Abstract

This page introduces an a priori language called Sgai, known informally as SIGIL. The acronym stands for Scenic Intuitive Glomerating Ideal Language, which hints at some of the important aspects of this project. In its development, all efforts have been made to avoid arbitrary linguistic effects so that, like a true adamic language (original and naive), every detail is worked out from first principles. Also, the internal logistics of structure and phonology were carefully tempered by practical and aesthetic measures; it was always intended to be a real human language, not a conworld conlang curiosity.

From early on, various dedicated writing systems were invented, to help clarify phonological concepts. Alongside development too, the various problems and solutions were recorded in book form, to become Language for the World.

NB. As from 2021, this language is being referred to as proto-Sgai, to make way for Sgai 2.0, the language which has evolved from this.

1 Introduction

1.1 Typology

Originally, this language was to be grammatically simple, and users would be free to frame their ideas in a loose, poetic manner. But, as the language developed and other specifications came into play, the grammar (among other things) became naturally more complex.

There are elements of both agglutinative and isolating phrase formation, with clauses generally verb final. There is some case marking, and both left and right branching order of dependencies. In many cases grammatical rules are loose, where semantics can come from almost any meaningful juxtaposition of rootwords and particles. The reason for what sounds like a hybrid grammar, is that different rules are found to operate at different levels, and SIGIL has several levels, from the level of phoneme up to the level of sentence.

1.2 Phonology

It took a few years to arrive at a final phoneme inventory. SIGIL required distinctness but also a reasonably large range of sounds. There are 33 consonants plus a number of co-articulations including affricates. The vowel-space is roughly 2×2×2, with front–back, open–close and rounding. There is vowel nasalization and tone, both for grammatical purposes.

SIGIL may be spoken as an uninterrupted stream without ambiguity. In English and many languages, we often need gaps and contours of stress to separate words and clauses. In SIGIL, regular “phoneme grammar” ensures proper framing and extraction of meaning; there is no equivalent to the ice cream vs I scream problem.

Two outcomes (desirable by-products) of the phonotactics carefully developed, are that word-building is reasonably intuitive, and there is an interesting flow of sounds. A sentence spoken in SIGIL is perfectly natural in terms of pronounceability, but the combination of sounds is unlike any language I have heard.

1.2.1 Inventory

main vowels:	a æ e i o ø u ɯ y
word-building consonants:	ʔh b d f ɡ h hl j k l m n ŋ p ɹ s ʃ t θ v w ɸ x z ʒ
extra/morpheme consonants:	ɾ ʔ qʼ pʼ tʼ ʧ’ ʧ ʤ ʦ ʣ tθ dð pɬ pɸ
other vocalics/syllabics:	ə ɹ l n r
tonemes:	relaxed, rising, high, falling

---

The symbols used here to represent the phonemes are from the International Phonetic Alphabet.
Vowel nasalization may be shown with a following  N, since in most cases sandhi will result in one of [m n ŋ], and finally will usually sound with a following [ŋ].
More precise rendering(s) and options for certain written phonemes:

a  ⇒ [ a ɑ ]

æ  ⇒ [ æ ɛ ]

o  ⇒ [ o ɔ ]

k  ⇒ [ kʰ qʰ ]

p  ⇒ [ pʰ ]

t  ⇒ [ tʰ ]

ɸ  ⇒ [ ɸʷ hw ]

x  ⇒ [ ç x ]

[ə] is always short, and unvoiced after an unvoiced consonant. All other vowels are normally not short.
Tones are shown as a following letter, one of  R H F  for rising, high, falling — rather than as a diacritic upon the vowel.

2 Grammar - words

2.1 Roots & phoneme grammar

All words contain one or more rootwords. A rootword with an active sense has the form [consonant + vowel], and with a passive sense has [consonant + nasalized vowel]. In practice, a passive rootword will become [consonant + vowel + N] where N is one of [m n ŋ].

