Pseudolinguistics

Style over substance Pseudoscience

Popular pseudosciences

Alternative medicine Creationism Racialism Pseudopsychology

Random examples

Gaia hypothesis Pseudojournal Tachyon energy Hans Bender Ufology Eugene Mallove Shroud of Turin Hyperloop Expert à louer v - t - e

We control what you think with Language

Said and done

Sustainability George Orwell Shibboleth Deceit Ich bin ein Berliner Nobel disease

Jargon, buzzwords, slogans

WOMBAT Anti-intellectualism Feminist glossary Elitism Big words IBGYBG v - t - e

Pseudolinguistics is the study of language in a way that falls short of academic rigor. It is the linguistic cousin of the pseudoscientific family. Of course, pseudolinguistics has strong ties with other fields such as pseudohistory and pseudoarcheology.

Basics of linguistics

In linguistics, the comparative method is the accepted method to show that languages are related. Related languages should descend from a definite proto-language, and the daughter languages should have words that transform according to definite sound laws from this proto-language. For example, in Uralic languages, the Proto-Uralic language split into at least two subgroups called Finnic and Ugric, where each 'p' in Finnic is replaced by 'f' in Ugric, e.g. Finnish pata corresponds to Hungarian fazék.

In contrast, pseudolinguists usually fail to use the comparative method, relying on false or folk etymologies, spurious similarities, various other kinds of pseudoscience (such as racialist arguments), or flat-out wild speculation.

Motivations of pseudolinguistics

Nationalism

Nationalists of various stripes are especially proficient abusers of linguistics. Many like to make up family relationships between languages or language families and then claim the newly discovered kin as a part of their nation. For example, a Turkish linguist might claim that the Finno-Ugric languages are related to Turkish and that therefore the Finns are actually Turks (for some reason, you'll rarely hear it the other way around, that my language is related to your language and therefore my people don't exist but are members of your people). Language isolates (i.e. languages not demonstrably related to any others, such as Basque, Ainu, or many Native American languages) seem to be particularly popular choices for such theories, as well as any ancient language of some long-gone civilization (Sumerian, Egyptian, etc.). A typical pseudolinguistic language family claim will show a list of similar words with similar meanings in both languages. As the Zompist has already eloquently torn apart such claims, there is no need for debunking them here.

One example of the 'other way around' can be found in the book Hebrew is Greek by Joseph Yahuda, who claimed that he was able to break a cypher that enabled him to prove that ancient Hebrew and ancient Greek are the same language, and that "the Hebrews were of Hellenic descent, and that the Arabs were of Hittite (Scythian) origin; that they were both intimately related to the Greeks by religion and custom…" This is despite the consensus view among linguists that Greek and Hittite are in the Indo-European language family, and that Hebrew and Arabic are in the Afroasiatic language family.

It's popular for nationalists across much of Eurasia to claim their country or region was the Proto-Indo-European homeland.

Religion

If mixed with an unhealthy dose of religious whackery, you might find pseudolinguists looking for an "Adamic language", as allegedly spoken by all of mankind before the abandonment of the Tower of Babel. In the Middle Ages, perceptive scholars identified Basque as the Adamic language, as it was one of the few known languages clearly not related to any other language. (The search for Adamic languages should not be confused with the Proto-Human language theory, which is a legitimate - though controversial - hypothesis of all languages in the world having one origin.)

Religion often provides the basis for a lot of pseudolinguistics:

• Isaac Mozeson created something he calls "Edenics", which purports to trace all languages to old Hebrew by ignoring vowels and permuting consonants.
• Some Mormon apologists claim that the Native American Uto-Aztecan languages are linked in some way to the Semitic languages, a supposed ground-breaking discovery that has attracted little attention in the wider linguistics sphere. The supposed language of the Book of Mormon (considered by non-Mormons to be a fabrication), "Reformed Egyptian" is also nonsense.
• Young-Earth creationists sometimes claim that Chinese characters prove the veracity of the Biblical account of creation.
• The ideas of channelled languages and glossolalia.

Alternatives to the comparative method

Glottochronology

“” Glottochronology must be about the only scientific technique where the accuracy of one’s results is enhanced by the removal rather than the augmentation of data!

