Auxlang 2012: Bravo LdP! Bravo Sambahsa!

Update: this approach has since been abandoned. Upon closer examination, using LdP vocabulary with Sambahsa grammar was found to be unsuitable since Sambahsa grammar demands an etymological orthography. Eventually I completely abandoned Indo-European grammar and settled upon using Swahili grammar with vocabulary from English, French and Swahili. The tentative name for this project became Lugha ya Afrika, which means "Language of Afrika" in Swahili.

Auxlang 2012 has been given a temporary name: Prototip. This is a simplification of Prototỳpe. I launched a website on which the progress of the language can be followed. You'll start to see some experimental grammar there over the next few weeks. The name of the site is Prototip: An Auxlang.

The reason for the simplification of the name comes from my recent experimentation with vocabulary. Seeking a way to replace some of the superb but unfamiliar vocabulary of Sambahsa with words more familiar to English speakers, or at least with modern words rather than Proto-Indo-European words, I found to my surprise that the vocabulary of Lingwa de Planeta was consistently better than anything I could come up with, despite it containing many words from unfamiliar languages such as Hindi, Russian, Chinese and Arabic. The way LdP words are written makes them reasonably easy to remember and easy to pronounce, and they look harmonious and believable. Trying to include too many English and French words, as I was doing, does not work because English and French are highly etymological in their orthographies; in many cases a better and more harmonious result comes from other languages rather than, as I was, slavishly using an English, French, or Germanic word. Sometimes the best word really is Swahili or Persian. It just works better.

In other words: if you want to use English vocabulary, speak English; if you want to use French vocabulary, speak French; if you want to use a mixture of modern Indo-European vocabulary and historic Proto-Indo-European vocabulary, speak Sambahsa; if you want to use a simplified worldlang vocabulary, speak Lingwa de Planeta. After this experiment, I can see why Sambahsa doesn't use a much greater proportion of English and French words: because doing so wouldn't work. One needs a very diverse Indo-European vocabulary to make Sambahsa work optimally.

Anyway, bravo Sambahsa!

And bravo LdP!

Once again, I just keep returning to these two languages in particular. But now, for my current needs at this particular point in time, it is mainly the grammar which keeps bringing me back to Sambahsa and mainly the vocabulary which keeps bringing me back to LdP, although overall LdP is my current favourite.

I realised today, as a result of recent experimentation, that the best way forward for further experimentation with Prototip is to use LdP vocabulary for experimentation with elements of a Sambahsa-like grammar.

There is no harm in this, since it means I get to continue learning the vocabulary I am currently learning, which is that of LdP, rather than switching to a different vocabulary, and simultaneously I get to continue learning about the grammar of Sambahsa, as I experiment with a related dialect. If the experiments all fail, I will just go back to using LdP and Sambahsa separately, having meanwhile gained a better understanding of both.

It's a win-win scenario.

Onward...

Auxlang 2012: Sambahsa orthography, English vocabulary

The beauty of experimentation is that it can tell you when you are on the wrong path. Sometimes a failed experiment is more valuable than a successful one. And as the Turkish proverb says, "No matter how far you have gone down the wrong road, turn back." Excellent advice.

My recent experiments with trying to create a vocabulary very clearly failed due to orthographical problems. Clearly it would take years for me to develop the necessary expertise to be able to design an orthography capable of reasonably faithfully importing words from English, French and German (in which Romance-language grammar will not be used for verbs).

Now, let's do a thought experiment:

(1) Imagine I already have a capable orthographical system, one which is not only capable of reasonably faithfully absorbing words from Western European languages but also from other languages such as Arabic and Persian, and imagine that orthography has already passed extensive testing;

(2) Imagine I wish to take that orthography and use it to import words from various source languages in order to create a prototype vocabulary;

(3) Imagine I then want to experiment with applying various grammatical systems to that prototype vocabulary in order to test a prototype language in different forms.

The key point here is that what I am really doing is prototyping. A prototype is something you build and then potentially might throw away; several prototypes might be built before you decide to keep one and turn it into something more permanent. Hmmm...

