Comparative philology is a term for the study of the historical development of the languages of the Indo-European language family. This technique was evolved by the group of linguists working in the second half of the 19th century, mostly in Germany. Going on a German term — Junggrammatiker — used first in disrespect for young linguists, the English translation Neogrammarians was devised. What is here called the Neogrammarian view refers to the methods used for linguistic reconstruction in the 19th century. It first arose towards the end of the 18th century with the discovery that classical languages like Greek, Latin and Sanskrit are related to each other. Soon after this a number of scholars independently established interconnections between the languages of the Germanic, Romance, Celtic, Slavic and Baltic groups of languages. Notable among the authors of this earlier group is Jakob Grimm who established a series of sound laws which applied to Germanic in its earlier stages.
Sound law This is a term which refers to a change or a series of related changes in the phonology of a language. The term law (German Lautgesetz) is used to stress the regularity of the change. The classic example of a sound law is the so-called Germanic sound shift or Grimm’s law. This states that in the early stage when Germanic was differentiating itself from the remaining dialects of Indo-Germanic, all voiceless stops were shifted to voiceless fricatives, i.e. /p/ became /f/, /t/ became /θ/ and /k/ became /x/ (among other shifts).
| Latin | English | Old English |
| pes | foot (p ~ f) | fot |
| tres | three (t ~ þ) | þreo |
| collis | hill (k ~ x) | hyl |
| quod | what (kw ~ xw) | hwæt |
With hill and what one must consider the development of the /x/-sound in the history of English. Within the Old English period /x/ was weakened to /h/ in word position and represented as h in writing. From this fact one can conclude that those words which have been inherited from Germanic and which are written with initial h originally had /x-/. The sequence /xw/ in Germanic later developed analogically to /hw/. Through assimilation of the /w/ to the preceding voiceless glottal fricative the voiceless [ʍ] which was voiced in the early modern period in southern English but which is retained in many conservative dialects.
Verner’s law The sound law was the central theoretical concept in comparative philology so that exceptions to laws represented a considerable difficulty to linguistic interpretation within this model. The results of a sound law could be veiled by later analogy which did away with irregularity in morphological paradigms but still, for Germanic, there were disturbing exceptions to the operation of established sound laws which could not be explained by an appeal to analogy.
Scholars had noticed at an early stage that the Germanic consonant shift did not affect all instances where it might have applied. Grimm called these exceptional cases grammatischer Wechsel ('morphological alternation') because they were to be found in verb paradigms, for example).
In 1877 the Danish scholar Karl Verner published an article entitled "Eine Ausnahme der germanischen Lautverschiebung" ('An exception to the Germanic sound shift'), in which he demonstrated that the apparent irregularity here constituted in fact a clear rule. Since then this phenomenon bears his name. In essence Verner’s Law runs as follows: If a voiceless consonant occurred in a syllable which was preceded by another which did not carry the main accent then this consonant was voiced unless the phonetic environment blocked this. Now because accent was relatively free in Indo-European (as still to be seen in Russian) there were many words which were affected by Verner’s Law, for instance in the preterite of strong verbs (but not in the first and third person singular) the accent was on the second syllable. Later with the rise of initial accent in Germanic the environment for Verner’s Law was no longer available but there were reflexes of the earlier situation.
| Indo-European | Old English |
| *pətér | fæder ‘father’ |
With the original word for ‘father’ the accent occurred after the /t/. This was changed as part of the Germanic sound shift to /θ/ and underwent Verner’s Law because the accent was not on the preceding syllable, yielding /ð/ which was later fortified to a plosive.
Another major sound shift in Germanic is the High German sound shift which explains many differences between English and German. The changes can be shown in a table with sample words as follows.
| word initially | word medially | ||
| /p/ | /pf/ | /p/ | /f/ |
| /t/ | /ts/ | /t/ | /s/ |
| /k/ | /kx/ | /k/ | x/ |
| English | German | English | German |
| path | Pfad | open | offen |
| tooth | Zahn | eat | essen |
| church | Kirche | make | machen |
‘Exceptionless’ sound laws In the course of the 19th century examples for sound laws were described for many languages of the Indo-European family. Linguists stressed the exceptionless character of sound laws (German Ausnahmslosigkeit). According to this view the only exceptions to a valid sound law can occur due to the force of analogy. This is where an expected change is not found because another element in a grammatical paradigm has caused a change to be blocked or has masked a change at some later point Thus the change of /s/ to /r/ in the past tense of the verb sein in German has been masked by the analogical spread of /r/ to all elements of the verbal paradigm, e.g. ich war, du warst, etc. In English we can see that an alternation originally existed: I was, you were, etc.
Reconstructing the proto-language One of the major concerns of comparative philology was with the reconstruction of the proto Indo-European language. The technique for reconstruction is as follows: going on a series of attested forms one postulates a common original form which is phonologically most likely. As an example take the postulation of Germanic /ɑ:/. In German one has words like Geist, Heim and in English the corresponding forms ghost, home. Assuming (correctly) that both languages are near relatives, the most likely common form of the vowel in these words is /ɑ:/. This is then called a protosegment and written with a preceding asterisk, i.e. *a (note that this sound does not occur in any present-day language). The sum of all the protosegments for this common stage of English, German is called West Germanic; a stage further back would be Germanic which would include Gothic (the only attested member of the East Germanic branch) and Old Norse (the Northern Germanic branch) before these were separated out into recognisably different languages.
The method for language reconstruction outlined here is termed the ‘comparative philology’ method and is associated with the school of linguists who originated in Leipzig in the last quarter of the 19th century and who are called ‘Neogrammarians’ (German ‘Junggrammatiker'). The most notable representatives of this direction in linguistics are Karl Brugmann and Berthold Delbrück; the main theorist of this school is Hermann Paul (see the relevant section in the history of linguistics section).
Conditioned and unconditioned sound changes A conditioned sound change is one which is caused by some segment in the environment of another. A clear instance of this is i-umlaut which is caused in a given syllable by the high vowel or /j/ in a following syllable. It is important to note that this change leads to morphological irregularity as in the following cases in German.
| jung | jünger |
| gut | Güte |
| Hohn | höhnisch |
In English old : older ~ elder shows the semantic exploitation of the analogically regularised form and the original umlaut form. The latter is found when referring to siblings and in one or two set phrases: my elder sister, an elder statesman; otherwise older is used.
Unconditioned sound change affects every possible segment which matches its input, i.e. it is not dependent on — conditioned by — environment. An example of this would be the diphthongisation of Middle English /i:/ and /u:/, which does not cause any grammatical irregularity; the loss of /x/ in Middle English is another instance of unconditioned sound change.