—Douglas Adams and J.P Mallory

Glottochronology, also called lexicostatistics, is a method of estimating the date of splitting between two languages. It involves the comparison of several languages' basic vocabulary and then deriving a date of diversion. In some ways, glottochronology can be compared to genetic analyses or carbon dating. Glottochronology was invented by Morris Swadesh, a historical linguist in the 1950s.

Glottochronology rests on the following assumptions:

1. That there is a basic vocabulary that is culture-free. This is assembled into a Swadesh list, a list of 100 (originally 200) words. These words include various body parts, basic verbs, colors, taxonomic terms like "fish", "bird" or "tree", various pronoun forms, Sun, Moon, Star, and geographic terms like "mountain".
2. That this basic, culture-free vocabulary is lost at a rate of 14% every thousand years.
3. That this rate is constant across languages
4. That one can compute the time of divergence between two languages when the amount of basic vocabulary cognates between them becomes known.

Unfortunately, glottochronology's assumptions are flawed:

• A lot of items on the Swadesh list such as "person" are not actually culture-free or are easily borrowable. Additionally, languages can have multiple words that each fit the meanings on the Swadesh list. Conversely, some languages have no equivalent words on the Swadesh list (colors are especially susceptible to this). Words can also be changed for cultural reasons or plain taboo.
• The 86% retention rate is challenged by comparing English and Icelandic. Icelandic retained 97.3% of its basic vocabulary while English only retained 67.8%. The wide gap severely removes confidences in the 86% number.
• Borrowings from related languages would skew the date of splitting forwards.

Other methods

Mass comparison, like the comparative method, assembles a list of proposed cognates between a large number of languages. Unlike the comparative method, it relies only on superficial similarities between languages and ignores the possibility of coincidences or borrowings. This makes results derived from it more error-prone, and connections between languages derived from it less trustworthy. This method has been used by Joseph Greenberg, John Bengtson and Merritt Ruhlen et al. to argue for their far-reaching macrofamilies and for proto-world.

Some people make proposals for language relations based on non-linguistic evidence, like genetics. These proposals are invalid for several reasons:

• A person can learn multiple languages while only having one set of genes their entire lives.
• Individuals and communities can abandon one language and adopt another.

Macrofamilies

See also: Indo-European languages#Proposed external connections

Macrofamilies are proposals to link large numbers of language families (and sometimes isolates).

• Altaic holds that the Turkic, Mongolic, and Tungusic language families are related. Occasionally the Korean and Japanic languages are included as part of a broader "Macro-Altaic" family with the other three being "Micro-Altaic". Even more rarely Ainu is included. Many of the proposed cognates may not be cognates, but borrowings and many of its proposed characteristics are typologically common, such as vowel harmony, or can be distributed areally. Because of the way Altaicists handle evidence, like their ignorance of the sociopolitical and cultural history of languages, reliance on wordlists and dictionaries instead of looking at texts, lack of regular phonological correspondence between proposed cognates, treatment of ghost words, treatment of loanwords, stretching of the semantics of compared words, and occasional misrepresentation of data it has been called a cult by at least one former supporter.

• Ural-Altaic, which combines Altaic and Uralic, inherits most of the problems of Altaic. This idea actually predates Altaic proper.

• Nostratic has many varied forms. In the form proposed by Vladislav "Slava" Illich-Svitych, who originated the hypothesis, it included: Dravidian, Altaic, Kartvelian, Afroasiatic, Uralic, and Indo-European. A different version proposed by Allan Bomhard includes Indo-European, Afroasiatic, Uralic-Yukaghir, Elamo-Dravidian, Altaic, and possibly Sumerian. There is a wide variety of families included or excluded from Nostratic, so it may not make sense to speak of Nostratic as a single hypothesis. A similar hypothesis proposed by Joseph Greenberg is Eurasiatic. It includes Indo-European, Uralic, Yukaghir, Altaic, Ainu, Korean, Japanese, Nivkh, Chukotko-Kamchatkan, and Eskimo—Aleut. It was meant to be a competing hypothesis, but includes numerous families that have at various times been thought to be Nostratic. Allan Bomhard's version of Nostratic includes a different reconstruction of Proto-Indo-European with the Glottalic hypothesis, where traditionally reconstructed b, d, g are instead p' , t' , k' . This means his version includes different cognates than Nostracists who use the traditional reconstructions.

Proto-World

Proto-World, or Proto-Human is the hypothetical ancestor of all languages today. While the idea that all living human languages go back to a common ancestor has merit, claims to have reconstructed what it looked like do not. Reconstructions of Proto-World typically use the following methodology:

A. Ignore vowels entirely: Any vowel matches any other vowel

B. For consonants, roughly similar places of articulation suffices to match cognates [though non-initial consonants are sometimes allowed drastic differences]. Minor place changes are acceptable: velars match uvulars, palatals, etc. Other features play no role whatsoever, so that oral stops correspond to nasals, etc.

C. Any differences in place which parallel widely attested sound changes such as lenition are acceptable, so that any consonant can be reflected by [h]

D. In semantics, any narrowing or any metaphorical extension is acceptable without further justification (such as cultural or historical arguments), so that 'dog' corresponds to 'fox, lynx, deer' etc. and 'arm' to 'elbow/hand, fingernail, foot, armpit, shoulder/arm,' and so forth.

This methodology creates the potential for numerous false cognates. *kuna "woman", one of the most commonly cited examples of a supposed Proto-World word, exemplifies these problems:

• The initial k sound can be "k, k', g, x, q, x, h, w, b, ž, ʔ, č etc."
• The final n sound can be "n, r, m, ã, w?, ʔ, and ∅ etc."
• The words to support the existence of *kuna have a variety of meanings: "‘wife,’ ‘woman,’ ‘lady,’ ‘mother,’ ‘female’ (of any species), ‘spirit of dead woman,’ ‘girl,’ ‘daughter,’ ‘maiden,’ ‘daughter-in-law,’ ‘small girl,’ ‘young woman,’ ‘old woman,’ etc."

Lyle Campbell and William Poser found that numerous Spanish words with known etymologies and histories fit the patterns established by *kuna.

Reconstructions of Proto-World face one major problem: After 6000 years, two languages are expected to share only 7% of their words. After 14000 years, nearly all of a languages basic vocabulary can be expected to change. This would make it very difficult to find true cognates even in situations where it is known that two languages are related (such as Hindi and English). At the time depths Proto-World is presumed to have existed, there is more than enough time for this to have happened several times.

In nationalism

See the main article on this topic: Nationalist pseudohistory § Similar words prove we speak the master language

Many nationalists like to claim that the language they speak was Proto-World. Some of the languages that been claimed to be Proto-World include Sanskrit, Hebrew, Gaelic, Tamil, Guarani Romanian, and Dutch. The Sun Language Theory, claims that language was invented by the Turks as a way to convert ritual blathering into a means of meaningful communication. This "theory" was promoted by the Turkish government under Atatürk in the 1930s.

A particularly stupid example is Valery Chudinov. He concluded that Russians are more ancient than most civilizations by "noticing" monosyllabic Russian words enscribed on ancient artifacts, the ocean floor, the sun's surface and drying plaster.

Occasionally, a lost continent like Mu, Lemuria, or Atlantis is invoked to explain how a language family spread. Tamil has at one point been claimed to have spoken on the lost continent of Lemuria.

Themes in pseudolinguistics

Connections between languages

Some people like to make connections between languages. These attempts are protoscience at best. Some examples include:

• Celtic and Amazigh (this idea is quite popular in some circles in France).
• Basque and a bunch of other languages. This is particularly common with apparent linguistic isolates.
• Some Romanian nationalists advance the theory idea that Latin is actually a dialect of Dacian.
• Native American tribes speak or spoke Welsh, Chinese, or Hungarian. Proponents of these ideas may appeal to assertions in historical documents, avoiding the more obvious route of juxtaposing texts in the supposedly identical languages in question, even if the languages are still alive or at least have reasonably sized bodies of texts available for comparison. A milder version of this is the idea that certain Native American languages have countless words from Old Norse.
• Various theories about Pictish, now generally considered P-Celtic (or Gallo-Brittonic).

Disconnections between languages

Conversely, some people like to claim that their language isn't related to another:

• Indian nationalists claim that Indian Indo-European languages are actually unrelated to European languages and that the language family was created as an excuse for European colonialism. Their claims may stem from the fact that the man who coined the term Indo-European, one Sir William Jones, was a British colonial judge and philologist who made the observation that Sanskrit, Latin, and Greek share word roots and so paved the way for the field of comparative linguistics. The nationalists are in some ways as wrong-minded about Indian Indo-European languages being unrelated to European languages as Jones was right about a great many things.
• Modern Hebrew has been claimed to not be a Semitic language, and that it is instead a Yiddish based creole of some sort, or a constructed language.
• At least one book alleges that the Romance languages are not descended from Latin.
• Some Hungarian nationalists believe that the Hungarian language is actually related to various non-Uralic languages, such as Hunnic or Sumerian. Usually mixed with conspiracy theories (Like claiming "the Finno-Ugric theory was pushed by the Hapsburgs" or the Bolsheviks), wishful thinking, and national mysticism. This idea is usually predicated by charlatans with expensive books.

Undeciphered writing systems

Undeciphered writing systems like Linear A from Crete or Rongorongo attract people who claim to have deciphered them but invariably have not.

The Voynich Manuscript also attracts people who claim to have deciphered it, but invariably, they have not.

Languages and the mind

Weak interpretations of the Sapir-Whorf hypothesis (the concept that language has an effect on how we interpret the world) are considered to be a part of legitimate linguistics. However, many linguists now regard the original proposition of the hypothesis, and stronger interpretations (suggesting that language defines and constrains our thoughts) as pseudolinguistics. (Modern linguistics regards the thought-control through language in Nineteen Eighty-Four as impossible.) Related to this are linguistic myths such as the claim that the Inuit have hundreds of words for snow, supposedly reflecting their deep interest in the subject.

Neurolinguistic programming also incorporates some pseudo-linguistic claims.

xenoglossy, the idea that an individual can spontaneously learn a language, is nonsense.

Auxiliary languages

Some claims used by proponents of international auxiliary languages (such as Esperanto) can be classified as pseudolinguistics.

Aliens

Some people think that ancient astronauts taught us writing or that dead languages such as Egyptian are in fact descended from alien tongues. Aliens have in all likelihood never visited Earth for much the same reasons that we have not visited them, and if they did, it is highly unlikely their languages would be pronounceable by humans. Think Chewbacca.

On a related note, essentially anything specific about extraterrestrial languages, i.e. astrolinguistics, is pseudolinguistics since it's all protoscience and speculation in reality.

Then there's this guy:

Linguistic discrimination

See the main article on this topic: Linguistic discrimination

Pseudolinguistics has been used a tool to discriminate. Until a few decades ago, the French claimed the regional languages of France were dialects of French in the Vergonha in an effort to discredit them and weed them out from daily use. The French were mostly successful in that effort, as the minority languages (Occitan, Arpitan, Gallo, etc.) are almost extinct today and are kept going only by a special effort of their respective provinces. African American Vernacular English has also been discriminated against.

Prescriptivism

Real linguistics is about understanding how real people use language, and the chief reference is the language as it is really spoken. In contrast, prescriptivists want to tell people how language should be used, based not on real-life examples but on a variety of bizarre and illogical arguments, including historical arguments, comparison with other languages (usually Latin for English-language prescriptivists), and other logical fallacies. Oxford Bibliographies Online says of linguistic prescriptivism: "this ideology and its practices are now usually ascribed to nonlinguists or nonacademic linguists, whereas modern academic linguists ... restrict themselves to the study and description of the structure of language and its natural use."

One common example is the so-called ban on splitting infinitives in English, such as in the phrase "to boldly go" which is considered bad writing by some. These have existed in English at least since the 14th century, but have been criticised by pedants since the Victorian age. Some people say this is based on a comparison with Latin because there is no such thing as a split infinitive in Latin: an infinitive in Latin is a single word, while in English it is two words ("to go"), so it is impossible to split an infinitive in Latin but easy in English. On the other hand, others point out that "go", not "to go", is actually the infinitive in English, and therefore suggest that there's no logic behind the prohibition: most of the infinitive police think it's uneducated or just don't like it.

See also

• Linguistics
• Alphabet
• Occam's razor — the majority of nationalist pseudolinguistics is based around the notion that this doesn't exist

External links

• Why is linguistics such a magnet for dilettantes and crackpots?, Gaston Dorren
• Folk linguistics on Wikipedia
• Of the Antiquity of the English Tongue, a satirical piece by Jonathan Swift

References

Articles on RationalWiki related to pseudo-studies

Pseudoarchaeology - Pseudoastronomy - Pseudohistory - Pseudolaw - Pseudolinguistics - Pseudomathematics - Pseudoscience - Pseudopsychology - Pseudoscience list - Pseudoscience in advertising - Pseudoskepticism - Pseudovitamin