Girish Mahajan (Editor)

Glosa

Updated on
Edit
Like
Comment
Share on FacebookTweet on TwitterShare on LinkedInShare on Reddit
Date
  
1972–1992

ISO 639-3
  
None (mis)

Sources
  
Interglossa

Created by
  
Ronald Clark and Wendy Ashby, based on the Interglossa of Lancelot Hogben

Setting and usage
  
international auxiliary language

Purpose
  
constructed language auxiliary language Glosa

Glosa is an international auxiliary language based on a previous draft auxiliary called Interglossa. As an isolating language, there are no inflections, so that words always remain in their dictionary form, no matter what function they have in the sentence. Consequently, grammatical functions, when not clear from the context, are taken over by a small number of operator words and by the use of word order (syntax).

Contents

History

Glosa is based on the draft auxiliary language Interglossa devised by the scientist Lancelot Hogben in the empty hours of fire-watching in Aberdeen during World War II. Interglossa was published in 1943 as a draft of an auxiliary.

Ron Clark came across the handbook of Interglossa: a draft of an auxiliary about 1960. Then he met Professor Hogben with the aim of developing the language. They worked to refine it, in order to make it more easily usable in all possible forms of communication. Wendy Ashby joined the project in 1972. When Hogben died in 1975, most changes had already been discussed. Hogben and Clark had agreed that the language should have a phonetic spelling (that is: each letter representing a single sound). This principle implied that the Greek CH, TH and PH now should be spelt K, T and F.

Finally a few further changes were introduced by Ron Clark and Wendy Ashby, who then gave the language the new name Glosa (from the Greek word for tongue, language [glossa being the English transliteration]), and thus founded a new auxiliary language.

Until about 1979, Ashby and Clark tested the use of Glosa using local volunteers in the town in which they were living. During this period, the vocabulary and some details of sentence formation were developed and revised. They had moved to another town by the time they had published the first Glosa dictionary.

From 1987, the charity-status organisation GEO (Glosa Education Organisation) has promoted the teaching of Glosa as a second language in schools worldwide.

GEO’s official website was set up by Paul O. Bartlett in 1996, and it is managed at present by Marcel Springer. It provides the Glosa Internet Dictionary (Glosa Inter-reti Diktionaria), as well as an introductory course, and other resources.

Unclear history

According to History behind Glosa, after Hogben’s death “a few further and trivial changes were introduced”. But there is no precise information about them. Concerning the turn to a phonetic spelling in Glosa, it is not clear whether Hogben would have agreed that the word Glosa is “fully phonetic” while Interglossa is not:

The single “s” [in Glosa] emphasises that the language is now fully phonetic.

Glosa is usually compared to two natural languages which are analytical in different degrees, Chinese and English.

Glosa is interesting among conlangs in that it is a completely analytic language: there are no inflections for noun plurals, verb tenses, genders, and what-not. Somewhat as in English, a word may be used as more than one part of speech.

While aspects of Hogben’s Interglossa were explicitly inspired by the auxiliary Basic English, Glosa tends to work like normal English. Interglossa works with a small number of essential light verbs (up to 20), which Hogben calls “verboids” or “verbal operators”, like the 18 verb operators of Basic English. In Glosa words from this special class can be elided if the context is clear. So the question again is whether this change is the result of the earlier Clark-Hogben discussions or of the later Clark-Ashby tests.

Overview

In Glosa, words always retain their original form, regardless of their function in a sentence. Thus, the same word can function as a verb, noun, adjective or preposition. Grammatical functions are taken over by a limited number of operator words and by the word order (syntax). Subject-Verb-Object order is the standard word order, and "adjectives" usually precede "nouns", and the "verbs" follow the tense particles and the "adverbs".

Glosa is written with the Latin alphabet without special characters, there are no double vowels or consonants and pronunciation rules are simple and regular.

Most words in Glosa are taken from Latin and Greek roots.

