The issue of Proto-Indo-European dialects is greatly complicated and
contains very many theories which suggest different forms of dialectal
division in Proto-times. The comparative method allows us to define
which groups of its language family are most similar to each other, lexical,
phonetic and morphological similarities creating the basis used in proclaiming
that "this group" at one time came from the same branch as "that group".
We have already presented an essay called Indo-European
Proto-Dialects online, so we will not be retelling the history of Indo-European
dialectology, but will touch on just a part of it -- the theory
of "Proto-European".
The first theory which appeared in the comparative linguistics of the
19th century was the division between "Eastern" and "Western" languages.
It was based mainly on a fact which was discovered very early by linguists:
the reflex of palatal Indo-European consonants k' and g',
according to which all tongues were divided into "satem" and "kentum" languages.
However, it has of late been proved that such division did not take place
in the Proto-Indo-European community, and moreover the very theory had
begun to appear not quite right. Some branches (Slavic, Baltic, Iranian)
gave two kinds of reflexes which showed both s, z and k,
g respectively, so that these branches cannot be regarded as purely
satem languages. On the other hand, this kind of division seems a
bit strange from a geographical point of view: the "kentum" representatives
were Celts in Western Europe and Tocharians in Central Asia, while satem
comprised the Indic and Baltic languages.
In order to make the theory of Proto-dialects more justified geographically,
the version of "European" languages was invented. It also appeared
in the previous century and was first based mainly on the Italic, Germanic
and Celtic languages, for other branches in Europe (Slavic, Baltic, Illyrian,
etc.) had not been studied enough at that time. The first book which
had touched a bit on the Slavic and Baltic material was Bopp's "Comparative
Grammar", finished and published in 1849. But even after the questionable
elements it contained became well known, the "European" theory continued
to exist in Indo-European dialectology. And even nowadays, when the
majority of proofs gathered by the supporters of the "European" proto-dialect
have been defeated, some linguists still believe it. Is it worth
believing, then?
Lexicon is the first foundation stone in this theory. A number
of words which can be found in Celtic, Italic, Germanic, and sometimes
also Baltic or Slavic, are not present (or not yet found) in Asiatic languages
such as Iranian, Indic or Anatolian. An interesting exception is
Greek, which was never regarded as a branch of "Proto-European" -- it had
appeared much closer to Indo-Iranian than to Germanic or Celtic.
We will try to analyze a pair of words which used to be or are still
regarded as solely European. This list can be called a lexicon of the "Proto-European
language", though it consists of just a few terms. Before drowning
ourselves in the sea of etymology, we must say that every (!) language
possesses a number of words which are unique and cannot be met in other
tongues, even in close relatives. English, for example, has the verb
clepe, "to call, to name", the origin of which is the Middle
English clepen, from the Old English clipian,
cleopian - no cognates in other languages, nowwhere in the
world. The same can be said about words which exist in only a pair
of tongues or in a single group of languages, but which either were lost
or had never existed in those others. The example here is the Celtic
*makwo- (Irish mac, Gaulish maponos,
Welsh mab).
Still, the small collection below is of great interest to us.
1. *more, *mare
The word is curious just for its possible proto-form with *a
in the root. It is known, that Indo-European vocalic system contained the
basic structure of ablaut which included *e and *o
vowels and also two semi-vowels *i and *u which
could be either vowels or sonants in the word. The number of roots with
*a was quite scarce in Proto-Indo-European, and, according
to Meillet, are mostly words from the popular speech, vulgarisms.
However, *mare, *more, *mar- is found in:
Latin mare (sea),
Lithuanian marios, mare.s (the Baltic Sea, the Curonian
Bay), Latvian mare (the Baltic Sea), Old Prussian mary
(a bay),
Old High German meri (a sea), Gothic marei
(a sea),
Irish muir (a sea), Gaulish more (a sea;
in Armorica < *are-more-),
Slavic *more (a sea);
another semantic chain:
Old High German muor (a bog), Old English mo'r
(marsh), Old Frisian ma'r (a pool), English marsh,
German Marsch and Moor; in several Slavic languages
the word more is used to denote a lake or a river: North
Russian Chudskoe more (lake Chudskoe), Bulgarian Maritsa
(the name of a marshy river), Slovak morske oko (a little
mountain lake).
2. *touta, *teuta'-
This stem is quite peculiar: its spreading over European languages
seems a little bit strange. It is present in the Italic group, but only
in one subgroup of it; it exists in all known Baltic languages but is absent
in Slavic, a language closely related to Baltic. The noun this root denotes
is always feminine. See the cognates:
Oscan touto (people, a tribe, Latin civitas), Umbrian
totam (a tribe, acc.sg.)
Lithuanian tauta (land, people; old meaning 'Westland',
'Germany'), Latvian tauta (a kin, a land, people), tautas
(foreign people; pl.), Old Prussian tauto (people),
Illyrian teut- (people; in the name Teutaplos),
Common Germanic *thiut-, Gothic thiuda
(people, German Volk), Old High German diota (people,
Volk), Old Saxon thioda (people, Volk), Old
English the'od (people), Old Norse thiod (people,
Volk, Leute),
Old Irish tuath (people, folk, a tribe), Welsh tu'd
(people), Cornish tus, Middle Breton tut, Breton
tud (people).
Mind that in Baltic speech the word sometimes means 'foreign land', 'foreigners', or even more exactly 'Germany'. Could it come from outside to Baltic people?
