Inlis: Grammatical Overview

[I don't know what I was thinking. Though this is only an overview, it's still hard to fit in enough detail to justify the title. I ought to list the personal pronouns, verbs, and such, but that would require a post on individual features. I may get to that eventually, but I need to resume the other topics I had already begun. I'll limit myself to a post or so per month on Inlis.]

There have been a lot of changes since I first unveiled Inlis, and more changes lie ahead. Unfortunately I don't have time to just sit down and elaborate the system properly, and there have been long stretches of inactivity.

It's a pity Interglossa isn't better known; I could explain Inlis in terms of Interglossa both quickly and easily. (No, Glosa isn't close enough to Interglossa for this: Inlis is closer to Interglossa than Glosa is.)

To begin with: no flexions. Once you learn a word, you're done with it. Ia ("ear") never becomes *ias; bi never becomes am, was, etc. (though these forms do exist: "arm," "wasp").

Like Interglossa, Inlis has functional classes for its words, though the divisions are different. I've combined "amplifiers" and "substantives" as "general words," which can function as modifiers (adjectives and adverbs) and noun phrase cores (nouns as such), but never as verbs. Prepositions as such are few and function only as prepositions and modifiers. Verbs are also restricted and have no other function.

Basic Structure
One of the things that bothered me about Interglossa, despite Hogben's criticism of Latino Sine Flexione's lack of proper syntactic markers, is that word functions were largely unmarked, making parsing unnecessarily difficult. So in Inlis, I've built in some markers. They constitute a few of the major divergences from Interglossa.

Noun phrases are marked in the standard language. Hogben does this too: Inlis da is almost the same as IG u(n), though it isn't always singular. Da is not a definite article! It just marks the beginning of a noun phrase. The end of a noun phrase is marked by a preposition, a verb, or some other element that just won't occur in a noun phrase, including the end of the sentence. Am I rigorous about this? No. Making the system completely self-parsing would be a pain for everyone. I want something that will seem fairly natural and simplify the parse.

Pronouns and names are not marked, so a sequence such as da mi is incomplete: it begins a possessive, such as da mi buk ("my book"). Similarly, da Frans gavan ("the French government") and da Laplas transfom ("(the) Laplace transform").

(Yes, adjectives precede nouns, and I prefer to render names phonemically. Realistically, we'll see names written out as in English with pronunciation in parentheses. That also means "Moscow (Moskau)" instead of "Moskva." Sorry, Dmitry.)

Verb phrases are marked by verbs. (Who knew?) As in IG, there aren't many real verbs, though they produce predictable (non-idiomatic) phrasal verbs. These are always verbs, too: unlike IG verboids, they are never nouns or modifiers. They do have predictable corresponding general words: bi has bien, hav has havin, and so on. Most verbs have two related meanings, one literal (triggered by a following noun phrase) and one more figurative (followed by one or more general words not preceded by da or its equivalent). Thus

Mi mek da haus I make/build a/the house.
Mi mek klin da haus I clean (make clean) a/the house.

Next we'll look at the name game in auxlanging.

Inlis: Importation and Homonyms

The main thing to remember when importing vocabulary into Inlis is that you will be importing nouns--not verbs as such nor adjectives as such. However, the forms may look like adjectives or (less often) verbs.

Verbs are imported as verbal nouns, and they generally end in -in: tinkin (thinking), wakin (working), wokin (walking), and so on. The reason for this ending is that -ing can suggest a noun or participle (and -en a participle), thus indicating the usage, and the marking as a verbal noun can avoid homonyms: rid (reed), ridin (reading); rait (right, privilege), raitin (writing). To avoid unwanted diphthongs, use -en instead of -in for roots ending in -a (rare--in fact, I can't think of an example right off) or -o: noen (knowing), goen (going). We also use -en after -i: sien (seeing, sight), not *siin. After -e we use -yen: breyen (braying), freyen (fraying), preyen (praying, prayer).

On the other hand, phrasal verbs don't use -in: bloap (not *bloenap, much less *bloapin), setap, wakaut (workout), etc. This generally coincides with English usage.

We seldom use -ion forms unless they are more common than the verb: neshan, opinyan, etc. Thus "creating, -ion" is krietin, not krieshan, which may however be used for the religious idea: Dem tok dibetin da krieshan oa da evolushan (They debate creation or evolution). Similarly, "sight, vision, seeing" is (as noted above) sien; vijan is vision in the mystical sense or that of what a visionary has.