It is at this point in the thought experiment that my brain finally switched on. The truth is, it does not matter what source language I import words from into the prototype language, providing the orthography can handle words from the chosen source language and that the words are reasonably compatible with the grammatical systems I might try. I can make a small vocabulary from a single source language, that would be fine for building and testing a prototype auxlang. If it works, I can throw that vocabulary away and make a new vocabulary later using the same orthography.

So, what I need for prototyping is a suitable source language for vocabulary which (taking into account my interests) has both French and Germanic words. There is such a language. It's called English.

Ah. Penny drops. Thought experiment is now getting somewhere.

The necessary orthography already exists: Sambahsa orthography. The only thing I need to do is just add a grave accent to that orthography, which shall indicate the stressed syllable in polysyllabic words. No other changes.

All right, so I can easily build a prototype vocabulary simply by respelling English words in Sambahsa orthography: this will result in some phonetic changes to the English words but will remain reasonably etymological.

There is a major ancilliary benefit: I will soon have learned how to read Sambahsa texts aloud simply by the practice I will get from transliterating English words into Sambahsa.

Now let's do another thought experiment. Imagine a fictitious scenario in which a language closely related to English in vocabulary, but with a different grammar, could have developed. Here's one scenario...

First, let's imagine as suggested a couple of years ago on an excellent blog, that a country exists and has existed for many centuries, a country in which Sambahsa originated and where it is still fluently spoken by millions today. The imaginary country is located in Asia Minor, near eastern Turkey.

Second, let's imagine that there is a friendly professor of Sambahsa who lives there and who has a keen interest in auxlangs. He decides that he would like to enlighten English speakers concerning the benefits of Sambahsa grammar, by creating a language called Sambahsa-Anglodialect (as opposed to Sambahsa itself, whose full name is in fact Sambahsa-Mundialect). The Anglodialect uses grammar very similar to the Mundialect but its vocabulary is chiefly taken from English words transliterated into Sambahsa orthography. Thus, Anglodialect forms a kind of a bridge or gateway language, introducing students to Sambahsa grammar without any vocabulary difficulty, because once they have learned the orthography they will easily recognise nearly all words.

So the professor transliterates a few thousand English words into Sambahsa orthography, publishes a grammar for the new language, and voilà! Now all English-speaking students have to do is learn the orthography and they can quickly understand nearly all words. They thus quickly learn Sambahsa grammar. Later, they learn Sambahsa-Mundialect with ease.

So that is my idea. To help myself learn Sambahsa, I am going to play the role of the imaginary professor and transliterate 1000 English words into Sambahsa orthography (with the sole addition of the grave accent to indicate the stressed syllable). Then I will play around with those words to do two things: (1) properly learn Sambahsa grammar; (2) experimentally see if a useful prototype Auxlang 2012 can be thus built. That's win-win.

I was wondering (hint, hint) if some kindly scholar (hint, hint) could perhaps give me a flying start by offering to transliterate the following 100 English words into Sambahsa orthography, using the grave accent to show stress (using whatever stress pattern is necessary for the Sambahsa orthography, not necessarily English stress). (The French words are only shown to eliminate ambiguity, clarifying the sense of the English word.) Thus, I will see how to write English phonemes in Sambahsa orthography (approximately, of course, since not all English phonemes exist in Sambahsa orthography).

If not, no worries, it will just take me a bit longer.

Update: Indeed some kind scholarly passer-by did assist me with filling in the first 100 words, see below. Many thanks to Dr Olivier Simon! This is a very good start, although mixed results. More discussion later...

