Prescription is typically contrasted with description, which observes and records how language is used in practice, and which is the basis of all linguistic research. Serious scholarly descriptive work is usually based on text or corpus analysis, or on field studies, but the term "description" includes each individual's observations of their own language usage. Unlike prescription, descriptive linguistics eschews value judgments and makes no recommendations.
Prescription and description are often seen as opposites, in the sense that one declares how language should be while the other declares how language is. But they can also be complementary, and usually exist in dynamic tension. Most commentators on language show elements of both prescription and description in their thinking, and popular debate on language issues frequently revolves around the question of how to balance these.
Standardized languages are useful for interregional communication; speakers of divergent dialects may understand a standard language used in broadcasting more readily than they would understand one another's. One can argue that such a lingua franca, if needed, will evolve by itself, but the desire to formulate and define it is very widespread in most parts of the world. Writers or communicators who wish to use words clearly, powerfully, or effectively often use prescriptive rules, believing that these may make their communications more widely understood and unambiguous. The vast popularity of books providing advice on such matters shows that prescription meets a real, or at least widely perceived need.
However, in some language communities, linguistic prescription can be regulated formally. The Académie française (French Academy) in Paris is an example of a national body whose recommendations are widely respected though not legally enforceable. In Germany and the Netherlands, recent spelling reforms were devised by teams of linguists commissioned by government and were then implemented by statute. See for example German spelling reform of 1996. The Russian language was heavily prescribed during the Soviet period, deviations from the norm being purged by the Union of Soviet Writers.
When writing is introduced into a culture, new avenues for standards are opened. Written language lacks voice tone and inflection, and other vocal features that serve to disambiguate speech, and tends to compensate for these by stricter adherence to norms. And since writers can take more time to think about their words, new avenues of standardization open up. Thus literary language, the specific register of written language, lends itself to prescription to a higher degree than spoken language.
The introduction of writing also introduces new economies into language. A body of written texts represents a sunk cost; changes in written language threaten to make the body of preserved texts obsolete, so writing creates an incentive to preserve older forms. In many places, writing was introduced by religious authorities, and serves as a vehicle for the values held to be prestigious by those authorities. Alphabets tend to follow religions; wherever western Christianity has spread, so has the Latin alphabet, while Eastern Orthodoxy is associated with the Greek or Cyrillic alphabets and Judaism with the Hebrew alphabet, and Islam and Hinduism go hand in hand with the Arabic and Devanagari scripts respectively. Similarly, the prestige of Chinese culture has preserved the usage of Chinese characters and caused their adaptation to the very different languages of Korea and Japan; the prestige of Chinese writing is such that, even when the Hangul alphabet was devised for Korean, the shapes of the letters were designed to fit the square frames of Chinese calligraphy.
Bureaucracy is another factor that encourages prescriptive tendencies in language. When government centres arise, people acquire different forms of language which they use in dealing with the government, which may be seated far from the locality of the governed. Standard writs and other legal forms create a body of precedent in language that tends to be reused over generations and centuries. In more recent times, the effects of bureaucracy have been accelerated by the popularization of travel and telecommunications; people grow accustomed to hearing speech from distant areas. Eventually, these several factors encourage standards to arise; this phenomenon has been observed since ancient Egyptian, where the spelling of the Middle Kingdom was preserved well into the Ptolemaic period in the standard usage of Egyptian hieroglyphics.
All language in developed societies therefore tends to exist on a continuum of styles. Privileged language is used in legal, ceremonial, and religious contexts, and tends to be prized over local and private speech. Written styles necessarily differ from spoken language, given the different stratagems used to communicate in writing as opposed to speech. Where the discontinuity between a high and a low style of language becomes marked, a state of diglossia arises: here, the privileged language requires special study to master, and is not instantly intelligible to the untrained. The very difficulty of the systems inspires a preservationist urge, since instruction in them represents a large effort. The writer who has mastered Chinese calligraphy or English spelling has put a great deal of time into acquiring a skill, and is likely to resist its devaluation through simplification.
However, prescription also involves conscious choices, privileging some existing forms over others. Such choices are often strategic, aimed at maximising clarity and precision in language use. Sometimes they may be based on entirely subjective judgments about what constitutes good taste. Sometimes there is a conscious decision to promote the language of one class or region within a language community, and this can become politically controversial—see below.
Sometimes prescription is motivated by an ethical position, as with the prohibition of swear words. The desire to avoid language which refers too specifically to matters of sexuality or toilet hygiene may result in a sense that the words themselves are obscene. Similar is the condemnation of expletives which offend against religion, or more recently of language which is not considered politically correct.