Adjectives are imported in the shortest, simplest form unless that creates a homonym. Thus "wise" is wais, not wisdam, though the latter is more distinctive. Also hai (not *hait), waid (not wid, which means "weed"), and so on.

For nouns as such, we sometimes resort to the plural to avoid a homonym: tri "three," tris "tree"; bi "be," bis "bee."

In some cases, however, homonyms are acceptable. Thus, both "coat" and "court" become kot, but "court" will normally be lo-kot ("law-court"). But in compounds lo- will normally be dropped: kot-haus ("courthouse"). This won't keep punsters and fantasy-writers from construing it as "coat-house," nor is there any danger in such a move.

We try to avoid final consonant clusters, so the final consonant usually drops: impotan ("important"), esperantis ("Esperantist"), but esperantisam ("Esperantism").

This doesn't cover everything, but it should make current choices more intelligible and help someone who wants to import a form do so accurately.

Next time I'll sketch the grammar, which should explain some of the quirks we've already encountered.

Inlis: Phonology and Consequences

In case anyone wonders, there is an actual basis for the vowel groups: Dr. J. C. Wells's lexical sets. The purpose of the sets is to describe phonological differences between dialects. I decided to use them in an attempt to simplify the importation process, since US and British vowels differ considerably. Inlis doesn't follow either of these dialects exclusively. Rather, it's meant to be a pan-English dialect that's closer to creoles than to the rival dialects. Speakers should always sound somewhat foreign and exotic: this is one time neither Received Pronunciation nor General American is welcome.

It should be obvious that a and o are overloaded, especially a. In my dialect, the seven mappings of o are just three distinct vowels--or technically just two. Such phenomena are common.

Still, a is overloaded, and I've tried to reduce the load, so far unsuccessfully. The current mappings seem to be the best.

Probably the most controversial items will be the mappings to o of the LOT, CLOTH, PALM, and THOUGHT sets. From a (Western) US perspective, it would make more sense to map these to a, further overloading that overworked vowel. But I chose o because

1. the sounds are commonly associated with that letter both in the main English dialects and in creoles

2. it's easier to map all of these to a single vowel, and a is already overloaded, while apart from these, o would be somewhat underloaded.

This does lead to some oddities. For example, if I imported wander and wonder directly, wander would become wonda and wonder wanda.

The big weakness is that the mappings make the English tendency toward homonyms even worse, but since I'm trying to rein in synonyms anyway, I can avoid a lot of the problem.

Next time I'll explain a bit about the grammar and importation mechanism.

Inlis: Phonology and Mappings

Vowels and diphthongs

I generally use Spanish vowels as the model, but in general

a as in father

e as in bet (or as in bait without the closing y sound)

i as in machine

o as in go

u as oo in moon


ai as eye

au as ou in house

ei as ey in obey

oi as in noise

Note that ei exists only to force the distinction with e, as in leta (letter) and leita (later, after). This distinction is uncommon.


The i of ai and oi is usually written y before another vowel. Technically, faya (fire) is pronounced FA-ya, but it's normal to say FAI-ya instead. Likewise au before a vowel is written aw: pawa (power).


The accent is on the first syllable.

Consonant mappings

Consonant mappings are mostly obvious; the exceptions are

Th in thin ([T]): t

Th in then ([D]): d

L or r before another consonant: omitted. (But l may be retained in advanced vocabulary.)

Final vocalic l sound (candle, ankle, etc.) -> al: kandal, ankal

Final vocalic r sound (canker, anchor, etc.) -> a: kanka, anka

Optional generic final vowel to help with otherwise final consonants: e.

In general, Inlis restricts consonant clusters. A form such as ?sprinklin should be considered extreme and worth avoiding, though it's technically allowable. Normally, the maximum should be only two consonants per cluster.


(The normal workaround for sCl- and sCr- initial clusters is to insert an a after C: sparinklin. I've not yet decided whether to write that or not; I'm tempted to consider it optional.)