ENGLISH	ANGLODIALECT	MUNDIALECT	FRENCH
abdomen	abdòmen	abdòmen	abdomen m.
belly	bèlly	gvènter, gwiwòt	ventre m.
above, over	abòve	ùper	au-dessus de
absent	absènt	absènt	absent
accept	accèpte	accèpte	accepter
accident	accìdent		accident m.
accompany	accompàny		accompagner
accomplish	accòmplisch		accomplir
accuse	accùse		accuser
actor, actrice	actòr, actrìce		acteur, actrice
add	add	addeih	ajouter
address	addrèsse		adresse f.
administrative	administratìve		administratif
adventure	adventùre	aventure	aventure
adventurer	adventùrer		aventurier, aventurière
adventurous	adventuròus	aventureùs	aventureux, aventureuse
advice, counsel	[ concìl ]	concìl	conseil m.
after	àfter		après
afternoon	afternòun		après-midi
age	àge		âge
agricultural	agriculturàl		agricole
agriculture	agricultùre		agriculture f.
air	air		air m.
air conditioning	air conditiòning		climatisation, Air conditionné
airplane	airplàne		avion m.
airport	airpòrt		aéroport
almonds	almònd		amandes
alphabet	alphàbet		alfabet m.
always	olwàiyse		toujours
ambulance	ambulànce		ambulance f.
and	aind		et
and	aind		et
ankle	àinkel		cheville f.
answer n., response	respònse		réponse f.
ant	àint		fourmi
apartment	apàrtment		appartement m.
apple	àpel	àpel	pomme f.
approximately, circa	cìrca		environ
apricot	apricòt		abricot m.
April	aprìle		avril
argument, dispute	argùment, dispùte		dispute f.
arm	arm		Bras m.
around	aràwnd		autour
arrive	arrìve		arriver
art	art		art m.
ashtray	aschtràiy		cendrier m.
ask	ask		demander
August	augùst		août
aunt	ont		tante (du côté du père)
available	aveilàble		disponible
bachelor, single	bàitschler, sìnghel		célibataire
back	back		dos
backpack	backpàck		sac à dos m.
bad	bad		mauvais, vilain
balcony	balcòn		balcon m.
ball	ball		balle f.
banana	banàne		banane f.
bank	bank		banque f.
basil	bàisel		basilic m.
basis, foundation	bàse, fundatiòn		base f.
bath	bath		bain m.
bathroom	bathròum		salle de bains f.
batteries	batterìe		piles
be able to	→	ghend	pouvoir
be, to	→	ses	être
beard	berd		barbe f.
beautiful	beautéitplen		beau
beauty	beautéit		beauté f.
become	→	bih	devenir
bed	bed		lit m.
bee	→	bei	abeille f.
beef	beuf		boeuf m.
beer	bir		bière f.
before	bifòr		avant
beginning	bighìning		début m.
behind	bihàynd		derrière
believe	bilìve		croire
below, under	bilòw, éunder		sous
better than	bètter dthan		meilleur que
between	bitwìhn		entre
bicycle	bic`ycle		biciclette f., vélo m.
big	big		grand
bill, invoice	bill, invòys		facture f.
bird	beurd		oiseau m.
birthday	beurthdèy		anniversaire m.
bitter	bìtter		amer
black	black		noir
blind	blind		aveugle
blond	blond		blond
blood	blod		sang m.
blue	blou		bleu
boat	boot	boot	bateau, -x
book	buk		livre m.
border	bòrder		frontière f.
boring	bòring		ennuyeux
bottle	botèl		bouteille f.
box	box		boîte f.
brain	bren		cerveau m.
brakes	braiks		freins
bread	bred		pain m.
break	→	bregh	casser
breakfast	[ brehgfàst ]		petit déjeuner m.

Auxlang 2012: Second 50 words: Pain

Update: I've realised there is a better way and accordingly I've abandoned these attempts, as it is pretty clear that I don't know what I'm doing when it comes to designing an orthography. Therefore I've decided instead to turn to the experts for orthographical help. See my next post for details. Everything below can be ignored, since it represents a failed experiment. I've moved on.

I'm in a world of pain.

I've now imported 100 words into the Auxlang 2012.

It's not working at all.

However, on the plus side, valuable lessons are being learned.

It is clear that either:

(1) I must abandon the attempt to retain so much etymological information in the orthography, in which case many or most written words will become unrecognisable to speakers of the native source languages, many written collisions will occur between homophones, and scientific vocabulary and taxonomic terms will be difficult to properly accommodate; and/or

(2) I must be willing to accept large differences between how a word is pronounced in Auxlang 2012 and how it is pronounced in its source language, in which case spoken words will become similarly unrecognisable and furthermore people will tend to ignore the pronunciation rules, using instead their native habits and causing much auditory confusion; and/or

(3) I must adopt extensive use of diacritics to indicate differences in vowel sounds (something which, quite frankly, is probably the best option); and/or

(4) I must simplify the orthography of the whole language, perhaps in the direction of a kind of imaginary Scandinavian language, with words from all three source languages (English, French, German) being passed through a kind of imaginary filter which transforms them into simplified forms in a consistent scheme with a sort of Scandinavian or maybe Dutch flavour.

