Index to CAP Articles

1. Aims, Background and References
2. Roots and Their Forms
3. Applying Philological Knowledge to the Study of English
4. Non-Roots and the Study of English
5. Roots with IE *MEN-
6. Roots with IE * ME- , Vowel Length and Vowel Gradation
7. Roots with IE *EN- and NE-
8. Roots with IE *EM-, NEM- and MEM + extensions
9. Reducing the Number of Entries in Pokorny
10. The Roots in Articles 1-8 and Mann's References to Them

CAP Article 1

Aims, Background and References

Introduction

The purpose of the CAP articles is:

1. To produce a simple graded introduction to Indo-European (IE) philology through the study of the morphology of well-established IE roots.

2. To consider the various interpretations of the existing data.

3. To study how this knowledge may be applied to learn other related languages more effectively and efficiently.

4. To gain a better understanding of register in language use.

Background

In 2003, the total number of languages in the world was estimated to be 6,809 (http://www.krysstal.com/langfams.html). This huge number depends ultimately on what we consider is a language and what is a dialect.

Linguists have long attempted to reduce this large number of languages to a smaller number of language families. The late Joseph Greenberg was the most successful. He reduced the number to about twenty-three.
(http://www.ship.edu/~psych/languagefamilies.html).

The largest language family in the world is Indo-European.

A major linguist concerned with this area of linguistics was Julius Pokorny (1887-1970) (http://volny.cz/enelen/pok.htm). Although seriously dated, his two-volume dictionary is still in use today. The details are as follows:

1. Pokorny, Julius (1959), Indogermanisches Etymologisches Wörterbuch, Vol. I, Bern: Francke.

2. Pokorny, Julius (1969), Indogermanisches Etymologisches Wörterbuch, Vol. II (Index compiled by Harry B. Partridge), Bern: Francke.

This material also exists in English in a simplified form in:

Pokorny (1959) distinguishes some 14 different IE subgroups as follows.

For practical purposes these may be divided into 7 major groups (Indic, Iranian, Hellenic, Italic, Celtic, Germanic and Slavonic) and 7 minor groups.

Pokorny (1959) also postulates by implication 2,044 starred or hypothetical forms reconstructed from the extant written reflexes of the languages found in the following fourteen IE language groups.

Two of my own books use this basic material as their subject matter. These references are:

1. Bird, Norman (1982), The Distribution of Indo European Root Morphemes (A Checklist for Philologists), Otto Harrassowitz, Wiesbaden, Germany.

2. Bird, Norman (1990), A First Handbook of the Roots of English, Lapine Education and Language Services, Jersey (C.I.).

An English scholar, Stuart E. Mann, tried to make good the deficiency of no standard IE dictionary in English with the following study:

This dictionary has perhaps never received the attention it deserves and is at present out of print. One study, which tries to take Mann's findings into consideration, is:

Of the three books by Bird mentioned above, Bird (1982) is out of print. Bird (1990) and Bird (1993) are available from the author.

TOP

CAP Article 2

Roots and Their Forms

Introduction

IE philology and modern linguistic studies both began in the eighteenth century. These disciplines have clearly come far in the last two hundred years, but our knowledge of IE languages and their origins is still far from complete. As mentioned in Article 1, a highpoint in IE morphology was Pokorny (1959). This dictionary, however, is now seriously dated, and a new work is now in preparation, namely: The IED project.
http://iiasnt.leidenuniv.nl/ied/intro.html

Roots and Their Forms

Pokorny (1959) provides us with a useful database of IE root forms. We may note, for example, that most IE roots are monosyllabic and that they all contain a vowel (V) and one or two consonants (C) with the following distribution:

1. VC e.g. 456 em- "to take"
2. CV e.g. 1167 mē "big"
3. CVC e.g. 1213 men "to think"

The initial consonant is generally constant, the final one less so. The vowel is liable to a lot of change. Nevertheless, well-established laws such as vowel gradation or ablaut have been devised.

IE "m" and "n" are fairly stable consonants and "e" is the most common basic vowel. Consequently, the root 1213 "men-" serves in this article as a useful illustration.

Pokorny (1959:726-8) cites cognates of 1213 men- "to think" in all 14 categories except 2 (4. Phrygian and Dacian, and 6. Illyrian); Mann (1984) cites only 10 occurrences and does not cite 13. Tocharian, or 14. Hittite. Both Pokorny and Mann cite occurrences of this root in all seven major IE language groups.