Vowel mappings

Consonant mappings are fairly simple; the vowel mappings are

a TRAP (bad, cab, ham) (I've considered e here, but a works better)

a STRUT (cub, rub, hum)

a BATH (staff, clasp, dance)

a NURSE (hurt, term, work)

a START (far, sharp, farm)

a lettER (beggar, martyr, visor)

a commA (China, sofa, about)


e DRESS (step, ebb, hem)

e(i) FACE (weight, rein, steak)


i KIT (ship, rip, dim)

i FLEECE (seed, key, seize)

i happY (silly, Tony, merry)


o LOT (stop, rob, swan)

o CLOTH (cough, long, gone)

o PALM (calm, bra, father)

o THOUGHT (taut, hawk, broad)

o GOAT (soap, soul, home)

o NORTH (war, storm, for)

o FORCE (floor, coarse, ore)


u CURE (poor, tour, fury)

u FOOT (full, look, could)

u GOOSE (who, group, few)

u intO (influence, situation, bivouac)


ai PRICE (ripe, tribe, aisle)

au MOUTH (pouch, noun, crowd)

oi CHOICE (boy, void, coin)

ia NEAR (beer, pier, fierce)

ea SQUARE (care, air, wear)


Note that "long u" ([ju]) becomes a mere oo sound ([u]) except initially: duti, nu, but yusin. This is because dialects differ on the non-initial pronunciation.


I'll try to justify these mappings next time.

Inlis: History

By way of background, Inlis derived from several realizations.

First, I realized that English wasn't being sufficiently leveraged. It's widely known--one of the only common denominators of the auxlang target demographic--yet this leads to nothing more than a few lexical items in most auxlangs.

Second, Lingua Franca Nova gave me a new paradigm for at-sight (or rather at-hearing) projects: it follows sound rather than writing, yet it is fairly intelligible.

Third, various English-based creoles are likewise intelligible at sight and at hearing (Bislama being my favorite), though they retain typical natlang quirks and limitations that undermine their usefulness. (In general, despite their much-touted ease, creoles are just a bit too random for auxlang use. Intelligent Design, friends! Don't leave home without it!)

Fourth, as I was fiddling with Interglossa, working on a translation of the Babel text, I realized that relexing Interglossa to to English--no, to Bislama--roots would yield an at-sight, at-hearing angloclone: an English-based auxlang.

Such an auxlang would recognize and reward the efforts of those who strive to learn at least spoken English, unlike most auxlangs which ignore or patronize such efforts. It would produce a neutral version of English, an exotic, non-native, global English--standardized, accessible, immediately extensible to nearly the full range of the original if necessary, but limited for beginners.

We'll look at the basic phonological mappings next time.

Inlis: The Rationale

I admit I'm surprised. I hadn't meant to dump Inlis in your collective lap and run off giggling maniacally, but that's roughly what I accomplished. So I was expecting the Spanish Inquisition when I checked in again. (I know: "Nobody expects the Spanish Inquisition!" So I'm a nobody. This is news?)

There should have been a comment explaining that regular English is bad enough without caricaturing it as a kind of creole.

Simply put, the rationale for Inlis derives from the following premises:

1. Most of the people in the auxlang target demographic know English.

2. They don't know it well enough to feel comfortable in it.

3. English as such is too complicated in sounds, vocabulary, and idioms for most learners to achieve effectively native fluency in it.

4. As with most natlangs, the gulf between native and non-native speakers will intimidate learners.

5. In only another generation or so, most people will speak Bad English. It will be bad either because they aren't native speakers; because they natively speak a non-prestige dialect developed almost from a pidgin (Hinglish, Ebonics, whatever); or because they grew up in North America, Britain, or Australia and are out to sabotage English from within.

6. On the other hand, some will speak Funny English, though that's subjective: Brits in the US, Americans in Britain, Australians pretty much anywhere. Funny English can be fun, even cool--useful traits for an auxlang.

Question: What if there were an artificial Funny English that would make us all sound not just like L2 English-speakers but fun/cool ones? Non-native English-speakers wouldn't sound any sillier than natives, and we could all have a good laugh, loosen up, and get on with life.

This is the goal of Inlis. It's meant to be easier and more predictable than English, intelligible to those who already know English to any extent (the investment in learning English isn't wasted), not the property of any group (native English-speakers won't have any real advantage), and about as extensible as English proper.

Next up, I'll explain the history of the project