Examples:

début
anniversaire
Schönheit
brakes
to break
breakfast
become, became
blind
bicycle, biciclette
accompany
add
room
book
boat
bath
bed
bifore
below

(1) Abandon etymological orthography while retaining vaguely similar pronunciation [more extreme alternatives are shown in brackets]:

note: grave accent indicates stressed syllable (e.g. biciclètte), not any change in vowel sound; you can see that sometimes this produces ridiculous words (e.g. belò or bilò which are so confusing it is better to just exclude them from the language, in this case by using unter)

débu
aniversair          [anivesà]
shònnheit          [shònheit]
braks
tu brek
brek-fast
bekòm, bekàmm

blèind
beiseikel, bisiklètt

akòmpani
ad
rum
buk
bot
bath          [bat]
bed
bifòr          [bifò]
belò          [bilò]



(2) Mostly retain etymological orthography but accept unnatural pronunciation which does not resemble source languages:

note: these mostly look okay but some of these (e.g. break, breakfast, book, boat, below) would sound extremely bizarre, almost unrecognisable, to any native English speaker; confusion and inconsistency would result between different speakers unless they were already fluent in the source language. Similarly début and plural débuts would sound ridiculous to French ears.

début
anniversàire
shònnheit
brakes
to break
break-fast
becòme, becàme
blind
bìcycle, biciclètte
accòmpany
add
room
book
boat
bath
bed
bifòre
belòw


(3) Extensive use of diacritics [compromise in brackets]:

note: stressed syllable is indicated by a preceding apostrophe (a better scheme, using diacritics, would be devised for this); all accents indicate changes to vowel sounds; a silent letter at the end of a word is indicated by a following forward tick (´); for convenience we shall assume that as e at the end of a word, after a consonant, is always silent.

début´
anniver'saire
'Shönheit
brákes
to breák
breák-fast
be'cóme, be'cáme
blínd
'bícýcle, bici'clette
a'cómpany
àdd
room
bôk (?)
bõt (?)
bath
bèd (?)
bi'fore 


(4) An imaginary, simplified quasi-Scandinavian scheme:

débu
aniversär
schönheit
bráks
brekk
brekk-fast
bekom, bekam
bleind
bisýkel
akòmpany
add
room
bok
bot
bath
bedd
biför
belöw

There is another option: adopt Sambahsa orthography! Which perhaps is what I should have done in the first place. However I would much prefer if Sambahsa orthography marked stress. But I digress.

More to the point, I have to put some more thought into why I am designing this language anyway, which will help me to chose the best path.

For example, am I really trying to achieve a sort of English spelling reform which maintains etymological information, combined with a simplified Sambahsa-like grammar? Or am I trying to make a completely standalone language with little resemblance which just happens to draw its vocabulary from these three languages? I might even be better just to pick a Northern European language which has already absorbed words from the three source languages, and use its vocabulary instead (simplified, of course); I think what is stopping me is that those languages have too much Germanic influence, not enough French influence, and thus are further from English vocabulary. Needless to say I have great respect for Frenkisch here, although it too is more Germanic overall than what I am looking for in Auxlang 2012.

Meanwhile, the other thing I have learned is an even greater and deeper respect for the designers of Sambahsa and Lingwa de Planeta. Those designers really know what they are doing! Matters of how to deal with the verbs "to be" and "to become", as well as matters of how to handle vowel sounds for verbs present and past, and orthography in general, have been wisely considered in those languages (taking opposite paths).

How to handle the verb "to be", for example, is a massive decision; I now understand why Sambahsa went nuts here and threw a massive irregular structure at the problem. It works fantastically well. But wow, it's difficult. Anyway my point is that even if I map the English "to be" into the language (maybe using different words but schematically similar to English) this might be insufficiently precise to match the precision of Sambahsa. Meanwhile, Lingwa de Planeta works surprisingly well using context instead.