The form that a root may take varies in the cognate languages. This can be seen from even a cursory study of English words according to their derivations from Germanic, Italic, Hellenic, and Celtic.

The Latin word for mind "men-s, men-t-is", hence English: mental, mention, com-men-t, and the suffix –ment in equipment, etc. This use of a root as a suffix indicates that the difference between the concepts of root and non-root is often little more than a matter of convenience, although a useful distinction in applied linguistics.

Other Italic derivatives in English illustrate a variety of formal changes as follows:

1. Mon-ey: vowel gradation (e→o change). Likewise, de-mon-strate.
2. Me(n)-men-t-o: a reduplicative form with a reduced first element due to assimilation.
3. Mon-u-men-t combines the two features of reduplication and vowel gradation.

Hellenic gives us the following changes of form:

1. A-m(e)n-esia: vowel reduction.
2. Mne-mon-ic: a reduplicative form with a reduced first element as in 1. Above, and a econd element with vowel gradation.
3. Man-ia: a vowel change.
4. Ma-them-atics; root reduction and vowel change.
5. Muse, mus-ic: a vowel change.

Germanic gives the following two pieces of for consideration:

1. Min-d: the simple root plus a suffix. The e→i change is not due to vowel gradation as above, but to an assimilation (umlaut) that had originally taken place in the second element of the word but has since disappeared.
2. Re-min-d: This word is a hybrid with an Italic prefix re- and a Germanic root.

This example also raises an interesting question incidentally that will be dealt with in later articles, namely root co occurrence or overlap. My studies have shown that the overlap between Germanic and Italic, Germanic and Hellenic, and Italic and Hellenic, ranges between 62% and 65%.

TOP

CAP Article 3

Applying Philological Knowledge to the Study of English

The linguistic insights noted in the above articles are important in the field of pure linguistic research, but, in my opinion, in recent times another area of research has been virtually ignored, namely the application of the philological knowledge that we already possess to efficient language learning.

The size and importance of the problem can best be illustrated with two statistics with regard to one IE language, English:

1. If we assume that there are about one billion speakers of English in the world today, then only about 400 million are native speakers.

2. According to the Oxford English Dictionary, there are 600,000 words in the English language.

It should be clear to anyone who has ever studied foreign languages that the chances of ever mastering this huge vocabulary are extremely limited, unless the problem is approached systematically. The study of applied philology can provide this system.

The Problem and the Solution

The problem for students of English is, in short, coming to terms with a huge vocabulary. The problem is especially great for students who come from non-IE linguistic backgrounds, and the problem is further exacerbated, if the students write their native language in a non-Latin script, e.g. Arabs and Chinese.

The first stage in solving this problem is reducing the number of words that need to be learnt by ignoring the low-frequency words. This solution has long be recognised and accepted in the field of applied linguistics and all the best educational publishers observe it. One of the best general books on this subject is:

Nation, ISP (2001), Learning vocabulary in another language, Cambridge: Cambridge University Press.

This solution is often adequate in the initial stages of vocabulary mastery, but a time comes when it is necessary to learn lower frequency words. It is then that applied philology techniques become important.

Applied Philology Techniques

Students need to be aware that:

1. All words have origins, and in the case of English approximately 95% of the vocabulary derives from Indo-European, in which only just over 2,000 roots have been postulated.
2. All words contain at least one root, and perhaps other roots and non-roots.
3. By non-root is meant affixes, clitics, and inflections.

The Origins of English

English belongs to the West Germanic group of IE languages.

As the first English people (Anglo-Saxons) settled in Celtic territory in the 5th Century AD, they incorporated many Celtic words into their language. With the advent of Christianity and the later conquest by French-speaking Normans (former North Germanic people) numerous Italic (Latin and French) words entered the language. Later, in common with much of Western Europe, many Hellenic (Greek) words entered the language as a result of the spread of scholarship especially during the Renaissance.

As a result, based on my own etymological study of the 7,500 most frequent words in English, the percentage distribution of English words by origin in round figures is as follows:

1. Germanic: 39%
2. Italic 51%
3. Hellenic 6%
4. Celtic 2%
5. Other IE languages 2%
6. Non IE languages 2%

This is a major reason why English is difficult language to learn. English is not in fact a single language, but a triple language with many foreign accretions. It can further be argued that in broad terms spoken English is predominantly Germanic, written English is largely Italic, and technical English is Hellenic.

If this point is not fully appreciated, and if each and every word is treated as being equal in meaning, then appropriate register will never be mastered, and a non-native speaker, however fluent, will always sound like a non-native speaker. In order to illustrate this point, compare the following two sentences which both communicate the same basic meaning.