Alphabet and phonology

1The practice of pronouncing n before a velar sound (g or k) as /ŋ/ is generally non-preferred and controversial but is used commonly in order to simplify pronunciation. 218 Steps to Fluency in Euro-Glosa notes that x may be pronounced /z/ at the start of a word but this is non-preferred. 318 Steps to Fluency in Euro-Glosa indicates q (rather than qu) for the spelling of the kw sound combination. In the vowel-IPA section above, the first pronunciation is the preferred one.

Spelling is phonetic. PH replaced by F, hard Greek CH by K, Y by I, RH by R, TH by T. There are no diphthongs in Glosa. Where two or more vowels occur together they are pronounced separately.

  • Section Note: Some foreign names may include non-Glosa letters in order to retain original spelling, observe: Spanish = Español
  • Section Note: Unlike several other auxiliary languages, Glosa uses the letters q and x. C makes the 'ch' sound in "church". Glosa lacks a character representing the phone [ʒ]. It also lacks a single letter/symbol for the 'sh' sound in "short", unlike Esperanto. Glosa represents this sound by the letter combination sc. Like Esperanto, German, Polish etc. J makes the y-sound found in "yell" or "yak" in Glosa. G and S are always "hard" (goat and, respectively, hiss/snake). In Glosa, "R" should be trilled or "tapped" (the tongue lightly taps the pallate of one's mouth), never uvularized.
  • Accent & Vocal Inflection

    The stress/accent should be placed gently on the vowel before the last consonant.

    Vocal inflection in Glosa is generally comparable to that of English- there usually is a rising inflection before a comma, semicolon, or terminal if interrogative (that is, if it is a question, the voice tends to "go upward" towards the end). A falling inflection is to occur before a full stop.

    Words

    Glosa contains two major groups of words:

    Primitives: the small number of basic function words present in most languages—these allow us to describe the relationships between the major concepts we convey. These are basically prepositions and conjunctions, such as: de [of], e [and], pre [before], supra [above], sub [under; below; lower; beneath; lesser; somewhat]. Substantives: the list of words representing the more complex things, actions and descriptions (sometimes usable for all three) present in a language, such as: via [road], kurso [run], hedo [happy], vide [see], celera [swift], tako [fast; quick; swift; brisk; hasty; prompt; hurry; nimble; rapid; rapidity; rate; speed; haste; sprint; quick; speedy; velocity]; oku [eye]. Please note that many of these words have multiple meanings, based on how they are used in a sentence (verb, adjective, etc.), exempli gratia: "oku" can mean "eye", "optical", "to notice with the eyes", "see (look)", "perceive (with the eyes)", or "to peep".

    Parts of Speech

    Glosa words can often serve as more than one part of speech. Thus part of speech is a role that the word plays in a sentence, not a tightly-bound property of a word.

    Verbs

    Most words can act as verbs, depending on their places in the sentence (usually in the medial position).

  • 1What is meant by "Prior Word" is the word used immediately prior to the verb of the sentence or clause in order to demonstrate or affect its tense. For example:
  • To show that a verb is in the past tense, add "pa" before the verb. To indicate the future tense, add "fu" before the verb. To indicate the conditional, add "sio" before the verb.

    Adjectives

    Adjectives, like the rest of the language, are not inflected. They do not change to fit the tense, number, gender, formality, or etc. of the nouns that they modify. They generally precede the word that they modify. Sometimes an adjective's place determines its meaning:

  • Mi fu lektu mo bibli = I will read one book
  • Mi fu lektu bibli mo = I will read the first book
  • To create "opposites", one just places "no-" as a prefix to the adjective. This usage is similar to that of the prefix "mal-" in Esperanto which gives the word the exact opposite meaning. So the Glosa usage below means "not beautiful". It is the equivalent of some of the uses on in- or un- in English

    kali = beautiful no-kali = ugly termo = hot, heat meso-termo = warm no-termo = cold

    Demonstratives

    u-la = that plu-la = those u-ci = this plu-ci = these

    Adverbs

    po-kron = late pre-kron = early pa-di = yesterday nu-di = today fu-di = tomorrow imedia = immediately

    Conjunctions

    akorda-co = accordingly alo = or alo...alo = either...or alora = in that case... anti-co = however e = and fini-co = finally hetero-co = otherwise jam = already kaso = case... ko-co = also klu = even... ni....ni = neither...nor pene = almost po-co = after that posi = perhaps plus-co = moreover qasi = as if... sed = but si ne... = unless vice = instead of...