3. *ablu-, *@blu-
This stem denoting an apple could be borrowed from an outside language.
Still, it is not obvious - apples exist in Greece and in Asia as well,
so Indo-Europeans who came to Europe must have known this tree. The non-Indo-European
character of the root is usually stressed by the *-b- in
it, which is quite rare and unnatural in Proto-Indo-European, as the proto-language
was fond of *bh. Again an *a in the root, and
again all European languages keep the word, while Greek, Anatolian, Indo-Iranian
lack it. Here is the cognate list:
Lithuanian obuolas, obuolys (an apple), obelis
(an apple tree), Latvian abuols (an apple), abele
(an apple tree), Old Prussian woble (an apple), wobalne
(an apple tree), Sudovian a'bale' (an apple),
Common Germanic *apalaz, *apliz, Old High German apful,
aphul (an apple), Crimean Gothic apel, Old Saxon
appul, Old Norse epil, Old English aeppel
(an apple),
Old Irish aball (an apple tree), Scottish Gaelic abhall
(an apple tree), Irish ubhall (an apple), Welsh afal
(an apple), Cornish auallen, Breton avallen,
Oscan Abella - a city in Campania which was famous for
its apples (Vergilius wrote: et quos maliferae despectant moenia Abellae;
Aeneis 7, 740),
Common Slavic *abluko, Russian jabloko
(an apple), jablon'a (an apple tree).
This very root is possibly present in Dacian or Thracian in the gloss dinupula < *k'un-abola, but we are not sure about it. In other Indo-European branches, the word for 'apple' is often formed from the stem *ma'l- (Greek me'lon, Latin malus, Hittite mahlas 'vineyard', Slavic *malina 'raspberry').
4. *wa't-
This word was identified and researched quite recently, so not many
critical remarks appeared by now. Its cognates in European languages:
Latin vatis (a seer, a foreseer),
Irish fa'ith (a seer), fa'th (maxim), Welsh
gwawd (a song, poetry),
Old English wo'd (frenzy), wo'th (a poem).
Among other words of possible "European" vocabulary there are *ab-
'water', *loku- 'a lake, flowing water', etc.
However, and though we see that these words are actually present in just a few Indo-European branches, it is still too early to draw conclusions. There can be different explanations for the fact that they exist in European languages only. The first and easiest one is that Anatolian, Greek, Indo-Iranian and Tocharian branches simply lost these stems - so their presence in Europe is just mere coincidence. Some proofs for this can readily be noticed . The latest research shows that the words above might have their cognates in Asian groups. For example, the root *teut- or some root very much like it semantically and phonetically was found in Hittite tuzzi- (an army), which both sounds and means something similar. Concerning the root *ablu-, some researchers refer to the Old Indic word abalah (the plant Crataegus Roxburghil). Opinions will continue to differ until the universal proof for each stem is found.
Another view is that a number of words entered the languages of Indo-European migrants in Europe from the aboriginal nations of these lands, for it is undeniable that before Indo-Europeans came, Europe was populated by numerous and completely different peoples who inhabited both river valleys and mountainous regions from the Balkans to the Highlands of Scotland. The relics of these substratum nations are today found only in Spain, where the Basques are located, but in prehistoric times such people had not been assimilated: Etruscans and Picenes in Italy, Sicelians in Sicily, Iberians and Lusitanes in Spain and Portugal, Picts in Scotland, Pelasgians in Greece, Rhaetians in the Alps, Aquitanes in France and many other different tribes are known from history, though their languages and origins are in most cases yet a mystery to us (see more about substratum in European languages).
Obviously,
this is not a full list of those who lived in Europe before the second
millennium B.C. Indirect sources -- legends, fairy tales, archaeological
sources -- tell us that Germany in the 2nd millennium B.C. was populated
by a people which is traditionally called Folkish or Volkish (from the
Germanic word folk, the origin of which is uncertain).
These people, as some linguists believe, added no little to the Germanic
language, primarily its specific phonetic features, such as the rich system
of unvoiced consonants and the First Germanic Consonant Shift. The
lexicon of these languages, which contains a great many non-Indo-European
words (30% of the Old High German vocabulary contains words which are not
found in any other IE group), is also believed to have been influenced
by the Folkish language.
And so the substratum influence is one of the possible reasons for
these "European" (as against "Asian") words. It is possible that
pre-Indo-European people used one language which was spoken in modern Germany,
France, Italy and the Baltic states, so that Germanic, Celtic, Italic and
Baltic migrants, respectively, could borrow words from it. Because
of their geographic placement, Greeks and Indo-Iranians would not have
had this access allowed them.
And finally the theory called "Proto-European": there was a single
language belonging to those Indo-Europeans who decided to go westward from
the common homeland and reached Europe, later turning into Celtic, Baltic,
Italic, Germanic and maybe Slavic branches. This possible Proto-language
(and its descendants) is sometimes called Nordwestblock; this name
appeared after the works written by Hamp and Huld; recently another Indo-Europeanist,
Beekes, supported it, and Antoine Meillet also suggested a Vocabulaire
du nord-ouest.
A name alone does not matter when we are looking for facts But
the facts are still too scarce for us to be sure which of the opinions
above is right. No characteristic features of morphology or phonetics
distinguish European languages from their Asian relatives -- this cannnot
be so if we adopt the "Proto-European" theory. We can only hope for
future research to resolve our problem.