1. Come quickly! My friend has had a heart attack. (13 syllables, 9 Germanic words)

2. Come immediately. My companion has suffered a cardiac arrest. (20 syllables, 9 words (4 Germanic, 4 Italic, 1 Hellenic))

These problems can be overcome. But only with the help of applied philology.

TOP

CAP Article 4

Non-Roots and the Study of English

Definitions: Roots, Non-roots, and Inflections

The major difference between roots and non-roots is that roots provide lexical or referential meaning, e.g. *men- "to think, mind", whereas non-roots have a grammatical or lexical function. Inflections have a purely grammatical function, e.g. to swim // he swim-s; the –s on "swims" indicates the 3rd person singular present simple indicative active form of the verb "to swim". If we ignore zero as a morphological concept for the time being, we have only eight inflections left in Modern English. As they take on various forms, they are usually designated by means of a symbol set within morphological brackets { }, as follows:

1. {Z1}: plural of countable nouns, e.g. a book – book-s
2. {Z2}: indicator of possession (strictly speaking a clitic), e.g. a boy – a boy's book
3. {Z3}: 3rd person singular present simple indicative active form of the verb, e.g. to swim – he swim-s
4. {D1}: 3rd person singular past simple indicative active form of the verb, e.g. to walk – he walk-ed, to swim – he swam
5. {D2}: past participle, e.g. to walk – walk-ed, to swim – swum
6. {-ing}: present participle, e.g. to walk – walking
7. {-er}: comparative degree of adjective or adverb, e.g. hard – hard-er
8. {-est}: superlative degree of adjective or adverb, e.g. hard – hard-est

Affixes may be placed between these extremes of the continuum of root and inflection, and may derive from either of these two sources. An affix is designated as a prefix, suffix, infix or suprafix according to its position in relation to the root of a word. An example of an infix in English is hard to find. Examples of the other types of affix in English are as follows:

1. *prefix: e.g. un-wise, de-crease
2. *suffix: e.g. care-ful, care-less, care-less-ly, care-less-ness
3. suprafix: e.g. im-pórt (v), ím-port (n), i.e. there is a change in stress.

*Note. More complete lists of prefixes and suffixes in English are given in Appendices 1 and 2

Expanding Vocabulary Knowledge and Mastery with Non-roots

If we take a single root and combine it with the affixes listed in the appendices, we can build a lengthy list of words, which is easier to learn than a similar number selected at random. This is illustrated below:

The English word of Italic origin "trac-t-or" from IE 1885 *tragh is a machine that pulls. The process of pulling is known as "trac-t-ion".

From the base (root + extension) "trac-t" we can form the verbs: to ab-s-tract, at-tract, con-tract, de-tract, dis-tract, ex-tract, pro-tract, re-tract, sub-tract.

From "traction" are formed the nouns, abstraction, attraction, contract, detraction, distraction extraction, protraction, retraction and subtraction.

From the same base are formed the adjectives: attractive, distractive; and the adverbs: attractively, distractively (hybrid forms).

In the Italic forms derived from French after other local changes had taken place, we have in English: train, trail, trace, treat, trait, portrait, betray, trick, tricky, retire, and retirement.

The Germanic cognate form is: thrall, hence enthrall.

According to Pokorny (1959:1089) the apparent Germanic cognate form draw (and hence drawer, withdraw, draft, drag, draught) is due to contamination with another IE root 406 (ibid., 273). This sort of problem will be addressed in later articles.

Aids in Recognizing the Origins of Words in English

There are several reasons why it is useful to be able recognize the origins of a word in English, e.g. to develop one's understanding of register, to understand the spelling, or to use certain particular types of reference materials such as:

Bird, N. (1990), A First Handbook of the Roots of English, Lapine Education and Language Services, Jersey (C.I.).

The following guidelines are not absolute rules, but are usually adequate to deal with most English words as they occur without reference to an etymological dictionary.

1. The functional vocabulary of a text is usually Germanic (c.98% true). By functional vocabulary is meant: pronouns, prepositions, conjunctions, auxiliary verbs, determiners, and some adverbs. Exceptions to this "rule" are "very" and "sure". Functional vocabulary constitutes approximately 50% of any text.