Each consonant and vowel has a phonaesthemic, biophysical or poetic connotation (see the full list), and in rootwords they combine to give a meaning which is either verbal (an action) or nominal (a thing), depending on the context.

For example [ma] means “cares for”, or “one who cares for” or “mother”, using connotations of [m] “intimacy” and [a] “acceptance”. The passive version is [maN] meaning “is cared for”, or “one who is cared for”, or “child”. (The primary meaning for a rootword must be carefully chosen, in order for it to be useful in word-building, as a fundamental unit of our cognitive world.)

A passive version will exist for all cv roots, receiving the result or influence of a cv action/agent, even if only in the sense of “made to be something by a more active entity”.

2.2 Use of phoneme types

Each phoneme type has a role in the grammar. For example affricates are reserved for phrase tags, ejectives are reserved for verbal person-marking, and various other vowel-less consonants are used in particular ways. Again, there is connotational background to the choice and use of these elements. One outcome of this approach is that there is no ambiguity about where words begin and end in the speech flow.

Repetition infix. After plosives and some other consonants may be inserted a flap [ɾ] which adds connotations of repetition or duplication to the main cv meaning.

2.3 Combining rootwords

A word in the general sense is defined as a compoundWord, containing one or more rootwords. Their combinational sequence engenders a meaning, and between each root may be put a linkPhoneme to show the accumulating relationship in more detail:

linkPhoneme	meaning
nothing	which is, acting like
[θ]	and also, and then, so, to
[ʃ]	involving, using, by
[f]	part of, from, away
[s]	in quantity of (for counted entities)
NOTE there are some rootwords which may be shortened (effectively forming a diphthong with a previous vowel):
[ji]⇒ [i]	small
[ja]⇒ [a] after unrounded vowels	greater
[wa]⇒ [a] after rounded vowels	large
[wu]⇒ [u]	enter
An example can be seen in the language name [səɡai] which means “small actors”.

2.4 Prefixes

Before a compoundWord may be put a class prefix. This is an option to help establish a semantic distinction for the particular combination of rootwords:

[bə]	female
[ɡə]	male
[də]	tree, star, divine (compare the Sumerian determinative d)
[pə]	place
[pr̩]	time

2.5 Suffixes

At the end of a sequence of rootwords and interspersed linkPhonemes is an optional suffix, giving an extra dimension of meaning to the word:

[-f]	mobile, travel, swift
[-fs]	instantaneous, sudden; superlative
[-fʃ]	mixed, various, interactive, thoroughly
[-s]	abstract
[-x]	physical, inanimate, gross
[-θ]	color; sequence, divided, repeating
[-ʃ]	vague, -like, -ish
Finally, a word may have an intensity, showing its degree of substance or influence. The vowels used here are also used (with the same quinary associations) in phrase modifiers (see §3.3 below):
[ʔu]	0° = not
[ʔi]	1° = only just, minimally
[ʔæ]	2° = some
[ʔa]	3° = very, fully, all
[ʔo]	4° = overly, too

2.6 Plural

Coming first in a word is the pluralizer (if necessary). It is simply [sə], and similar to the use of final s in English, it applies to the entire built-up word. In some cases, a plural may take a slightly different meaning to its singular version.

2.7 Summary

3 Grammar - phrases

3.1 Verbs and their marking

Verbs are words whose action is primary to the phrase or sentence. They usually come at the end of a phrase. To show they have prime importance (as verbs) they have a person tag-suffix. This refers to the agent and patient roles filled by words (and maybe phrases) mentioned before the verb. There are four person tags, being ejectives:

[qʼ]	I, me
[pʼ]	you
[tʼ]	they, other
[ʧ’]	something, Fate

There may be two tags after a verb. The first tag points to the agent of the verb, the second to the patient. A single tag assumes patient is “other” (unwritten [tʼ]). In practice only [pʼqʼ] and [tʼqʼ] double tags are used – so “I” am forced to be passive (humble) towards “you” and “others”, by virtue of the more natural pronunciation. There are also many more pronominal possibilities for [qʼ] and [pʼ] than most languages account for.

