A prescriptivism with moral and political ends: The linguistic shalts and shalt nots I can get behind

In the field of linguistics, we tend to make a distinction between two ways of thinking about language and grammar: prescriptivism and descriptivism.

Prescripvitism asserts that there is one ‘correct’ or ‘proper’ way of using a language. To prescriptivists, grammar is rules handed down from authorities on the language. Novelists, journalists, and English teachers are often held up as authorities on matters of grammar. Likewise, prescriptivists often hold up dictionaries as authorities on what words mean. For prescriptivists, grammar is a matter of conformity to pre-determined rules.

In contrast, descriptivism assumes that all humans (who have not been seriously isolated or otherwise severely impaired) will acquire a fully functioning, complex language system. Because language is maintained socially, that language will show impressive similarity to the language of others. As a result, descriptivists are interested in describing the regularly occurring features of people’s speech. Descriptivists view data about how people actually speak as the source of the rules on grammar. For descriptivists, journalists, novelists, and English teachers are mere language users like the rest of us, and the dictionary describes the meanings of words according to how people actually use them. Descriptive grammatical rules are simply statements about the way people have been observed to use language.

In many cases, these two approaches are diametrically opposed. Take the following line from Eminem’s song “Rock Bottom”: “Cause I ain’t never went gold off one song”.

On the one hand, a prescriptivist would say that this sentence is not grammatical. It violates the prescriptive grammatical rule against double negatives (that is it has both ain’t and never in it). A prescriptivist might even insist that Eminem’s sentence literally means that he has gone gold off of a song (which is not his intention). The prescriptivist would also assert that ain’t is not actually a word and that the proper verb form would be haven’t gone instead of ain’t went.

On the other hand, a descriptivist would say that Eminem’s lyrics are perfectly grammatical according to the rules of his variety of English. Double negatives are common features in many varieties of English. In these varieties, two negatives in a sentence do not cancel each other out (at least not in the minds of the people who use and interpret such sentences), instead they might add emphasis to the speaker’s point. The descriptivist would also insist that ain’t is in fact a word as demonstrated by the fact that its meaning is widely known and we can use it accurately in a sentence (for a similar point describing how irregardless is a word, see this post). Also, the verb form ain’t went is also a common feature of Eminem’s English variety and simply differs from the forms expected in other forms of English. The descriptivist might go one step further and point out that ain’t went actually demonstrates features of regularization or redundancy reduction which are common in language change. For example, ain’t is a more parsimonious choice than am notare not, and is not, which requires the user to systematically select one of three different forms without gaining anything in terms of meaning. Also, using went here instead of gone shows consistency with other verbs in English where the same form used in the simple past is also used in perfect aspect, for example, “I thought” and “I have thought” or “I walked” and “I have walked“. Eminem’s grammar then actually shows signs of streamlining a purely grammatical distinction (one that provides the speaker no expressive advantage) in English between verbs that have a third form (for example, gowent, and gone) and those that don’t (for example, think and thought). In the end, the descriptivist arrives at the conclusion that we can infer grammatical rules from Eminem’s sentence even if these violate the prescriptivist’s arbitrary rules.

If you know something about the field of linguistics, then you probably know that it’s not the case that there are two camps: prescriptive linguists and descriptive linguists. I don’t find it a great exaggeration to say that linguists are by definition descriptivists, at least when speaking as linguists or language professionals. I have never met a linguist who identified themselves as a prescriptivist; indeed, the word is basically an insult in the field. It carries with it connotations of people with a shallow understanding of language insisting on rules that are entirely arbitrary and basically just being bullies and elitists. Within the field of linguistics (particularly sociolinguistics), prescriptivists are generally seen as looking for a rationalization for their own attitudes toward others, which might include racist or classist attitudes.