2. Content vocabulary includes nouns, verbs (except auxiliary verbs), adjectives, and most adverbs.

3. Short words (1-2 syllables) are usually Germanic, longer words are Italic, (2-4 syllables), and the longest words are Hellenic (3-10 syllables), e.g. G heart, I core, cordial, H. cardiac (c. 70% true)

4. Strong / irregular verbs are Germanic (98% true).

5. The origin of words can often be recognized from the non-roots as listed in the appendices, e.g. Italic con-vers-at-ion-al, but hybrids do occur, e.g. Italic re- + Germanic mind, in remind.

6. Certain spelling features, e.g. the letter "y" in the root of the word indicates Hellenic origin, as does the joining of roots with the letter "o", e.g. ot-o-rhin-o-laryng-o-log-ist.

If all these guidelines are considered together, the origin of a word can usually be guessed quite rapidly and accurately.

TOP

Appendix 1

English Prefixes and their Origins

TOP

Appendix 2

English Suffixes and their Origins

TOP

CAP Article 5

Roots with IE *MEN-

Introduction

In these articles, we have considered one root in detail, namely 1213 {men-} "to think, mind". Before continuing, it is worthwhile explaining some of the conventions in notation used in this linguistic discipline.

1. The brackets { } are used in morphology to indicate a morpheme, which may be either a root or a non-root.
   
2. In historical linguistics / philology the * indicates that the form is a hypothetical reconstructed form. In IE linguistics all the roots are reconstructions, as IE as such has been wholly reconstructed from extant written forms of cognate languages. Consequently, in books on IE linguistics the asterisk / star can often be omitted.
   
3. The use of upper or lower case letters to indicate IE roots depends on the authors' conventions, e.g. Pokorny and Mann use lower case, Bird upper case.
   
4. The number preceding the root indicates the number out of the 2,044 roots found in Pokorny (1959). These numbers are found only in Bird (1982, 1990, 1993) and are not standard or internationally recognised. I numbered the roots in Pokorny in this way as a matter of convenience. Every headword is included except *abo(n) "monkey" (ibid., p2), since, as Pokorny himself states, this is a Celtic new-formation probably related to 2 {ab-} "water".

Tabulating the Distribution of IE Roots

Bird (1993) tabulates the distribution of all the roots and non-roots in Pokorny (1959) and Mann (1984) in Pokorny's 14 categories (cf. Article 1) as follows:

The meaning of each section in the above table is as follows:

1. No.: the overall serial number of this entry within Pokorny (1959) as explained above. It is worthwhile keeping in mind that the numbers that concern us for the most part are: 5 Hellenic, 8 Italic, 9 Celtic and 10 Germanic. These appear in bold in the table.
   
2. 1-5 MEN: the first starred headform appearing in Pokorny (but possibly changed by Mann).
   
3. The IE groups in which derivatives of the starred forms occur. The number means that both Pokorny and Mann agree; P means the derivative occurs in Pokorny only and M in Mann only. Blank means that no occurrence in a particular language group has been recorded.
   
4. The occurrence of M1179 in 1212 means that Mann believes that 1212 should be subsumed under 1179, making 1212 redundant. Bird (1990) tends to follow these recommendations.

Consequences in English of the Distribution 1-5 MEN

*Notes:

1. The difference in the root vowel between "permanent" and "remain" is typical of the sort of changes that Latin underwent in passing through Vulgar Latin (VL) into the present Romance languages such as French.

2. Although the table frequently shows cognate forms of the root in Celtic, no examples appear in the list above. Although Germanic- and Celtic-speaking communities have lived in close proximity in Britain for over 1,500 years, the Celtic languages only account for about 2% of English vocabulary.

One Final Important General Point

Although the advantages of applied philology have been mentioned several times so far in these articles, one important point has been inadequately emphasised. In addition to helping to study English vocabulary, becoming root conscious, helps study the vocabularies of the languages from which the words derive. Applied philology constitutes in fact a major aid in learning foreign languages in general and IE languages in particular.

TOP

CAP Article 6

Roots with IE * ME- , Vowel Length and Vowel Gradation

Introduction

In article 5, we considered roots with the form *MEN- (1211-1215), i.e. CVC (consonant vowel consonant). IE roots may, however, be in the form CV (consonant vowel), e.g. * ME- . In this article we consider 7 IE roots with this form. In two them the vowel is short *{me-} (Nos. 1162-3) and in the remaining five, the vowel it is long *{mē-}(Nos. 1164-1168). The distribution of these roots across the IE languages is summarised in the following table. Explanations of the abbreviations are described in article 5.