A phrase may be meaningful with only a verb. If verb personage is obvious from the context, and if there is a phrase modifier (see below), the person tag(s) may be omitted.

3.2 Role tags

A word taking part in a verbal phrase (or even a more complex phrase taking a role in a larger sentence) will have a tag-suffix to show whether it is the agent or patient of the verb. The final plosives are to be clearly aspirated.

[tʰ]	agent, instigator of verbal action
[kʰ]	patient, recipient of verbal action
[ktʰ]	both agent & patient, action upon self

3.3 Phrase modifiers

Phrase modifiers come at the end of a phrase or sentence, and are formed very much like rootwords, their cv syllable having a phonaesthemic consonant but followed by a toned magnitude vowel. They act as clarifying or editorial comments on the statement. There are four kinds, and they may be used in sequence.

If there is only one phraseMod, it closes the entire statement with F a falling tone.
A sequence of phraseMods will form a tonal contour: rise (high (high)) fall.

For example, [fekʰmat’wæF] “wondering if they are looking after the bird”; and [fekʰmat’buF] “it looks after the bird (perhaps by eating it)”.

Phrase modifiers are optional. Or they may contain enough relevant information that they stand alone, without any verbal content beforehand.

3.4 Summary

[still to come ...]

• Grouping tags – clauses, logic, etc.
• Detachable pronouns
• Possession

4 Lexicon

4.1 Small talk

Here written in the Slinseng-Fi script, the IPA, and translated with the derivation in brackets.

4.2 Numbers

4.3 Vocabulary

4.3.1  Phoneme connotations – for building rootwords
4.3.2  Rootwords, 3-way sorted

[... in progress ...]

5 The wiki

Grammar

Although all details of the grammar and usage of Sgai appear in Language for the World, I felt little incentive to translate the often complex data and symbols from the book to HTML code, for this webpage. That, and the fact that the Lexicon was far from complete, put further developments on hold. And then the idea for a wiki arose.

Vocabulary

The task of completing a Lexicon is rather huge, exceeding the time to design and construct the grammar system. It can also be (as many conlangers have found) quite tedious. I have finished a reasonably full set of Rootwords, but only a small number of words and phrases derived from these. So I need help to fill out the language, as life is short, and Sgai is sunny.

Wikidot  

For these reasons I developed a wiki site at host Wikidot – sgai.wikidot.com. It is designed to allow perusal of the main details of the Sgai language, and creation of new words and phrases by interested parties. Please feel free to browse and contribute.

(I have since begun to transfer main points onto this page, so one doesn’t need to browse the relatively slow wiki to become familiar with the language.)

6 Examples

6.1 Tower of Babel excerpt

6.1.1 This is a translation from Genesis XI verses 1 and 2, using various scripts. It will be seen that the Slinseng-Fi script is quite extravagant in its use of space, however lovely a script. For this reason, I developed the highly compressed Slinsen-Yi. The other official script for Sgai is Pranish, an alphabet comparable to Latin in terms of its use of space.

Slinseng-Fi
Slinsen-Yi
Pranish

6.1.2 Here it is transliterated with the IPA. One can perhaps see why a dedicated script was desired from early on, when accurate depiction of phonemes using the IPA can look very awkward.

ʧɹ sbonʃɡaʔatʰ lity ʧa lantykʰ ni-ŋoF ||
ʧɹ ɡan-sæF ʔnn hæufukʰ sjuθvat’ ʔnnF ʧɹ
pnymwa ʦa pyŋ dðu ʃinaɹ ʦokʰ kiθxont’paF ||

6.1.3 The original in biblical English:

And the whole earth was of one language, and of one speech.
And it came to pass, as they journeyed from the east,
that they found a plain in the land of Shinar; and they dwelt there.