    Question and Answer Words

    A word used to ask or answer a question of who, where, what, when, why, how or how much. These words form a set in a semi-systematic manner with a particle of the compound indicating abstract quantity (what person or thing, what place, what time, for what reason, in what manner, what is the amount) and the prefix/other particle indicating the specific function of the word (exactly which, all, some, negating, etc.). There are other ways to say the following correlatives, the table just shows the most basic and systematic of these:

    What is the time? = Qo horo?
  • To change a statement into an interrogative, "qe" is placed at the beginning of the sentence.
  • Numbers

    The following table uses a period (.) is used to group numbers in threes.

  • Note: Some use "centi" the older form of "hekto" for "hundred". "Centi" is now used as "hundredth" in accordance with the ISO standard usage.
  • Compound Words

    In order to form a composite word in Glosa, one just combines existing words. For example:

    pe – person who does/person (short form of persona) an – male (from andros) fe – female (from femina) do – building where (from domo meaning house) lo – location, place of (from loko)
  • Therefore, a student is stude-pe (one who studies), a male student is stude-an, a female student is stude-fe and a building where students study (school, college, etc.) is a stude-do. Likewise a hospital is pato-do (from the word pathology but meaning sickness), literally meaning a house/building for the sick.
  • tegu – cover; ceiling; (to)shutter; deck; lid (cover); eclipse; (to) shelter; casing oku-tegu – eyelid agri – field, countryside agri-lo – farm a-nu – until now

    Meals can also be formed by noun-compounding:

    evening = vespera to eat, to devour = vora dinner, supper = vespera-vora

    Phrases and Clauses

    Phrases, the basic unit of recognizable meaning in Glosa, follow a "Subject+Verb+(Object)" order and noun phrases are "Substantive Final", which means that they start with the least important word, and are followed by additional words combining progressively to extend the meaning of the substantive, which comes last.

    Punctuation

    Full-stops end sentences. They can be the normal full stop (.), the interrogative point (?), exclamation mark (!), and, theoretically, the interrobang. Semi-colon separates clauses, principal and subordinate. Colon precedes items of a catalogue. If three or more items occur in a row, they should be separated with a comma and, prior to the final item, the word "e" or "plus" (and, plus). Fe stude: biologi, kemi, e/plus Français. Comma separates items from each other.

    Sample and Useful Words

    Hello, greetings, saltutations = Saluta! Ave! Welcome = Bene-veni Please! = Place! Sorry! = Pardo! Penite! What is your name? = Tu habe qo nomina/nima? (literally: You have what name?) My name is... = Mi nomina/nima es... Where am I = Qo-lo es mi? How much? = Qanto? Do you speak Glosa = Qe tu dice Glosa? I don't understand you = Mi ne logi/kompreni tu. Thank you = Gratia You're welcome = Es nuli. (literally: It's nothing) Here's to your health = A tu eu-sani. Bless you!/Gesundheit! = (Eu-)sani (a tu)! It is a nice day = Es u bene di. I love you = Mi amo tu. Goodbye = Vale. What is that? = Qo-ra es u-la? That is...? = U-la es...? How are you? = Komo tu? Good morning! = Boni matina/mana! Good evening! = Boni po-meso-di! Boni di! (literally: Good after mid-day, Good day) Good night! = Boni noktu! Good night, sweet dreams = Boni somni! Plu boni sonia! I can't find an error = Mi ne pote detekti u defekti. Well = Bene Be well = Vale Good/well = Boni/bene/eu Well (healthy) = Sani Ki = movement, to go, to move A cat, the cat = U feli(s) Cats = Plu feli(s) Dog = Kanis Pig = Sui Bovine (cow/bull) = Bovi (fe-bovi, an-bovi) Horse = Equs Frog = Rana Bird = Avi Bee = Apis Spider = Aranea Fish = Piski A/an/the (singular) = U (before all consonants but h); un (before vowels and h) The/some (plural) = Plu