However a number of writers have recently begun to point out that linguists are not entirely anti-prescriptivist (for example, here). There are many ways to suggest this. For example, as a linguist I’m currently writing largely according to the rules of prescriptivist grammar, so I must not therefore be a true descriptivist, right? No. I recognize that the vast majority of people using the internet buy into prescriptivism, so I cater my message accordingly (hoping not to be the victim of their attitudes). In addition, the rules of prescriptive grammar are also not coincidentally quite similar to the grammar I’ve always used, which would be less true if I had grown up speaking, for example, African American English.

Furthermore, it’s not the case that I (or most other linguists) am an anarchist when it comes to language and grammar. First, I believe it is quite possible for people to make grammar mistakes. This occurs when they say something that they do not intend, something that violates the rules by which they normally produce language. Typos are one way this occurs. In addition, we all (linguists included) have aesthetic preferences, and we make decisions about what is ‘better’ in a particular context, perhaps because it will be more stylistically pleasing to our audience or because it will be more easily understood or be less ambiguous. However, in general, linguists acknowledge that these decisions are matters of personal preference or convention. They’re hardly the stuff of iron-clad rules. One choice is not inherently illogical or wrong. Thus, there is no reason to launch some type of campaign against the opposite choice or any alternative usage.

Hence, most linguists find the intensity with which prescriptivists defend their preferences misguided at best. I and many other linguists also view ardent prescriptivism as motivated by oppressive ideologies such as classism or racism.

This leads to an interesting question, however. Does prescriptivism need to be inherently discriminatory? Might we insist that particular language forms are not grammatically wrong but are instead morally or politically wrong? I, of course, think the answer is yes. In my view, linguists and others should be concerned about the way language is used, not about whether the structure conforms to some external authority’s arbitrary rules but rather whether the discourse conforms to our political and moral ideals.

Thus, I believe that a prescriptivism motivated by moral and political concerns is worthwhile. We should not be policing others’ language for deviance from arbitrary rules. We should be policing others’ language for the way it represents the world and others in it.

So if you’re looking for a prescriptivist commandment on language, here it is:

1. Thou shalt not use language to harm.

Avoiding harm is, of course, a rather vague rule, so I’ll provide some examples of how language can harm.

First, the most obvious example is insult. While we’ve all been told not to call people names (which I suppose is a good rule and all), what I’m really thinking of here pertains more to insidious discourse designed to put entire groups down in covert ways. For example, the use of “no homo” propagates a homophobic ideology by positioning homosexuality as a socially undesirable phenomenon. Also, the use of Mock Spanish is a covert means of perpetuating racism against Latino/as. Hence, thou shalt not use “no homo” (with the exception of attempts to reclaim the phrase), and thou shalt not use Mock Spanish.

Another way in which language and discourse can harm is through systematic exclusion. As an example, a traditional prescriptive rule requires that generic third person references use he, him, or his. Take the following sentence as an example: “If an employee desires a raise, he should ask his supervisor”. Traditionally, prescriptivists have railed against the use of singular, generic they and their in this sentence (as in “If an employee desires a raise, they should ask their supervisor”). However, whether we choose to use the cumbersome he or she or s/he or simply to pluralize the sentence, the desire to be inclusive of women is what we should be concerned with. A prescriptivism concerned with political and moral issues should be outraged not by the use of an otherwise plural pronoun (they) as singular but by the implication that he (a man) is an acceptable all-purpose substitute for person at the expense of women. There is, of course, always the argument that he in this case means anyone male or female, and everyone knows that. This argument, however, fails to recognize the empirical evidence that suggests that the use of masculine pronouns as generic leads to disproportionate representation of men, when we examine people’s spontaneous interpretations. There’s plenty of reason then to say thou shalt use language inclusive of women.