I realise what I must immediately do is take a look at a Sambahsa paragraph and try to map it into an Auxlang 2012 scheme experimentally.

I'll do that in the next post.

Meanwhile, are any of the above alternatives (1) to (4) better than the forms shown in the table below? That is the big question...

ENGLISH	AUXLANG	FRENCH	GERMAN	Plural / past
back	back	dos	Rücken	backs
backpack	back-pack	sac à dos m.	Rucksack m.	Back-packs
bad	bad	mauvais, vilain	schlecht	n/a
balcony	bàlcony	balcon m.	Balkon	bàlconys
ball	ball	balle f.	Ball m.	balls
banana	banàna	banane f.	Banane f.	banànas
bank	bank	banque f.	Bank	banks
basil	bàsil`	basilic m.	Basilikum	no plural
basis, foundation	bàsis	base f.	Grundlage f.	bàsises
bath	bath	bain m.	Bad n.	baths
bathroom	bath-room	salle de bains f.	Badezimmer	bath-rooms
batteries	bàttery	piles	Batterien	bàtterys
be able to	kann	pouvoir	können	n/a
be, to	{En-like/Sb-like scheme?}	être	sein	{En-like/Sb-like scheme?}
beard	bart	barbe f.	Bart m.	bìerds
beautiful	shonn	beau	schön	n/a
beauty	shonnheit`	beauté f.	Schönheit f.	no plural
become	becóme	devenir	werden	becáme
bed	bett	lit m.	Bett n.	betts
bee	bee	abeille f.	Biene f.	bees
beef	beef`	boeuf m.	Rindfleisch	no plural
beer	beer`	bière f.	Bier n.	no plural
before	befòre	avant	vor(her)	n/a
beginning	début´	début m.	Anfang m.	débuts
behind	behínd	derrière	Hinter	n/a
believe	believe	croire	glauben	n/a
below, under	belòw´, unter	sous	Unter	n/a
better than	besser als	meilleur que	besser als	n/a
between	betwèen	entre	zwischen	n/a
bicycle	biciclètte	biciclette f., vélo m.	Fahrrad	biciclèttes
big	bigg	grand	groß	n/a
bill, invoice	bill	facture f.	Rechnung f.	bills
bird	vògel	oiseau m.	Vogel	vògels
birthday	anniversàire	anniversaire m.	Geburtstag m.	anniversàires
bitter	bìtter	amer	bitter	n/a
black	black`	noir	schwarz	no plural
blind	blínd	aveugle	Blind	no plural
blond	blond	blond	Blond	blonds
blood	blut`	sang m.	Blut	no plural
blue	bleu	bleu	blau	no plural
boat	boot	bateau, -x	Boot n.	boots
book	book	livre m.	Buch n.	books
border	bòrder	frontière f.	Grenze f.	bòrders
boring	bòring	ennuyeux	Langweilig	n/a
bottle	bottel	bouteille f.	Flasche f.	bottels
box	boxx	boîte f.	Schachtel	boxxs
brain	brain	cerveau m.	Gehirn	brains
brakes	bráke	freins	Bremsen	brákes
bread	bredd	pain m.	Brot n.	bredds
break	break	casser	brechen	broke
breakfast	break-fast	petit déjeuner m.	Frühstück	break-fasts

Key: Red and blue indicate problem words which I think need either to be reconsidered altogether or might need respelling; red ones are considered worse than blue ones. Green (the text, not the background shading) indicates words imported from Sambahsa... mostly small grammatical words. Backtick (`) at the end of a word indicates the word has no plural form (for number; it may have a plural form for varieties). Grave accent indicates the stressed syllable (e.g. àppel). Acute accent modifies the vowel sound, as in French (e.g. café); some accute accents modify the vowel sound, to reproduce English sounds. Pronunciation is similar to but often not exactly like the source languages.


Warning: the above table will quickly become out of date as the language changes during its development. This table will not be updated to reflect future changes in the language made after this post.