From Proto-Germanic to Burgundian: Grammar

Although the Burgundian language can be seen as quite archaic, its grammar was subject to dramatic changes over time, most notably in its nominal system.

In verbs, Burgundian lost several sets of forms, including all synthetic passive forms (nimada 'is taken', nimaidau 'let it be taken' etc.), past optative forms (nemi 'if it had been taken'), and the remains of the dual number (although they are absent as such even from the oldest stages of Burgundian).

The conjugation paradigm, though quite complicated today, also underwent simplification and modification. These processes include the following changes:
- old optative/subjunctive acquired the meaning of the future tense, thus ik nimou now basically means 'I'll take', and not 'I would take', although such forms are still used to express subjunctive mood;
- plural endings of the past tense of weak verbs were simplified, so old forms -idedum, -idedud, -idedun all changed into a uniform ending -edu (thus xausidedum 'we heard' became xouzjedu); in other tenses, however, the forms are distinct, e.g. nimaim 'we'll take' vs. nimait 'you'll take';
- in old weak verbs ending in -jan the last consonant was palatalized, and the letter -j- is present in all forms, in contrast to older stages of the language, where it disappeared in some forms, cf. OB. xausida 'heard' (from xausjan) and NB. xouzjda (from xouzjen).

Declension paradigm also underwent significant changes.
 - All inanimate nouns, including words denoting abstract ideas, natural phenomena, body parts, plants and mushrooms, food and drinks, tools and many other things, became Neuter, losing their specific Masculine or Feminine endings. It is worth noting that most animals (with the notable exception of the word deuz 'animal') preserved their gender.

 - Old Dative came to be used with most of the prepositions, thus forcing the Genitive case out of the language. In Middle Burgundian, Genitive was only used after preposition af 'of'; later on, Dative came to be used even after this preposition. In modern language old Genitive is preserved in some archaic set phrases like Gudis vilja 'God's will' etc.

- The Nominative ending -s was reinterpreted as Masculine marker, so it was lost in Neuter nouns (including those that used to be Masculine or Feminine, like dags 'day' and naxts 'night') and added to some Masculine words were it wasn't present earlier, e.g. brodrs 'brother' vs. older brodar.

- Masculine and Feminine nouns changed their old Nominative plural ending -os for adjectival -i (from earlier -ai), while old -os ending (now -as) became an Accusative marker, by analogy with the Neuter gender, employing Nominative -a in the Accusative. The use of the adjectival ending might have been triggered by the similarly sounding endings in Slavic languages and Latin.

- Vowels forming declensional endings of nouns and adjectives underwent some changes, most notably -o > -a (tuggo 'tongue' > tunga, windros 'eyelashes' > vindras) and -ai > -i (mildai leudos'kind people' > mildi leudi, wibai '(to) the woman' > vibi). Thus, Feminine nouns ending in -a may belong to two different declensions in the modern language, e.g. viba 'woman' (Dative vibi), but meurja 'cockroach' (Dative meurjin).

Pronominal system was also modernized. Firstly, possessive pronouns lost their declension, so they are unchangeable in the modern language. Secondly, a whole paradigm of interrogative, demonstrative and relative pronouns and adverbs was devised, similar to that present in Japanese (da-re 'who', a-re 'he, she', ko-re 'this', so-re 'that' etc.) The stems' meanings are as follows: þ- stands for the close axis of demonstrative pronouns, jen- stands for the far axis, j- has the relative meaning, and p- forms interrogative words, e.g. þar 'here', jenar 'there', jar 'where (relative)', par 'where?'. This system wasn't present in the oldest stages of the language's development, but it had developed by the end of the Middle Burgundian stage.

From Proto-Germanic to Burgundian: Vowels

Vowel system of Burgundian is quite archaic. The oldest known form of the language had the following vowels:
short a, (e), i, (o), u
long e:, i:, o:, u:
diphthongs ai, au, eu

Short e and o only occurred in closed syllables before r and h - technically they were allophones of ai and au. The writing system did not differentiate between short and long vowels, the only exceptions being e: (e) and o: (o) - the two sounds that were nearly always long. The letters i and u signified both short and long vowels, in contrast to Gothic, which had a special digraph for the long i: (ei).

In time, however, this system was modified. Quite early diphthongs started moving towards monophthongs. The pronuncation of vowel digraphs changed in the following way: 
ai: OB. [aɪ] > LOB. [æɪ] > MB. [æ:] (written ai or ae) > NB. [æ ~ ɛ]
au: OB. [aʊ] > LOB. [ɔʊ] (written ) > MB. [u:] > NB. [u]
eu: OB. [eʊ] > LOB. [əʊ] > MB. [ɯ:] > NB. [ɯ]

During Middle Burgundian stage, diphthongs ai and au became e and o respectively in closed syllables, so xaifsts 'quarrel' became xefst, and laistjan 'to follow, obey' became lestjen. The sequence auj became oj, e.g. Froja 'Lord' from Frauja.

The combination of the semi-vowel w and the short i (wi, and also the old sequence qi-) quite quickly came to be pronounced similarly to eu in open syllables and before one consonant. The resulting sound used to be spelled as ui, e.g. suimen 'to swim', kuik 'crazy'. The latest orthography reform, however, changed this state, so all words containing an [ɯ] sound are now spelled with the digraph eu (seumen, keuk, and also eula, leud). It is worth noting that this change did not involve words with a long i:, e.g.
svin  'pig', sviben 'to cease'.

As i and u were the only vowels that had both short and long variants, the whole opposition of long and short vowels was lost as early as in Middle Burgundian. Before that, the short i was nearly always dropped when unstressed, thus twalif 'twelve' became tvalf or tvalb, and iniþ 'sin' became ind.

Unstressed vowels were often lost (especially the short i), sometimes causing radical changes in the structure of a word, e.g. xaubiþ 'head' became xoft, eudilo 'baby' became eula.

Modification of combinations like -adi- and -egi- and mass borrowing from German gave rise to four new diphthongs in Burgundian: ai, ei, oi and au, e.g. gajlings 'relative' (from MB. gadilings), nej 'new' (from OB. neujis or niwjis), grojl 'shoot (of grass) (from OB. growils), galaubjen (from OB. galoubjan, influenced by German glauben).

From Proto-Germanic to Burgundian: Consonants

So, let's now see how the proposed Burgundian language developed over 1,500 years and what it has become (I shall make one final reservation, and later on I will refer to my language as if it were an existing idiom of a people living in Eastern Europe).

Let's first take a look at the phonetic system of the Old Burgundian (which is nearly same as Gothic) language.

Old Burgundian had the following consonants:
labial p, b, f, m
dental t, d, þ, s, z
glides w, j
alveolar r
lateral l
velar k, g, h, q, hw

Old Burgundian preserved the original velar nature of the sounds h and hw, which developed into glottal consonants in Gothic and Vandalic, h being eventually lost in the latter. Later on, however, closer to the Middle Burgundian stage, the sound x became glottal in most environments, except after sonorants and in initial position before r and n. In modern language this x became g: OB. xramns 'raven' > NB. granns, OB. marxs 'horse' > NB. margs, OB. xnakko 'nape' > NB. gnaka, but OB. axma 'spirit' > NB. axma.

It is worth noting that the sound [h] was signified with the letter h in the original Gothic alphabet, but the letter was lost since then and substituted with the one originally signifying the combination [ks], probably under the influence of the Greek χ (cf. the Gothic spelling of the word 'Christ', Xristus). During the Old Burgundian stage, the two letters were used  interchangeably, but later on the letter h came to be used only at the beginning of words, marking its change to a glottal consonant, and finally it was dropped out of the alphabet during the Middle Burgundian stage.

As for the letter 'hw', it was lost already in Late Old Burgundian, as the sound denoted by this letter quite quickly merged with p, passing through a short period of a [ɸ] pronunciation, similar to the -f- of Japanese words like Fuji. This is how the unique Burgundian interrogative stem p-, not found elsewhere in Germanic languages, came to exist, so OB. hwat 'what' became NB. pat, and OB. hwaiwa 'how' became NB. pep (with assimilation).

The plosive b, d, g where pronounced as fricative [β, ð, ɣ] after vowels, being devoiced at the end of the word, e.g. xlaif 'bread', but xlaiba 'loaves of bread'. During later stages of the language's development, these consonants acquired a stop articulation in all positions, but before that they were dropped in the position before -i-, which became the source of the modern 'vowel + glide' combinations. The same happened with w. Thus, OB. ragin 'advice' became NB. rajn, OB. galewjan 'to set free' became NB. galejen, and OB. gadilings 'relative' became NB. gajlings. Old fricatives were also lost between the vowel i and following sonorants, e.g. OB. rigns 'rain' became NB. rin.

Proto-Germanic voiceless fricatives became voiced stops after sonorants as early as in Old Burgundian, for example, *wilþijaz 'wild' became wildis (modern vild).

The distribution of reflexes of devoiced b, d, g in the modern language is quite uneven. While g always yields voiced g, the situation with b and d is more difficult. For example, leufs 'beloved' (stem leub-) became leufs, with an f-reflex, but lif 'life' (stem lib-) became lib, with a b-reflex. As for d, it has regularly been preserved, but sometimes it yielded a voiceless -t at the absolute end of the word, e.g. NB. dut from OB. duþ/dud 'to, for', verbal ending -t from -d/-þ,  and even at the end of the stem, e.g. NB. vitet from OB. witoþ 'law' (from earlier witodus).

The sounds þ, t, d were generally lost before l, for example maþl 'meeting place' became mal, and the word litils 'small' (Middle Burgundian litls) became lil.

The Burgundian language hasn't ever known the phenomenon of vowel umlaut before j, which is observed in all other modern Germanic languages. Instead, Burgundian consonants got palatalized before j, e.g. xauzjan 'to hear' ([xauzjan]) came to be pronounced [xu:ʑen]. Dentals were the first to become palatalized, as seen in the oldest borrowings that have č in their original forms, e.g. tjas 'time' from Proto-Slavic *časъ. Velars weren't palatilized until quite recently.

The sound x was lost between consonants, e.g. OB. alxs 'temple > NB. als, OB. fulxsn 'secret, riddle' > NB. fulsn.

The combination sk became sx very early, quite quickly coming to be pronounced [ʃ]. There are two exceptions from this rule: firstly, sk did not change into sx before j, thus ejsxen 'to be embarassed' (< OB. aiwiskon), but ejskjen 'to make embarassed' (< OB. aiwiskjan). Secondly, the combination skr came to be pronounced [ʃkr], e.g. skreunja ['skreunja] 'earth-house' became ['ʃkrɯ:ɲa].

What Do We Know about the Real Burgundian Language

So, what do we know about the real language of the Burgundians? Even if it hasn't survived, the Burgundians played a notable role in the history of Europe in their time, so their language must have left some traces of existence. Gothic, for example, has quite a big corpus sufficient for a proper description and research of the language. Vandalic has several phrases like  'scapia matzia ia drincan' recorded by Roman authors. The Burgundian language, unfortunately, wasn't so lucky.

Nevertheless, the data that we have is sufficient to classify the Burgundian language as belonging to the East Germanic branch and even to have some idea of its specific features.

The only surviving text in Burgundian is a short phrase on a fibula found in Charnay. It has been read as uþ fnþai Iddan Liano, which supposedly means 'may Liano find Idda'. Of course, the reading is not one hundred per cent correct, and the inscription itself is not necessarily made in Burgundian.

Ammianus (325/330-391) gives two Burgundian words belonging to the administrative sphere: sinistus 'chief priest' and hendinos 'king'. The former is clearly *sinists (cf. Gothic sinista 'elder') while the latter is hard to reconstruct, unless it is connected with the Germanic root *hind- 'back, behind' (with the suffix *-īn-?) or the word *kinþiz 'kind, race'.

Some Burgundian words are present in the Latin text of Lex Romana Burgundionum, which was supposedly written during the reign of King Gundobad, including wittimon 'marriage', morginagiva 'morning gift' and some others.

Even with these modern sources some conclusions can be made. Thus, Burgundian clearly was an East Germanic, as it demonstrates features typical for Gothic and (as far as one can judge) Vandalic:
- lack of rhotacism, typical for all other branches (Gaisaldus, but English gar 'spear'),
- development of *e to i in all positions except before *r, *h and *hʷ (Gibica, but German geben),
- lack of umlaut,
- lack of development of the Proto-Germanic long *ē to *ǣ/*ā (Leubaredus, but German Rat).

So, we now see that Burgundian can be safely classified as an East Germanic language. Using the abovementioned sources, we can also say something about particular features of the Burgundian language:
- Like in other East Germanic languages, voiced b, d, g were rendered as fricatives between vowels (morginagiva);
- Unlike Gothic, Burgundian preserved the Proto-Germanic diphthong *eu (Eudilana, Reudilo);
- Unlike Gothic and Vandalic, Burgundian preserved the long *ē, and *ō, which developed into ī and ū in those language (Wenaharii, Hrodehildis);
- Unlike Gothic, in Burgundian the sound *n did not disappear before *n, cf. Burgundian Hanhavald, but Gothic hahan 'to hang (akin with English hang);
- It seems that Burgundian preserved the Proto-Germanic diphthongs *ai and *au, which developed into ɛ and ɔ in Gothic (Gaisaberga, Baudulfum);
- Medial *-þ- was voiced to d in Burgundian (the are virtually no names with -th- between vowels).

There are also some words that are attested in Burgundian names but are absent from the Gothic corpus. Those include *hadus 'battle', *saims 'blood clot', *hroks 'crow', *gunds 'war, battle', *erps 'dark, brown' and a few others.

All these things were taken into account and played their part in the creation of the Neo-Burgundian language.

Basic Things: Negation

There are several types of negation.

The most simple one is the use of the conjunction ni 'not': ik ni vait þan guman 'I don't know this man'. This conjunction blends with the auxiliary verb visen 'to be' in all forms, so the negative forms are as follows: sa tjons nist in s'ruma 'the dog isn't in the room', izi atta nivas xoux 'her father wasn't tall', ik nisjou slepen xisnaxts 'I won't sleep tonight'.

Before nouns, another negative word is used, nain 'no, none', e.g. ik im nain guma, ik viba 'I'm not a man, I'm a woman'.

It should be mentioned here that the verb 'to be' may be dropped in constructions like 'I'm a woman', when the second part of the sentence is a noun. However, when we have an adjective in the second part, e.g. if we want to say 'I'm young', the verb is needed: ik im jung.

Other negative words include nevr 'never', niman 'no one', nain 'none', nipar 'nowhere'. Double negation usually does not occur, so phrases like 'I never drink coffee' sound like ik nevr drinka kafa, employing a single negative word.

Basic Things: Numerals

The set of cardinal numbers is typical for Germanic languages:
1 - ain
2 - tvai
3 - tri
4 - fidur
5 - fim
6 - ses
7 - sim
8 - axtu
9 - neun
10 - tain

Cardinal numbers from 11 to 19 are formed adding -ten to the stem of cardinal numbers (some changes occur in the process). The exceptions are the numbers 11 and 12:
11 enlif
12 tvalf
13 triten
14 firten
15 fimten
16 sesten
17 simten
18 axten
19 neunten

Tens are formed with addition of -tjus to the stem of cardinal numbers:
20 tvaitjus
30 tritjus
40 firtjus
50 fimtjus
60 sestjus
70 simtjus
80 axtjus
90 neuntjus

Cardinal numbers from 21 to 99 are formed in a manner similar to the one used in English:
24 tvaitjus fidur
62 sestjus tvai
98 neuntjus axtu

100 is xundred. This is a noun; when hundreds as numbers are formed, the form is -xund:
100 ainxund
200 tvaixund
300 trixund
400 firxund
500 fimxund
600 sesxund
700 simxund
800 axtxund
900 neunxund

1000 is þusend, so thousands go as follows:
1000 ain þusend
2000 tvai þusend
3000 tri þusend
4000 fidur þusend
5000 fim þusend
6000 ses þusend
7000 sim þusend
8000 axtu þusend
etc.

1 000 000 is milljon, 1 000 000 000 is milljard.
Large cardinal numbers are formed in the following way:
2 675 346 - tvai milljon sesxund simtjus fim þusend trixund firtjus ses.

Ordinal numbers are formed with the suffix -d/-t: sest 'sixth', tvaitjus-fidurd 'twenty-fourth'. The only exceptions are: frum 'first', andr 'second' and tridj 'third'.

Fraction numerals are formed with the suffix -(e)da with the exception of 1/2 (xalb), 1/3 (tridail) and 1/4 (firdail): tri simdas 'three sevenths', fim axtedas 'five eights'. One tenth is tenda, not *taineda.

Basic Things: Pronunciation Details

The reading rules have been roughly described in one of the earliest posts, and now it's time to describe the phonological processes and other details present in the language.

First of all, it is worth noting that there are sounds that might seem identical to other ones. Thus, there are two 'u' sounds (u [ʊ] and ou [u], in addition to the quite similiar sound eu [ɯ], which sounds both like an 'i' and an 'u'), two 'e' sounds (e [e] and ai [æ ~ ɛ]), two 'sh' sounds (sj [ɕ] and sx [ʃ]), two 'l' sounds (l [ɫ] and lj [lʲ]), two 'ch' (tj [t͡ɕ], kj [c]) and two 'dj' (dj [d͡ʑ], gj [ɟ]) sounds, which makes the overall phonetic system quite complex.

Nevertheless, the difference between the 'identical' sounds must be observed, as different sounds sometimels yield different meanings, as is the case with English 'stuff' and 'staff', 'slit' and 'sleet'.

A short list of such 'paronyms' includes the following:
buna 'scene' and bouna 'bean',
kjeka 'doe' and tjeka 'Czech',
tila 'aim' and tilja 'line',
xugjen 'to count' and xudjen 'to hide',
graidjen 'to order' and gredjen 'to starve'.

As for the details of pronunciation, they are given below.
- The letter j not only makes the preceding consonant soft - it also makes the following vowel closer narrower and closer to the front. Thus, the word tjons 'dog' is pronounced [t͡ɕøns], and the word gjuka 'parable' sounds like ['ɟʏka].
- The letter j may in fact affect consonants that weren't listed in the first post. For example, þafjen 'to feel' and farbjen 'to colour' may be pronounced either as ['θafjen] and ['farbjen] or as ['θafʲen] and ['farbʲen], with the latter being the more preferable variant.
- The letter x has two pronunciations: it is [ħ] or [χ] at the end of closed syllables and [h] in all other cases, so xand 'hand' is [hand], but naxt 'night' is [naħt] or [naχt] ([ħ] is a voiceless pharyngeal consonant found in Arabic (ح) and some other languages, mainly Afro-Asiatic and North Caucasian). This letter is always pronounced as velar [x] before ou, e.g. xoun [xu:n] 'low'.
- The sound [n] becomes [ŋ] before g and k, like in other Germanic languages.
- The letter l, which is regularly pronounced as a 'dark L' ([ɫ]) is softened to [l] before front vowels, so the word lib 'life' sounds almost as if it where spelled *ljib.
- The vowel combination ai has two distinct pronouncations: it is [ɛ] before two or more consonants and [æ] in other cases. Thus, laisvext 'school subject' is pronounced ['ɫɛsveħt], xaim 'home' is [hæ:m], and aiz 'copper' is [æ:z].
- Old diphthongs (ai, eu, ou) are usually long in open syllables and in monosyllabic words ending in one consonant, so faik 'denial' tends to be pronounced [fæ:k], and eula 'baby' is [ɯ:la].
- Voiceless plosives (p, t, k) are aspirated and more tense than voiced plosives, a phenomenon similar to what is observed in other Germanic languages. Voiced plosives (b, d, g) are devoiced in the final position, but, unlike the corresponding sounds in German, they don't sound as aspirated and tense as the voiceless ones, so the words þat 'this' and þad 'here' really sound differently, the former being [θatʰ], and the latter [θat].
- Voiced plosives are not devoiced in the final position if they follow a nasal consonant. In this case, two consonants blend and yield a prenasalized stop. Thus, and 'end' is pronounced [an͜d], not *[ant].
- Syllabic sonorants are not uncommon. Thus, some words are pronounced with a sonorant as the nucleus of the syllable, e.g. fingr ['fɪŋgr̩] 'finger', regl ['regɫ̩] 'rule', sokn ['sɔkn̩] 'search'.
- The letter v has two pronunciations: [v] and [ɪ]. The former is used in most cases, while the latter is only used to transcribe Greek words with 'υ' or other non-native words with 'y', eg. fysika 'physics', krystal 'crystal', ryþm 'rhythm'.

Stress
The stress is typical for a Germanic language: in native words, it falls on the first syllable of the word for most nouns and on the first syllable of the root for verbs, adjective, participles and verbal nouns. In complex words, double stress is frequent, e.g. firdailfinal [ˌfɪrdɛɫfɪ'naɫ] 'quarter final'.

Basic Things: Use of Articles

As it has been said before, there are two articles: the definite sa and the indefinite ain. It should be noted that they aren't used as frequently as in English or French. As is the case with the continuous verbal forms, use of the two articles is largely optional, especially in spoken language, where they aren't always consistently used.

There are, however, certain cases, where the use of the articles is obligatory:
- Before subordinate clauses beginning with relative pronounes (je, jan, jar etc.), e.g. sa xuz, in jam ik liba 'the house in which I live', ain boka, je ik las 'one of the books that I have read';
- When a countable object is introduced for the first time, the indefinite article is obligatory (ain guma inlidit sa rum 'a man enters the room');
- Once a countable object has been introduced, the next time it is used with sa: ik nevr sap sa guman erza 'I've never seen the man before'.

In other cases, the countable objects may be referred to without any articles, sometimes with the use of demonstrative pronouns or words like sum 'some', engen 'any, some kind of'.

There are some cases, when absence of articles is obligatory:
- Abstract nouns. There are virtually no cases in which they have an article, cf. sxouni, je ik in izi sap with English 'the beauty that I saw in her';
- In Genitive constructions, e.g. tjons a Poula 'Paul's dog', and a stillin 'the end of silence'.
- When a countable noun is referred to as an abstract category, and not as a discrete object. Consider the following sentence: min aiþi ist doktors 'my mother is a doctor'. In this case the word doktors signifies a profession, and not a healthcare worker. The situation is different in the following context: min aiþi ist ain doktors in s'nepst seukxuza 'my mother is a doctor at the nearest hospital', because in this case the word means 'one of the employees of a particular institution'.

Similar difference is observed in the following example: frjonds ist xispas, jas at þis pesjt 'a friend is someone who cares about you' vs. is ist ain frjonds f'Poula 'he is a friend of Paul's',

Basic Things: Function Words and Use of Cases

As it has already been said, there are three cases: Nominative, Dative, and Accusative. In older stages of the language there used to be as many as five of them: Genitive and Vocative did not make it to this day.

All the constructions with the Genitive case where substituted by the combination of the preposition f' and the word in the Dative case, e.g. f'Petra keuna 'Peter's wife', ougna f'min aiþin 'my mother's eyes'. As seen in the examples, the word in the Dative ('Peter' in the former, 'my mother' in the latter) may be placed either before the word in the Nominative or after it. The former variant is a bit more preferable, while the second is more often use when the group in the Genitive is too long, e.g. sxini f'þin granim sxounim virdim ougnam 'the glance of your beautiful big green eyes'.

As for the Vocative case, although it seems to have disappeared many centuries ago, there are special vocative forms for masculine words of the first declension (those ending in -s) in the modern language. Those words drop the final -s in the Nominative when the object is addressed, so the form becomes identical to that of the Accusative case. For example, the word 'friend' is frjonds in the general Nominative, but when we address somebody with the words 'my friend!', the word drops the Nominative ending: min frjond!

Moreover, the forms without -s are used in forms of address and titles even when the person in question is not being directly addressed. Thus, the word doktors 'doctor' becomes doktor when it precedes a name: doktor Jons 'Dr. Jons'. Some examples of this phenomenon include kung Gundar 'King Gunther', professor Kvernjens 'Professor Kvernjens', froj Simonsun 'Mr. Simonsun' etc.

The same is true for given names: they generally don't have the -s ending, so Peter is Petr (not *Petrs),  John is Joxan or Jon (not *Joxans, Jons), and James is Jakob (not *Jakobus). The only exceptions are the names Maþais 'Matthew' and Andrais 'Andrew' - they drop the final -s in the Dative and the Accusative.

As for the cases existing in the modern language, their usage is as following:
- Nominative - as the subject of a sentence (sa burs itit ain apl 'the boy eats an apple');

- Dative - after a preposition (sa burs ist in s'xuza 'the boy is in the house') or - only for pronouns - when the sense of the Dative case is implied (Ik gab þis ain sxilling 'I gave you one schilling')
- Accusative - after a verb (sa burs itit ain apl 'the boy eats an apple') or after prepositions of movement (in apa doupjen 'to dive into the river', þer gaba gangen 'to go through the forest').

Different prepositions can be seen in the table below.



In addition to these, there is also an important preposition at meaning 'about', e.g. ain spill at drakam 'a story about dragons'.

A couple of words should be said about the preposition f' 'of'. It has an alternative form, a, which is used to break bulky consonant clusters, e.g. and a stillin 'end of silence' (not *and f'stillin). This form is also used before words beginning with þ-, f-, v-, e.g. xivi a vina 'the colour of wine'.

When the preposition f' precedes the definite article sa, they blend together, a contracted form as is used: vorda as poetin 'the poet's words'. This is similar to French forms 'du' (de + le) and 'des' (de + les) and German forms 'vor' (von + der), 'vom' (von + dem).

Conjunctions include the following words: i(d), jax'and', eþ 'or', ibi 'if' þo, ak 'but', unte, þej 'because', und 'for', i 'that', sveþox, þobi,  jabi 'although', dut 'to, so that', þi 'than'.

There are two words for 'and': i(d) and jax (i becomes id before vowels). There is some difference in how they are used:
- When more than two objects are enumerated (my mother, my father, my sister and my brother), the word jax (not i) is used before the last enumerated word: min aiþi, min atta, min svistra jax min brodrs.

- When only two objects are enumerated (war and peace), both words are acceptable, but jax is a little more preferable than i in this case.
- With verbs, i is used two equivalent parts of a sentence: ik im drinkend kafa i sxounend TV 'I'm drinking coffee and watching TV'.
- When these words connect two parts of a complex sentence, the meaning expressed by them differs: i signifies a cause-and-effect relationship (is kam, i si idja 'he came, so she left'), while jax signifies an opposition (is kam, jax si idja 'he came, while she left'). In this aspect, the difference is the same as the one between the Russian words и and а.

As for the two words with the meaning 'but', the difference between them is more subtle, but one can say that ak is generally used to denote a stronger contrast than þo, being closer to English words like 'however', 'although' etc.

The most important interjections include the following: ja 'yes', ne 'no', ej or oj 'hey', nu 'well'.

Basic Things: Complex Sentences and Word Order

Basic word order and subordinate clauses
The basic word order is SVO, just like in most of the European languages, e.g. ik xazja banana 'I like bananas'. In subordinate clauses, however, the verb is usually placed at the end of the sentence, e.g. is kam und mis anxaims nimen 'he's come to take me home'.

Placing the verb at the end of the sentence is not obligatory if the group accompanying the word is too long, e.g. is kam und mis nimen t'unsr granam aldam xuza 'he's come to take me to our big old house'. In this example there are three meaningful words before the objects, which would make the sentence bulky in case of the verb being placed at the very end.

The verb generally does not appear after the object in subordinary clauses after the conjunction i 'that', e.g. ik vait, i si friet mis 'I know that she loves me', but also (occasionally, mainly in written language) ik vait, i si mis friet.

Inversion
Inversion is also present. It occurs in two cases:
- Before the enclitic particle ba 'if', e.g. vissa ba ik þat vord, ik maxta it in min rodin botjen 'if I knew this word, I would use it in my speech';
- In interrogatory phrases (see below).

Interrogatory phrases
Interrogatory sentences are basically formed with the help of inversion, just like in English, German or French: vaist þu min nama? 'do you know my name?'

Indirect questions have no inversion, e.g. eskje ina, jefle jera is aig 'ask him, how old he is'. It is important to note that all the interrogatory words beginning with a are changed into relative ones beginning in j-: thus, pe 'what' becomes je etc.

Questions beginning with the word an 'whether, really' also have inversion: an axjes þu, je þu keudend is? 'do you really mean what you're saying?'

In spoken language, the interrogatory character of a phrase may be shown with intonation.

Basic Things: Tenses, Moods and Voices

There are three grammatic tenses: past, present, and future. The grammatical markers of the tenses are given in the material on verbs.

The perfective/imperfective aspect is not directly expressed, in contrast to the existence of the  'auxiliary verb + past participle' forms typical for other European languages. For these purposes, the word noux 'yet' and jut 'already' are generally used. Some verbs have special perfective counterparts with ga- or other prefixes, for example bringen and gabringen, both meaning 'to bring'. It should be noted, however, that the prefix ga- not always has the perfective meaning, cf. bergen 'to keep, guard' and gabergen 'to accumulate, economize'.

There is, however, a special way of expressing perfective aspect for intransitive verbs. This includes a combination of the verb visen 'to be' in the present tense and the past participle of the corresponding verb. For example, to express the idea 'I've got tired' one can use the verb visen in the present tense and add the past participle of the verb afmajen 'to get tired' to it: ik im afmajd.

Unlike the perfective aspect, the continuous aspect is very frequently expressed and underlined. The formula for the continuous verb forms is the same as in English: the verb visen + active participle, for example ik im itend 'I'm eating'. The same is right for the past and the future tense, where one can say ik vas itend 'I was eating' or ik sjou itend 'I will be eating'.

One should note that there are two ways of expressing the future tense. One of them has already been discussed: there is a special set of verbal endings which descend from old subjunctive/optative forms. There is, however, an analytical formula for the future tense, which is as follows: the verb visen in the future + infinitive, for example ik sjou iten 'I'll eat'.

The general difference between these two forms can be roughly described as follows:
- The analytical form can be described as used to refer to the further future than the synthetic one, e.g. if we want to say 'I will be studying at the university after five years', it is more correct to use the former (ik sjou in s'universitaita studiren afta fim jeram), while if we want to say 'I'm going to Greece this summer', the synthetic formula is better (ik gangou ta Kreklanda þat axsen).
- The synthetical form denotes an action occurring only onse, while the analytical one underlines the repeated or continuous character of the action. For example, if we want to say 'I'll jump', ik leupou will mean 'I will make one jumping movement', while ik sjou leupen will mean 'I will be jumping, I will make many jumping movements'.
- The synthetical form might be used to express sunbjunctive mood, while the analytical one can not. For example, if we say is keumai, it could mean not only "he will come", but also "he might come" or "he would come". The use of the synthetic plural in the place of subjunctive is bookish and does not occur in the spoken language.

The passive forms of verbs are analytical, consisting of the auxiliary verb visen and the past participle, for example 'I'm being helped' is ik im xilped, and 'the house will be burnt' is sa xuz sjai gabrannjed.

As for the moods, there are three of them: indicative, subjunctive, and imperative.

The imperative mood has its own grammatical form, which is the pure stem of the verb, e.g. sxak 'move!' from the verb sxaken 'to move'. The verbs with infinitives ending in and have imperatives on : e.g. xouzje 'hear!' from the verb xouzjen 'to hear' and sorne 'take care!' from the verb sornen 'to take care'.

The subjunctive mood forms are more or less identical to that of the indicative mood. In each case a separate construction is used. The most important ones are given below (on the example of "If you give me water, I'll drink it").
ibi þu mis vata gibais, ik drinkou it - 'if you give me water, I'll drink it' - for something that can still happen (both verbs are in subjunctive mood; the word ibi 'if' is used)
gabt ba þu mis vata, ik maxta it drinken - 'if you gave (had given) me water, I would drink (have drunk) it' - for something that can't happen any more, but once could have (unreal condition) (first verb is in the past with the enclitic ba 'if', the second verb is used with the modal verb magen in the past)
ibi ains mis vata gibt, ik drinka it - 'if someone gives me water, I drink it' - for something that usually happens (real condition) (both verbs are in the present, the word ibi 'if' is used)
ik maxta vata drinken - 'I would drink water' - "pure" subjunctive (more colloquial)
ik drinkou vata - 'I would drink water' - "pure" subjunctive (more bookish)

Basic Things: Irregular Verbs

Irregular verbs include the following groups:
- preterite-present verbs, which have past endings of strong verbs in the present and past endings of weak verbs in the past (most of them are modal verbs);
- irregular verb viljen 'to want', which has future endings in the present;
- irregular verb visen 'to be', which has its own set of endings

The inflectional paradigm of the preterite-present verbs is as follows (e.g. the verb viten 'to know'):

person	Present	Past	Future	example Present	example Past	example Future
1s (I)	-0	-ta/-da	-ou	ik kann	ik kunda	-
2s (thou)	-t	-ta/-da	-ais	þu kannt	þu kunda	-
3s (he/she/it)	-0	-ta/-da	-ai	is kannt	is kunda	-
1p (we)	-u	-(t)edu	-aim	vis kunnu	vis kunnedu	-
2p (you)	-u	-(t)edu	-ait	jus kunnu	jus kunnedu	-
3p (they)	-u	-(t)edu	-ain	jai kunnu	jai kunnedu	-

The following verbs belong to the preterite-present group:
kunnen 'can' (Present: singular kann-, plural kunnu, Past: singular kunda, plural kunnedu)
magen 'may' (Present: singular mag-, plural magu, Past: singular maxta, plural maxtedu)
viten 'to know' (Present: singular vait- (þu vaist), plural vitu, Past: singular vissa, plural vissedu)
þorben 'to need' (Present: singular þarb (þu þarft), plural þorbu, Past: singular þarfta, plural þarftedu)
sxulen 'should' (Present: singular sxal, plural sxulu, Past: singular sxulda, plural sxuledu)
gadorsen 'to dare' (Present: singular gadars, plural gadorsu, Past: singular gadorsta, plural gadorsedu)
aigen 'to have' (Present: singular aig, plural aigu, Past: singular exta, plural extedu)

There are also two words that used to be preterite-present verbs, but became unchangeable in the modern language: mot 'must' and doug 'to be able'. They descend from the old verbs *motan and *dugan, but today their usage is that of adjectives, e.g. is vas mot it touen 'he should have done it', is vas doug it touen 'he was able to do it'.

The verb viljen 'to want' has the same endings as the preterite-present verbs with the exception of the present tense, where it uses the 'normal' future endings:

person	Present	Past
1s (I)	ik viljou	ik vulda
2s (thou)	þu viljais	þu vulda
3s (he/she/it)	is viljai	is vulda
1p (we)	vis viljaim	vis vuljedu
2p (you)	jus viljait	jus vuljedu
3p (they)	jai viljain	jai vuljedu

As for the verb visen 'to be', it has its own paradigm, quite different from that of the rest of verbs.

person	Present	Past	Future
1s (I)	im	vas	sjou
2s (thou)	is	vast	sjais
3s (he/she/it)	ist	vas	sjai
1p (we)	sum	vesu	sjaim
2p (you)	sut	vesu	sjait
3p (they)	sind	vesu	sjain

In addition to irregular verbs, there is one weak verb, gangen 'to go', which has a suppletive paradigm: it is ganga, gangs, gangt etc. in the present, but idja (singular) and idju (plural) in the past. The participles are nevertheless formed regularly: gangend 'going' and gangd 'gone'.