One final example of how discourse can harm is through deception or obfuscation. Of course, we all acknowledge that lying is bad. Telling someone that your name is Pat when your name is actually Sam is lying, and most people would agree that this is not acceptable under most circumstances. However, what I am referring to here is something more insidious. I’m referring to discourse designed to represent the world in such a way as to render human processes unchangeable. In this analysis of the word austerity, another blogger suggests that the word intentionally obfuscates issues related to cuts to government spending. Specifically, by using the word austerity, proponents of decreased government spending are able to mask their own contribution to the hardships created by spending cut policies and present austerity more as an unchangeable state of being rather than as the effect of politicians’ reversible actions. This type of obfuscation and deception in the discourse of politicians and others in power is harmful as it makes it more difficult to recognize what their contributions to shaping realities are, thereby making it more difficult for us to challenge their authority. This rule presents a great deal more difficulty in determining how we should use language. For example, is it wrong to use the word austerity? It’s unclear, but I think we can still celebrate the intent here: Thou shalt not obfuscate and deceive.

As a linguist, I’m not opposed to the notion that we would be concerned with the quality of our language and discourse. However, I am opposed to classism, racism, and other discriminatory ideologies. Traditionally, prescriptivism has gone hand and hand with discrimination. Nonetheless, I believe there is room for a prescriptivism that more closely reflects our moral and political ideals. It is interesting to note that the types of prescriptivist language rules I suggest above are often trivialized. It is a fascinating contradiction that language prescriptions such as the use of gender-neutral language are quickly dismissed as ‘unnecessary’ or ‘overly sensitive’ often by the same people who in the next breath will not hesitate to take a hardline stance on the non-word status of irregardless, the absolute need for the Oxford comma, or the ‘proper’ number of spaces following a period. Why these things are of greater importance or more worthy of our attention than gender-neutral language or discourse that acknowledges the role of powerful authorities in creating our realities is a mystery to me.

Tagged with: , , , , , , , , , , ,
Posted in Ideology and social change, Prescriptivism and language prejudice
3 comments on “A prescriptivism with moral and political ends: The linguistic shalts and shalt nots I can get behind
  1. Giovanna says:

    Some of what you describe as “prescriptive” was called instead “proscriptive” in my Intro to Linguistics course a few decades ago. In particular, I would say that condemnation of “ain’t” is based on an overreaction to incorrect usage, and this observation is supported by “authorities,” such as the following.

    From the Online Etymology Dictionary:

    ain’t
    1706, originally a contraction of am not, and in proper use with that sense until it began to be used as a generic contraction for are not, is not, etc., in early 19c. Cockney dialect of London; popularized by representations of this in Dickens, etc., which led to the word being banished from correct English.

    http://www.etymonline.com/index.php?term=ain%27t

  2. Mark Lewis says:

    Hi Nic! Not sure if we’ve made you a regular Speech Events reader yet, but in case you missed it, I’m sure you’d be interested in our recent roundtable considering the prescriptivism of medical abbreviation guidelines. We consider a case of language use that is said to be not morally or politically wrong, but potentially deadly!

    http://speechevents.wordpress.com/2013/11/06/email-roundtable-medical-abbreviations-and-linguistic-regulation/

  3. Life Glug says:

    Great post! Said I’d reply, apologies it’s taken me so long! Working on a few of my own and it’s eating all my time away.

    There’s another way in which I find the binary distinction between pre/descriptivism misleading – the implication of objectivity. It seems to imply that PREs are imposing some sort of ideology where DEs are being absolutely removed, only ever unproblematically describing what’s happening. But linguists (and enthusiasts) are regularly stepping into debates to call out prescriptivism (see http://www.reddit.com/r/badlinguistics) – at which point they cease to be descriptivist, and you could argue are engaging in a form of applied linguistics!

    The rigid pre/de divide ignores the fact that these prescriptive attitudes, whether we’re repulsed by them or not, are a sociolinguistic reality, and therefore worthy of description themselves. There’s a whole interesting conversation on social power that people miss out on when they reduce prescriptivists to nothing more than individual idiots to be lampooned. Just as you advocate above, I think when we’re armed with a broader theoretical understanding of how prescriptive attitudes reflect and reproduce oppression, and an acknowledgment of our own political perspectives, we can much more effectively use linguistics to enact positive social change.

    Thanks again for the mention!

    Graham

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 412 other followers

Follow linguistic pulse on Twitter
%d bloggers like this: