In 409 AD the last Roman legion left British shores, and in fifty years the Islands became a victim of invaders. Germanic tribes from Southern Scandinavia and Northern Germany, pushed from their densely populated homelands, looked for a new land to settle. At that time the British Isles were inhabited by the Celts and remaining Roman colonists, who failed to organize any resistance against Germanic intruders, and so had to let them settle here. This is how the Old English language was born.

Celtic tribes crossed the Channel and starting to settle in Britain already in the 7th century BC. The very word "Britain" seems to be the name given by the pre-Celtic inhabitants of the island, accepted by first Indo-Europeans. The Celts quickly spread over the island, and only in the north still existed non-Indo-European peoples which are sometimes called "Picts" (the name given by Romans). Picts lived in Scotland and on Shetland Islands and represented the most ancient population of the Isles, the origin of which is unknown. Picts do not seem to leave any features of their language to Indo-European population of Britain - the famous Irish and Welsh initial mutations of consonants can be the only sign of the substratum left by unknown nations of Britain.

See "Picts and Pictish language" article
See "Initial Mutations in Indo-European languages" article

At the time the Celts reached Britain they spoke the common language, close to Gaulish in France. But later, when Celtic tribes occupied Ireland, Northern England, Wales, their tongues were divided according to tribal divisions. These languages will later become Welsh, Irish Gaelic, Cornish, but from that time no signs remained, because the Celts did not invent writing yet.

Not much is left from Celtic languages in English. Though many place names and names for rivers are surely Celtic (like Usk - from Celtic *usce "water", or Avon - from *awin "river"), the morphology and phonetics are untouched by the Celtic influence. Some linguists state that the word down comes from Celtic *dún "down"; other examples of Celtic influence in place names are tne following:
  cothair (a fortress) - Carnarvon
  uisge (water) - Exe, Usk, Esk
  dun, dum (a hill) - Dumbarton, Dumfries, Dunedin
  llan (church) - Llandaff, Llandovery, Llandudno
  coil (forest) - Kilbrook, Killiemore
  kil (church) - Kilbride, Kilmacolm
  ceann (cape) - Kebadre, Kingussie
  inis (island) - Innisfail
  inver (mountain) - Inverness, Inverurie
  bail (house) - Ballantrae, Ballyshannon,
and, certainly, the word whiskey which means the same as Irish uisge "water". But this borrowing took place much later.

In the 1st century AD first Roman colonists begin to penetrate in Britain; Roman legions built roads, camps, founded towns and castres. But still they did not manage to assimilate the Celts, maybe because they lived apart from each other and did not mix. Tens of Latin words in Britain together with many towns, places and hills named by Romans make up the Roman heritage in the Old English. Such cities as Dorchester, Winchester, Lancaster, words like camp, castra, many terms of the Christian religion and several words denoting armaments were borrowed at that time by Britons, and automatically were transferred into the Old English, or Anglo-Saxon language already when there was no Romans in the country.

In 449 the legendary leaders of two Germanic tribes, Hengist and Horsa, achieved British shores on their ships. The Anglo-Saxon conquest, however, lasted for several centuries, and all this period Celtic aborigines moved farther and farther to the west of the island until they manage to fortify in mountainous Wales, in Corwall, and preserved their kingdoms in Scotland. Germanic tribes killed Celtic population, destroyed Celtic and former Roman towns and roads. In the 5th century such cities as Durovern in Kent, Virocon, Trimontii, Camulodunum, were abandoned by the population.

Angles settled around the present-day Noridge, and in Northern England; Saxons, the most numerous of the tribes, occupied all Central England, the south of the island and settled in London (Londinii at that time). Jutes and Frises, who probably came to Britain a bit later, settled on the island of White and in what is now Kent - the word Kent derives from the name of the Celtic tribe Cantii. Soon all these tribes founded their separate kingdoms, which was united after centuries of struggle only in 878 by Alfred, king of Wessex. Before that each of the tribes spoke its language, they were similar to each other but had differences which later became the dialectal peculiarities of Old English.

Now a little bit about the foreign influence in Old English. From the 6th century Christianity start activities in Britain, the Bible is translated into Old English, and quite a lot of terms are borrowed from Latin at that time: many bishops, missionaries and Pope's officials come from Rome. The next group of foreign loanwords were taken from Scandinavian dialects, after the Vikings occupied much of the country in the 9th - 11th centuries. Scandinavian languages were close relatives with Old English, so the mutual influence was strong enough to develop also the Old English morphology, strengthening its analytic processes. Many words in the language were either changed to sound more Scandinavian, or borrowed. Below we will touch this problem in more details.

The Old English language, which has quite a lot of literature monuments, came to the end after the Norman conquest in 1066. The new period was called Middle English.
 

        § 2. The Old English Phonetics.
 
 The phonetic system of Old English preserved in general the Common Germanic structure of sounds. Main phonetic features of Germanic languages - Grimm's Law and Verner's Law - are clearly seen in Old English, as well as many processes which took part among vowels and diphthongs. However, Old English is sometimes moving further in developing the phonetics, and that is why some of its models are a bit hard to trace back to the Common Germanic period. That is why we decided to follow the way suggested by many linguists in description the Old English phonetics: to offer the explanations comparing to the Gothic language, the most ancient of witnessed Germanic tongues and the most archaic of them.

Though Gothic is referred to the East Germanic subgroup of languages, its similar sounds, morphological forms and vocabulary with Old English, Old High German, Old Scandinavian and other ancient Germanic languages are quite frequent. In the early Medieval Age, when both Gothic and Old English were spoken, single languages of the group did not go too far from each other and remained very much alike.

As a whole, Old English phonetics included the majority of sounds typical for all Germanic speech; and only some of its phonemes are unique and require a special acquaintance.

First we will talk about the vowels, which could be either short or long in Old English, just like in Modern English or German. The difference between them is also familiar to Modern English speakers: the distinction between open and close syllables. This distinction is quite Germanic, and obviously did not existed in Proto-Indo-European where vowels could be short and long in every position. In Germanic a long sound in a closed syllable can be seldom met. The table below explains all Old English (OE) vocals compared to those of Gothic and followed by examples (sounds for them will follow later):
 

The Gothic language used to have much more diphthongs than Old English. Usually this is the general trend in Indo-European languages - diphthongs turn into simple vowels, and the more developed the language, the less diphthongs it has. The Old English tongue had two original diphthongs, both of which were composed of long vowel components:

éá - equal to Gothic au, found both in nominal and verbal parts of speech: béám, céás (he chose)
éó - equal to Gothic iu: céósan (to choose); in some dialects and varieties of the language it was written like íó, but we are sure this was the same sound in fact.

One of the main phonological and morphological instruments in Common Germanic and practically in all Germanic languages was the Ablaut, the vowel interchange in the root of nouns and verbs. This specific feature, though known in all Indo-European groups as a phonetic means, was of great importance particularly in Germanic, where it was sometimes used instead of verb endings and noun inflections. Interesting, that the same way of "infixation" of different vowels into the root is known in Semitic and other Afroasiatic languages: compare the Arabic language which has kataba (he wrote), kutiba (written), katib (writing), kitab (a writing), 'aktaba (he made smb write) as different forms of the root k-t-b, and the English language which uses sing, sang, sung, song as different forms of the root s-ng. This is the Germanic Ablaut.
 
The Ablaut in its classical Germanic forms is present in Gothic, Old High German and many other ancient Germanic languages. But English, though keeping this interchange, slightly changed the rules of the Ablaut. For instance, if the Germanic classical verb conjugation represent the Ablaut in i - a - zero forms, in Old English it looks like rísan - rás - rison - risen (to rise) in the I class of Strong verbs, and like béodan - béad - budon - boden in the II class. But still the Ablaut played an important role in the morphology of the verb and noun (for nouns it could be also the means of word composition: faran (to go, to travel) produced the noun fór (a trip)).

Speaking about vowel mutations which took place in Old English words through its period of existence, we do not wish to describe thoroughly all of them, just to point the most important processes:

1. Breaking
  æ > ea before combinations of "r+consonant", "l+cons.", "h+cons.", and also before h final:
        ærm > earm, æld > eald, æhta > eahta, sæh > seah
  e > eo before "r+cons.", "lc, lh, h +cons.", and before h final:
        herte > heorte, melcan > meolcan, selh > seolh, feh > feoh

2. Palatalization
 This is the process which went under the influence of g, c, sc before vowels in the beginning of the word:
  e > ie (gefan > giefan)
  æ > ea (cæster > ceaster)
  æ' > éa (gæ'fon > géafon)
  a > ea (scacan > sceacan)
  o > eo (scort > sceort)
Interesting to know that this palatalization (or softening) is thought by some linguists to influence not vowels but consonants themselves. This means that in some particular position sounds g, c, sc became respectively [g'], [k'], [sk'], and this was marked by a soft vowel after them. So opinions vary on this problem.

3. i-mutation
 This interesting feature changed many of Old English words on a very early stage of the language's history. It is caused by i (ot j) in the next syllable, it affects all vowels, except i and e. Vowels move from their back position to the new front one:
  a > e (framian > fremman)
  æ > e (tælian > tellan)
  á > æ' (lárian>læ'ran)
  o > e (ofstian > efstan)
  ó > é (dómian > déman)
  u > y (fullian > fyllan)
  ú > ý (cúþian > cýþan)
  ea > ie (earmiþu > iermþu)
  éa > íe (geléafian > gelíefan)
  eo > ie (afeorrian > afierran, afyrran)
  éo > íe (getréowi > getríewe, getríve)

4. Back Mutation
   Appears before liquids and labial consonants (i.e. r, l; p, b, f, m):
  i > io (hira > heora)
  e > eo (herot > heorot)
  a > ea (saru > searu)

5. Contraction of vowels due to a dropped h
   After the consonant had dropped, two vowels met, and they collided into one long vowel:
  ah + vowel > eah + vowel > éa, (slahan > sleahan > sléan)
  eh,ih + vowel > éo (sehan > seohan > séon)
  oh + vowel > ó - (fóhan > fón, hóhan > hón)
Once I saw this feature was called the "secondary laryngeal drop" in one book. Really, this reminds the common situation of the Proto-Indo-European laryngeal sound [h] which was dropped in all Indo-European languages except Anatolian (Hittite hartagga, Greek arktos, Welsh art - a bear). The drop of this sound usually caused the lengthening of neighbouring sounds. We see that this trend was preserved in Old English as well - h was not stable enough to remain between vowels. The English language was moving towards the analytism.

The consonants in the Old English language are simple to learn for a nowadays English-speaker - and we are all, aren't we? They look the following way:
 

Consonants could also be subject to several kinds of mutations which we place here:

1. Voicing of fricative sounds (h, f, s, þ) appears, if a fricative is surrounded by vowels:
        wíf (a wife; unvoiced) - wífes (voiced); wearþ (a becoming; unvoiced) - weorðan (to become; voiced).
2. Palatalization appears only in Late Old English, but significantly changes the pronunciation making it closer to today's English:
        cild [kild] > [child]; scip [skip] > [ship]; everywhere [g], [cg] sounds turn into [dj]: bricg [bricg] > [bridj]
3. Other changes
  any velar cons.+ t > -ht-:  sócte > sóhte
  any labial cons.+ t > -ft-:  sceapt > sceaft
  any dental cons.+ t > -ss-:  witte > wisse
  n was lost before h, f, s, p: bronhte > bróhte, sonfte > sófte

Certainly there were other changes as well, but they are not so important to be placed in our short grammar.

In general, Old English phonetics suffered great changes during the whole period from the 5th to the 11th century. Anglo-Saxons did not live in isolation from the world - they contacted with Germanic tribes in France, with Vikings from Scandinavia, with Celtic tribes in Britain, and all these contacts could not but influence the language's pronunciation somehow. Besides, the internal development of the English language after languages of Angles, Saxons and Jutes were unified, was rather fast, and sometimes it took only half a century to change some form of the language or replace it with  another one. That is why we cannot regard the Old English language as the state: it was the constant movement.
 

        § 3. The Old English Substantive.

The substantive in Indo-European has always three main categories which change its forms: the number, the case, the gender. It ias known that the general trend of the Indo-European family is to decrease the number of numbers, cases and genders from the Proto-Indo-European stage to modern languages. Some groups are more conservative and therefore keep many forms, preserving archaic language traits; some are more progressive and lose forms or transform them very quickly. The Old English language, as well as practically all Germanic tongues, is not conservative at all: it generated quite a lot of analytic forms instead of older inflections, and lost many other of them.

Of eight Proto-Indo-European cases, Old English keeps just four which were inherited from the Common Germanic language. In fact, several of original Indo-European noun cases were weak enough to be lost practically in all branches of the family, coinciding with other, stronger cases. The ablative case often was assimilated by the genitive (in Greek, Slavic, Baltic, and Germanic), locative usually merged with dative (Italic, Celtic, Greek), and so did the instrumental case. That is how four cases appeared in Germanic and later in Old English - nominative, genitive, accusative and dative. These four were the most ancient and therefore stable in the system of the Indo-European morphology.

The problem of the Old English instrumental case is rather strange - this case arises quite all of a sudden among Germanic tongues and in some forms is used quite regularly (like in demonstrative pronouns). In Gothic the traces of instrumental and locative though can be found, but are considered as not more than relics. But the Old English must have "recalled" this archaic instrumental, which existed, however, not for too long and disappeared already in the 10th century, even before the Norman conquest and transformation of the English language into its Middle stage.

As for other cases, here is a little pattern of their usage in the Old English syntax.

  1. Genitive - expresses the possessive menaing: whose? of what?
        Also after the expression meaning full of , free of , worthy of , guilty of,  etc.
  2. Dative - expresses the object towards which the action is directed.
        After the after the verbs like "say to smb", "send smb", "give to smb"; "known to smb", "necessary for smth / smb", "close to smb", "peculiar for smth".
        Also in the expressions like  from the enemy, against the wind, on the shore.
  3. Accusative - expresses the object immediately affected by the action (what?), the direct object.

Three genders were strong enough, and only northern dialects could sometimes lose their distinction. But in fact the lose of genders in Middle English happened due to the drop of the case inflections, when words could no longer be distinguished by its endings. As for the numbers, the Old English noun completely lost the dual, which was preserved only in personal pronouns (see later).

All Old English nouns were divided into strong and weak ones, the same as verbs in Germanic. While the first had a branched declension, special endings for different numbers and cases, the weak declension was represented by nouns which were already starting to lose their declension system. The majority of noun stems in Old English should be referred to the strong type. Here are the tables for each stems with some comments - the best way of explaining the grammar.

a-stems
                   Singular
Nom. stán (stone)  scip (ship)  bán (bone)  reced (house)  níeten (ox)
Gen.  stánes           scipes       bánes           recedes           níetenes
Dat.   stáne            scipe         báne            recede             níetene
Acc.  stán              scip           bán              reced               níeten
                  Plural
Nom.  stánas         scipu         bán              reced                níetenu
Gen.    stána          scipa         bána            receda             níetena
Dat.    stánum        scipum      bánum          recedum          níetenum
Acc.   stánas         scipu         bán              reced               níetenu

This type of stems derived from masculine and neuter noun o-stems in Proto-Indo-European. First when I started studying Old English I was irritated all the time because I couldn't get why normal Indo-European o-stems are called a-stems in all books on Old English. I found it a silly and unforgivable mistake until I understood that in Germanic the Indo-European short o became a, and therefore the stem marker was also changed the same way. So the first word here, stán, is masculine, the rest are neuter. The only difference in declension is the plural nominative-accusative, where neuter words lost their endings or have -u, while masculine preserved -as.

A little peculiarity of those words who have the sound [æ] in the stem and say farewell to it in the plural:
          Masculine                         Neuter
     Sing.              Pl.             Sing.               Pl.
N  dæg (day) dagas     fæt (vessel)  fatu
G  dæges       daga       fætes           fata
D  dæge         dagum    fæte            fatum
A  dæg           dagas     fæt              fatu

Examples of a-stems: earm (an arm), eorl, helm (a helmet), hring (a ring), múþ (a mouth); neuter ones - dor (a gate), hof (a courtyard), geoc (a yoke), word, déor (an animal), bearn (a child), géar (a year).

ja-stems
                         Singular
               Masculine                             Neuter
N  hrycg (back)   here (army) ende (end)  cynn (kind)  ríce (realm)
G  hrycges           heriges        endes          cynnes        ríces
D  hrycge            herige          ende           cynne          ríce
A  hrycg              here            ende           cynn            ríce
                         Plural
N  hrycgeas        herigeas       endas         cynn            ríciu
G  hrycgea          herigea        enda           cynna          rícea
D  hrycgium        herigum       endum        cynnum       rícium
A  hrycgeas        herigeas       endas         cynn            ríciu

Again the descendant of Indo-European jo-stem type, known only in masculine and neuter. In fact it is a subbranch of o-stems, complicated by the i before the ending: like Latin lupus and filius. Examples of this type: masculine - wecg (a wedge), bócere (a scholar), fiscere (a fisher); neuter - net, bed, wíte (a punishment).

wa-stems
               Singular                                Plural
      Masc.        Neut.                    Masc.         Neut.
N  bearu (wood)  bealu (evil)   bearwas      bealu (-o)
G  bearwes           bealwes      bearwa        bealwa
D  bearwe            bealwe        bearwum     bealwum
A  bearu (-o)        bealu (-o)    bearwas      bealu (-o)

Just to mention. This is one more peculiarity of good old a-stems with the touch of w in declension. Interesting that the majority of this kind of stems make abstract nouns. Examples: masculine - snáw (snow), þéaw (a custom); neuter - searu (armour), tréow (a tree), cnéw (a knee)

ó-stems
                         Sg.
N  swaþu (trace) fór (journey)  tigol (brick)
G  swaþe            fóre               tigole
D  swaþe            fóre               tigole
A  swaþe            fóre               tigole
                        Pl.
N  swaþa           fóra               tigola
G  swaþa           fóra               tigola
D  swaþum        fórum             tigolum
A  swaþa           fóra               tigola

Another major group of Old English nouns consists only of feminine nouns. Funny but in Indo-European they are called a-stems. But Germanic turned vowels sometimes upside down, and this long a became long o. However, practically no word of this type ends in -o, which was lost or transformed. The special variants of ó-stems are jo- and wo-stems which have practically the same declension but with the corresponding sounds between the root and the ending.

Examples of ó-stems: caru (care), sceamu (shame), onswaru (worry), lufu (love), lár (an instruction), sorg (sorrow), þrág (a season), ides (a woman).
Examples of jó-stems: sibb (peace), ecg (a blade), secg (a sword), hild (a fight), æx (an axe).
Examples of wó-stems: beadu (a battle), nearu (need), læs (a beam).

i-stems
        Masc.                          Neut.
                    Sg.
N  sige (victory) hyll (hill)  sife (sieve)
G  siges              hylles     sifes
D  sige               hylle       sife
A  sige               hyll         sife
                    Pl.
N  sigeas           hyllas      sifu
G  sigea            hylla        sifa
D  sigum           hyllum     sifum
A  sigeas          hyllas       sifu

The tribes and nations were usually of this very type, and were used always in plural: Engle (the Angles), Seaxe (the Saxons), Mierce (the Mercians), Norþymbre (the Northumbrians), Dene (the Danish)
  N Dene
  G Dena (Miercna, Seaxna)
  D Denum
  A Dene

                Fem.
        Sg.            Pl.
N  hyd (hide)   hýde, hýda
G  hýde           hýda
D  hýde           hýdum
A  hýd             hýde, hýda

This kind of stems included all three genders and derived from the same type of Indo-European stems, frequent also in other branches and languages of the family.

Examples: masculine - mere (a sea), mete (food), dæl (a part), giest (a guest), drync (a drink); neuter - spere (a spear); feminine - cwén (a woman), wiht (a thing).

They can be either masculine or feminine. Here it is seen clearly how Old English lost its final -s in endings: Gothic had sunus and handus, while Old English has already sunu and hand respectively. Interesting that dropping final consonants is also a general trend of almost all Indo-European languages. Ancient tongues still keep them everywhere - Greek, Latin, Gothic, Old Prussian, Sanskrit, Old Irish; but later, no matter where a language is situated and what processes it undergoes, final consonants (namely -s, -t, often -m, -n) disappear, remaining nowadays only in the two Baltic languages and in New Greek.

Examples:  masculine - wudu (wood), medu (honey), weald (forest), sumor (a summer); fem. - nosu (a nose), flór (a floor).

The other type of nouns according to their declension was the group of Weak nouns, derived from n-nouns is Common Germanic. Their declension is simple and stable, having special endings:

      Masc.            Fem.                  Neut.
                    Sg.
N  nama (name) cwene (woman)  éage (eye)
G  naman           cwenan               éagan
D  naman           cwenan               éagan
A  naman           cwenan               éage
                    Pl.
N  naman          cwenan               éagan
G  namena        cwenena             éagena
D  namum         cwenum              éagum
A  naman          cwenan               éagan

Examples: masc. - guma (a man), wita (a wizard), steorra (a star), móna (the Moon), déma (a judge); fem. - eorþe (Earth), heorte (a heart), sunne (Sun); neut. - éare (an ear).

And now the last (but least?) one which is interesting due to its special Germanic structure. I am speaking about the root-stems which according to Germanic laws of Ablaut, change the root vowel during the declension. In Modern English such words still exist, and we all know them: goose - geese, tooth - teeth, foot - feet, mouse - mice etc. At school they were a nightmare for me, now they are an Old English grammar. Besides, in Old English time they were far more numerous in the language.

           Masc.                                           Fem.
                          Sg.
N  mann      fót (foot) tóþ (tooth)  | hnutu (nut)  bóc (book)  gós (goose)  mús (mouse) burg (burg)
G  mannes   fótes       tóþes          | hnute          bóce            góse             múse            burge
D  menn      fét           téþ             | hnyte          béc              gés               mýs              byrig
A  mann      fót          tóþ             | hnutu          bók              gós               mús              burg
                          Pl.
N  menn      fét          téþ              | hnyte          béc             gés               mýs              byrig
G  manna    fóta        tóþa            | hnuta          bóca            gósa            músa             burga
D  mannum fótum     tóþum          | hnutum       bócum         gósum          músum          burgum
A  menn      fét          téþ              | hnyte          béc             gés               mýs               byrig

See the rule? The general rule is the so-called i-mutation, which changes the vowel. The conversion table looks as follows and never fails - it is universally right both for verbs and nouns. The table of i-mutation changes remains above.
 
Examples: fem. - wífman (a woman), ác (an oak), gát (a goat), bróc (breeches), wlóh (seam), dung (a dungeon), furh (a furrow), sulh (a plough), grut (gruel), lús (a louse), þrul (a basket), éa (water), niht (a night),  mæ'gþ (a girl), scrúd (clothes).

There are still some other types of declension, but not too important fro understanding the general image. For example, r-stems denoted the family relatives (dohtor 'a daughter', módor 'a mother' and several others), es-stems usually meant children and cubs (cild 'a child', cealf 'a calf').

The most intriguing question that arises from the picture of the Old English declension is "How to define which words is which kind of stems?". I am sure you are always thinking of this question, the same as I thought myself when first studying Old English. The answer is "I don't know"; because of the loss of many endings all genders, all stems and therefore all nouns mixed in the language, and one has just to learn how to decline this or that word. This mixture was the decisive step of the following transfer of English to the analytic language - when endings are not used, people forget genders and cases.

In any solid dictionary you will be given a noun with its gender and kind of stem. But in general, the declension is similar for all stems. One of the most stable differences of masculine and feminine is the -es (masc.) or -e in genitive singular of the Strong declension.

For you to have at least a general idea after reading so much tables I am giving another table, the general declension system of Old English nouns. Here '-' means a zero ending.
 

ád (masc.) - pile
eofor (masc.) - a boar
fácen (neut.) - crime, evil
feging (fem.) - conjunction
gewrit (neut.) - a letter
prýd (fem.) - pride
pund (neut., weak) - a pound

If you are ready with this, take the adjectives. Answers for the exercise are here.