    Example Text

  • Glosa text (From: Prof. Hogben's Language Planning.[8])
  • A prima vista posi id feno u no-spe ergo de face u verba-lista; qi fu sati panto nece volu de interkomunika; sed inklude ne ma de, posi, u kilo basi verba. U nova-papira uti minimo 20,000 verba; e in English mero de mikro English - French lexiko proxi 10,000 gene lista. Pe ne nece studi id mega tem te detekti u mega mero de lista es ne-nece. U logika ge-face verba-lista sio apo multi sinonima alo proxi-sinonima, de qi Anglo-Amerika lingua es ple. Ex. little-small, big-large, begin-commence. Id ne nece tolera funktio imbrika homo band - ribbon - strip. Plus, id sio evita excesi specializa per face mo verba akti qod in Plu Palaeo Lingua gene face per tri alo ma. Exempla, u France demo nima un extra tegu de homi soma la peau, u-la de cepa la pelure; e u-la de botuli la cotte. Anti na es mei precise de France demo, na auto supra-kargo u lexiko per ko-responde seri skin - rind - jacket - peel. Kron na vide u difere inter thread - twine - cord - string - rope - tow na solo kumu nima epi nima pro qo es, a fini u metri-difere.
  • English Translation:
  • At first sight it may seem a hopeless task to construct a vocabulary that would cover all the essential words of intercommunication, yet contain not more than, say, a thousand basic words. A newspaper uses at least 20,000, and in the English section of a small English-French pocket dictionary some 10,000 are listed. It requires no lengthy scrutiny to discover that a large portion of the material is not essential. A rationally constructed word-list would discard many synonyms or near-synonyms, of which the Anglo-American language is full. For example, little - small, big - large, begin - commence. It need not tolerate functional overlap as with band - ribbon - strip. Also, it would avoid over-specialization by making one word do what in natural languages is often done by three or more. For example, the French call the outer cover of the human body la peau, that of the onion la pelure, and that of the sausage la cotte. Though less precise than the French, we ourselves overburden the dictionary with the corresponding series skin - rind - jacket - peel. When we distinguish between thread - twine - cord - string - rope - tow we are merely heaping name upon name for what is ultimately a difference in size.

    Language Sample for Comparison

    The following is the Lord's Prayer in Glosa and other languages:

    notice that in Glosa the word "sky" is derived from Greek (Ουρανός (God of the sky) > Urani (sky)) while Esperanto uses a Latin derived word (caelum-caeli)

    Word Derivation [9]

    Generally, the following derivation rules apply when creating new words for Glosa. Some basic words (often that act as specificational prefixes) are shortened (such as "an", "fe", or "pe").

    Indefinite words remain as they are (ad, de, si, kata).

  • Latin o-declination-words become the nominative plural. Therefore:
  • -us, -i ending are adapted to -i ending (rami, soni, tubi) -er, -ri become -ri (libri) -um, -i are -a ending in Glosa (exempla)
  • Words built from the perfect-tense-radix become -i (cepti, fluxi, komposi)
  • Latin -io, -ionis are not changed to the ablative-ending (-ione) but keep the nominatives -io (natio, okasio, petitio, religio, tensio).
  • The same occurs when deriving from Greek (however, Greek lacks an ablative so the dative is used instead):
  • -os, -u become -o (fobo, orto).
  • Occasionally the Greek aorist-root is taken instead of present-tense-root (gene).
  • Greek verbs become -o (1st person singular) such as: skizo.
  • Species names keep nominative (equs, ursus).
  • Any time Greek CH, Y, RH, TH and PH occur they become K, I, R, T and F respectively in Glosa.

    References

    Glosa Wikipedia