It is sometimes claimed that in centuries past, English prescription was based on the norms of Latin grammar, but this is doubtful. Robert Lowth is frequently cited as one who did this, but in fact he specifically condemned "forcing the English under the rules of a foreign Language". It is true that analogies with Latin were sometimes used as substantiating arguments, but only when the forms being thus defended were in any case the norm in the prestige form of English. A good example is the split infinitive: supporters of the construction frequently claim the old prohibition was based on a false analogy with Latin, but this seems to be a straw man argument; it is difficult to find a serious writer who ever argued against the split infinitive on the basis of such an analogy, and the earliest authority to advise against the construction, an anonymous American grammarian in 1834, gave a very clear statement basing his view on descriptive observation.
Foreign language teaching is necessarily prescriptive. Here the students have no prior idiom of their own in the target language and are entirely focused on the acquisition of norms laid down by others.
One of the most serious of these is that prescription has a tendency to favour the language of one particular region or social class over others, and thus militates against linguistic diversity. Frequently a standard dialect is associated with the upper class, as for example Great Britain's Received Pronunciation. RP has now lost much of its status as the Anglophone standard, being replaced by the dual standards of General American and British NRP (non-regional pronunciation). While these have a more democratic base, they are still standards which exclude large parts of the English-speaking world: speakers of Scottish English, Hiberno-English, Australian English, or AAVE may feel the standard is slanted against them. Thus prescription has clear political consequences. In the past, prescription was used consciously as a political tool; today, prescription usually attempts to avoid this pitfall, but this can be difficult to do.
A second problem with prescription is that prescriptive rules quickly become entrenched and it is difficult to change them when the language changes. Thus there is a tendency for prescription to be overly conservative. When in the early 19th century, prescriptive use advised against the split infinitive, the main reason was that this construction was not in fact a frequent feature of the varieties of English favoured by those prescribing. Today it has become common in most varieties of English, and a prohibition is no longer sensible. However, the rule endured long after the justification for it had disappeared. In this way, prescription can appear to be antithetical to natural language evolution, although this is usually not the intention of those formulating the rules.
A further problem is the difficulty of defining legitimate criteria. Although prescribing authorities almost invariably have clear ideas about why they make a particular choice, and the choices are therefore seldom entirely arbitrary, they often appear arbitrary to others who do not understand or are not in sympathy with the criteria. Judgments which seek to resolve ambiguity or increase the ability of the language to make subtle distinctions are easier to defend. Judgments based on the subjective associations of a word are more problematic.
Finally, there is the problem of inappropriate dogmatism. While competent authorities tend to make careful statements, popular pronouncements on language are apt to condemn. Thus wise prescriptive advice may identify a form as non-standard and suggest it be used with caution in some contexts; repeated in the school room this may become a ruling that the non-standard form is automatically wrong, a view which linguists reject. (Linguists may accept that a form is incorrect if it fails to communicate, but not simply because it diverges from a norm.) A classic example from 18th-century England is Robert Lowth's tentative suggestion that preposition stranding in relative clauses sounds colloquial; from this grew a grammatical dogma that a sentence should never end with a preposition.
For example, a descriptive linguist working in English would describe the word ain't in terms of usage, distribution, and history, observing both the growth in its popularity but also the resistance to it in some parts of the language community. Prescription, on the other hand, would consider whether it met criteria of rationality, historical grammatical usage, or conformity to a contemporary standard dialect. When a form does not conform—as is the case for ain't—the prescriptivist will recommend avoiding it in formal contexts. These two approaches are not incompatible, as they attempt different tasks for different purposes.
However, description and prescription can appear to be in conflict when stronger statements are made on either side. When an extreme prescriptivist wishes to condemn a very commonly used language phenomenon as solecism or barbarism or simply as vulgar, the evidence of description may testify to the acceptability of the form. This would be the case if someone wished to argue that ain't should not even be used in colloquial spoken English. Prescriptive statements will sometimes be heard which suggest that a word is inherently ugly; a descriptive approach will deny the meaningfulness of this judgment. In such instances of controversy, most linguists fall heavily on the descriptive side of the argument, accepting forms as correct or acceptable when they achieve general currency.
On the other hand, some adherents of a strongly descriptive approach may argue that prescription is always undesirable. Sometimes they see it as reactionary or stifling. A "pure descriptivist" would believe that no language form can ever be incorrect and that advice on language usage is always misplaced. However, this is a very rare position. Most of those who claim to oppose prescription per se are in fact only inimical to those forms of prescription not supported by current descriptive analysis.
Some prescriptive institutions adjust their decrees according to the native speakers. In an article entitled "Realistic Prescriptivism" (2008), descriptive linguist Ghil'ad Zuckermann "provides a critical analysis of the Academy of the Hebrew Language's mission, as intriguingly defined in its constitution: 'to direct the development of Hebrew in light of its nature'". He describes various U-turn decisions made by the Academy in recent years, and argues that the Academy of the Hebrew Language "has begun submitting to the 'real world', accommodating its decrees to the parole of native Israeli speakers, long regarded as 'reckless' and 'lazy'.