge of how humans think. This is the hermeneutic circle.
My public texts are roughly synonymous with Saussure’s (1916) parole and my private texts with his langue. Saussure emphasized the communal nature of langue but I emphasize the personal nature of private texts. Private texts are unique to each user, but the needs of mutual understanding result in mutual information, and convergence on shared conventions, among the private texts of a linguistic community. Private texts develop over the course of a life from the cumulative perception and interpretation of public texts in linguistic and nonlinguistic context. The forms of public texts are arbitrary conventions unique to each language, but the initial bootstrapping—which allows a private text to be informed by public texts—requires key inputs from genetic texts.
Public texts are designed to evoke rich associations in the private texts of intended auditors, just as artfully placed splodges of an impressionist painting evoke interpretive mechanisms in viewers that “fill in the details.” The richness of information resides in the auditors’ private texts, not in the public texts that function as conventional pointers to contents of private texts. This indicative function of public texts depends on communicants sharing similar-enough private texts because of common humanity, similar life experiences, and shared membership in a linguistic community (they need to speak the same language). A speaker anticipates that a listener will interpret the public text in much the same way as the speaker would interpret the public text.
Words are defined using other words. For the writing of this paragraph, I opened a Magyar dictionary at random and found the word csendülni followed by other words that I assumed to be a definition of csendülni in Magyar. The public text csendülni pointed to nothing in my private text. It evoked no associations in my mind. If you are not a Magyar speaker, there is probably little you can say about csendülni except that you believe it to be a word of Magyar on my dubious authority. Most Magyar speakers could probably use csendülni in conversation, but it would have different private associations for each speaker. My friend Apari Péter offered “voice of the bell” as his spontaneous translation of csendülni. Some Magyar speakers might have no more than a vague idea what the word “really” means and might consult a Magyar dictionary for the “correct” definition. When I consult an English dictionary, I am not infrequently surprised to find that the public definition differs markedly from the sense of my private text.
This public text is an attempt to persuade; to rearrange the associations of meaning and information in your private text; to change how you interpret and use these words not reveal what they “properly” mean. It is an attempt to explain so that you will understand my intention when you consult your private text. My intent is to construct abstract attractors in your private text that will give form to your future thought.
Think of words as instruments characterized by their use, and then think of the use of a hammer, the use of a chisel, the use of a square, the use of a glue pot, and of the glue.
—Ludwig Wittgenstein (1958)
Dilthey (1883) distinguished the Naturwissenschaften (natural sciences) as sciences of Erklären (explanation) from the Geisteswissenschaften (spiritual sciences) as sciences of Verstehen (understanding). For Ricoeur (1971), the dialectic of explanation and understanding was the hermeneutic circle (“I understand what you intended to do, if you are able to explain to me why you did such-and-such an action”). Understanding is often used with connotations of recognition of an intention. An author considers a reader to have understood his text when the reader’s interpretation is close to that intended by the author and the author believes he has been misunderstood when the reader’s interpretation differs markedly from the author’s intended interpretation. A reader understands where the author “is coming from” when the reader recognizes the author’s intention. Authors and readers often disagree about whether the text has been understood.
In all human discourse there is ineliminable freedom of interpretation. Consider translation of a Latin text into English by a human interpreter. The inputs to translation are of two kinds: text and context. The output is a text in English. Different translators produce different translations of the same text because of differences of personal context. The Latin text does not dictate the English translation. There is, in principle, an extraordinarily complex synchronic account of the molecular mechanism by which the Latin text is translated as an English text given the molecular state of the translator, but an explanation of this molecular state depends on the even more complex evolutionary, developmental, and cultural history of the translator (what one might call the diachronic sources of the translator’s personal identity or personality).
Definition is use. A word can denote subtly different things for different speakers within a linguistic community. Some speakers may have completely “wrong” definitions because they have misunderstood what they have heard. Some variants leave few descendants, perhaps they are used and corrected in the classroom, or the word itself falls into disuse because denoted things disappear from the environment. Other variants spread though the community and the “wrong” definition becomes standard usage. The survival value of a definition is determined by its role and use within the linguistic community. Some words are “living fossils” whose definitions change slowly or not at all, whereas other words possess rapidly changing definitions. The meanings of gene and gender have changed dramatically during this century, with several usages currently in competition. Some older usages may disappear and be replaced by new words as their proponents become tired of being misunderstood.
Each speaker’s association between a word and something in the world is a personal convention, but the convention must have survival value within the community of speakers if it is to persist. The attributes that favor survival of a personal convention are various, but two can be singled out. First, the convention is more likely to persist if similar conventions are possessed by other speakers and listeners. Second, a convention is more likely to persist if the word corresponds to some significant property of the world. By some such process, gestures and sounds (or their subvocalizations) came to be about things in the real world. Origin, function, causes, and consequences are inextricably entwined in recursive systems.
The definition of a word is an evolving entity with a history but without essential attributes. Within the philosophy of biology there has been a shift from defining species as classes to defining them as individuals. An organism belongs to a particular species by virtue of its ancestry rather than by the possession of defining features. From a diachronic perspective, words have the same kind of individuality. But, from a synchronic perspective, word tokens are members of classes with consensus definitions. Words are historical kinds.
Languages are technologies of the self that have evolved as elaborate systems of reciprocal altruism to allow humans to coordinate their activities to avoid dangers and exploit opportunities. But languages are also used by language users to exploit other language users. What distinguishes morally legitimate means of persuasion from morally questionable manipulation and immoral coercion? A full exploration of this question is beyond my current intent and I will limit myself to mentioning two important factors. One criterion of morally questionable means is the imposition or threat of large costs if another individual does not make a desired choice. Another criterion is deceit: hiding from the individual you are attempting to influence the reasons for your intervention.
Es ist heute unmöglich, bestimmt zu sagen, warum eigentlich gestraft wird: alle Begriffe, in denen sich ein ganzer Prozeß semiotisch zusammenfaßt, entziehen sich der Definition; definierbar ist nur Das, was keine Geschichte hat.
—Friedrich Nietzsche (1887)
Today it is impossible to say precisely why people are actually punished: all concepts in which an entire process is semiotically concentrated defy definition; only something which has no history can be defined.
—Friedrich Nietzsche (2007)
German Prozeß has a secondary meaning of criminal trial that eludes translation by the English process.
There are two general ways of persuading readers to describe and interpret the world in your terms. One is to persuade them to adopt your definitions of existing terms. Another is to persuade them to use new terms of your invention. In this supplement of the supplement, I consider successful redefinitions of gender and, in the supplement hereto, the forlorn failure of madumnal.
In the first chapter of David Copperfield, Charles Dickens (1849) wrote: “In consideration of the day and hour of my birth, it was declared by the nurse . . . first, that I was destined to be unlucky in life; and secondly, that I was privileged to see ghosts and spirits; both these gifts inevitably attaching, as they believed, to all unlucky infants of either gender, born towards the small hours on a Friday night.” Gender has long been used as a synonym for sex, although this sense was described as “now only jocular” in the first edition of the Oxford English Dictionary (1899). At that time, sex was the usual term for the distinction between males and females, but, within a century, sex was replaced by gender as the preferred term of many English speakers.
I have studied recent shifts in the use of sex and gender as a test case for thinking about memetic evolution (Haig 2004b). In brief, social psychologists and psychoanalysts introduced a distinction between “socially constructed” gender and “biologically determined” sex in the 1960s. Gender became a term of art in these fields, both as a way of marking a theoretical distinction and of signaling to informed listeners that the speaker believed the social to be more important than the biological. Animals had sex; only humans had gender. From this base, gender entered general discourse via its adoption by feminists of the 1980s to signal their belief in the predominant influence of social factors on sex differences, but a sex–gender distinction is now rarely maintained. Gender commonly now refers to all differences between males and females whether these are social or biological. Even hamsters have genders (Robins et al. 1995).
Haig (2004b) suggested two factors played a role in the reconvergence of meanings of gender and sex. First, biological and social factors often interact to determine differences between men and women, and, in such situations, there was no neutral term. Gender became the safer default in cases of overlap, thus undermining the distinction between biological sex and cultural gender. Second, many listeners heard gender being used to refer to male–female differences but were uninformed of the theoretical distinction and thus adopted gender as the fashionable general term. I now suspect another factor contributed to the rise of gender and concomitant decline of sex. As a student explained her preference for gender to me: “Gender is a category but sex is an action.” Some speakers avoid using sex because sex has copulatory connotations. Sex became a widely used euphemism for fucking only during the twentieth century. Did you wince when you read the “F-word”? I hesitated as I wrote it and still hesitate when I read it. Euphemisms are polite ways of saying things we feel uncomfortable saying. But, as a euphemism becomes widely adopted, our discomfort with the subject inevitably rubs off on the euphemism, which becomes less polite (consider the linguistic fate of “toilet,” now strongly associated with the “S-word”). Our hesitance in talking openly about “sexual activity” has tainted sex with “sexual connotations” and created a preference for a word in which these associations are less direct. Oddly, an obsolete definition of to gender is to copulate.
The public forms of sex and gender, as spoken and written texts, as vibrations in air and marks on paper, have remained unchanged through substantial changes of usage in context. Change has occurred in the private associations of the public forms. Meanings of words have histories not essential attributes. At any one time, different speakers would have given different interpretations of what the public forms meant and used the public forms in different contexts. If enough speakers use a word “wrongly” then their private definition becomes the accepted norm of their linguistic community. The heterodox becomes orthodox. Meaning is use.
Across this sequence of supplements a necessity is announced: that of an infinite chain, ineluctably multiplying the supplementary mediations that produce the sense of the very thing they defer.
(citation deferred)
An example of failed neologism is my attempt to persuade workers in the field of genomic imprinting to recognize a distinction between maternal alleles (in the mother) and madumnal alleles (from the mother in the offspring), with a parallel distinction between paternal and padumnal alleles. A mother possesses two alleles at each locus, one of which is transmitted to her offspring via an egg. The adjective “maternal” is commonly used to describe both alleles of the mother and a single allele of the offspring. These two uses of the adjective describe different situations that are subject to distinct selective forces. As an example of confusions that can arise, consider an imprinted gene that is expressed only from the allele that a daughter inherits from her mother that affects the maternal care that the daughter provides to her offspring. To which material genes would maternal gene refer in this case? Greater clarity could be achieved if one had some succinct way of distinguishing a gene inherited “from a mother” from a gene “in a mother.” I needed a linguistic difference to mark this semantic distinction.
My first solution was to use maternal for “in the mother” and maternally derived for “from the mother” (Haig and Westoby 1989), but maternal continued to be used in the scientific literature for both senses of in and from the mother. Readers considered the final three syllables of “maternally derived” to be superfluous. I next proposed a distinction between maternal (in mothers) and madernal (from mothers) (Haig 1992a). This had a certain minimalist elegance, but no one adopted madernal and the distinction between maternal and madernal was easy to miss in spoken English (there is no discernible difference in accents with soft t’s). I next proposed maternal (in a mother) and madumnal (from a mother) (Haig 1996b) the latter term modeled on autumnal. This neologism had the advantage that readers and listeners were forced to note the difference in pronunciation of maternal and madumnal, but the disadvantage that madumnal was almost universally unloved. Was this because madumnal was intrinsically cacophonous or because the unfamiliar was perceived as uncomely and might be warmly embraced with better acquaintance? If my ugly duckling were to transform into a graceful swan, then you would be seeing the world more nearly as I see it. The most important reason for the failure of my terminological innovation was that few readers saw the need for such a distinction. I am stubborn and still think the distinction is important but, for this volume, I have returned to the use of maternal and paternal because, on balance, I decided that removing the distraction provided by unfamiliar terms was worth the resulting loss of precision.
All meaning is metaphor. One thing stands in the place of another. A spoken word is transcribed as a written word that is translated as a written word in another language. The signifier is a metaphor of a signified that has been a metaphor of a signifier. The significance of texts is that texts perpetuate past choices for guidance of future choice. The interpretation of interpretations is endless. No one has the final word.