Tabulating the Distribution of IE Roots

Bird (1993) tabulates the distribution of all the roots and non-roots in Pokorny (1959) and Mann (1984) in Pokorny's 14 categories (cf. article 1 ) as follows:

Consequences in English of the Distribution 1-7 ME- (Nos. 1162-1168)

*Notes

1. Although the table frequently shows cognate forms of the root in Celtic, only one example appears in the list above, "claymore": a large sword. As mentioned in article 5 , "although Germanic- and Celtic-speaking communities have lived in close proximity in Britain for over 1,500 years, the Celtic languages only account for about 2% of English vocabulary". The area of activity, from which a loanword comes, in this case "warfare", is often indicative of the state of relations that the speakers of the two languages concerned had for each other.

The IE Vowel System

As may be noted from the above examples, the initial consonant /m-/ is constant but the vowels vary considerably. Although this is so, the vowels change according to quite well established rules.

These rules are explained in their simplest form in the following quotation from the Encyclopedia Britannica:

The vowel system included a, e, o, i, and u, of which the first three, and probably all five, occurred both long and short. A characteristic feature of Proto-Indo-European was the pattern of grammatically conditioned vowel alternation called "ablaut," by which the basic vowel e (normal grade) of a root or suffix could be lost (zero-grade), lengthened to e (lengthened grade), changed to o ( o -grade), or changed to o and lengthened (lengthened o -grade).

Many of these changes are found in root 1213 MEN- "to think". These are discussed in article 2 . Examples of the various variations found in this root are as follows:

Normal grade "e" in Hellenic derived "mentor"; -ment in Italic "fragment"
Lengthened grade "e" in Latin "mēns – mentis" = "mind"
Reduced grade "e" in Hellenic derived "amnesia".
Ablaut "e o" in Italic derived "demonstrate" and "money" ;
          Hellenic derived "mania"

Other examples of this phenomenon can be seen in the words derived from the seven IE forms, 1169-75 ME-, cited above, e.g. G. me, my, middle, mow, more, mood.

The use of vowel gradation for grammatical purposes is of particular interest in the Germanic languages, e.g. signaling of tense: to forget – forgot – forgotten; to meet – met – met. All three ablaut grades, e, o, weak/zero grades, occur in English, although they may not be immediately apparent, e.g. to sing – sang (and "a song") – sung. (cf. http://darkwing.uoregon.edu/~l150web/weblec2.2.html ).

TOP

CAP Article 7

Roots with IE *EN- and NE-

Introduction

In article 5, we considered roots with the form *MEN- (1211-1215), i.e. CVC (consonant vowel consonant), and in article 6 IE roots in the form CV (consonant vowel), e.g. * ME- . In this article we consider 5 IE roots with the basic forms *en- and *ne-. In two of the *en forms the vowel is short *en- (Nos. 458-39 and 459-40) and in the remaining one the vowel is long *ēn- (No. 460-41). In the case of the form *ne-, the vowel is short in both cases. The distribution of these roots across the IE languages is summarised in the following table. Explanations of the abbreviations are described in article 5.

Tabulating the Distribution of IE Roots

Bird (1993) tabulates the distribution of all the roots and non-roots in Pokorny (1959) and Mann (1984) in Pokorny's 14 categories (cf. article 1) as follows:

Consequences in English of the Distribution
1-3 EN- (Nos. 458-460) and 1-2 NE- (1283-1284)

*Notes

1. In the Germanic examples in the above table, it will be noted that the vowel is frequently dropped, e.g. n- in be-n-eath (No. 458), n- in n-either (1283).

2. In the English loanwords from Italic, the "n" in both in- (1) and in- (2) is frequently subject to partial contiguous regressive assimilation, e.g. in 458 in addition to in- in in-cline, we also have illuminate, immigrate and irrigate. In those words deriving from French, vowel change also occurs, e.g. enable. This frequently coincides with the same assimilation as described in the case of Latin, e.g. employ and annoy. In 1283 we find in-visible but also illegible, impatient and irregular. From French we find en-emy but also nice and nonsense.

3. A similar situation occurs with regard to words of Hellenic origin: 458 in addition to en-erg-y, we have emphasis and even -s- in epi-s-ode. In 1283 in addition to the full form in an-o-nym-ous and an-o-mal-ous, we also have the generally reduced form as in a-tom and a-path-y, i.e. complete contiguous regressive assimilation .

4. Examples of complete contiguous regressive assimilation also occur in words derived from Latin, e.g. i-gnor-ant. The reason for the disappearance of the /n/ of {in-} is presumably the reduction of the consonantal cluster */ngn/.

5. The examples of regressive assimilations described in 2-4 above, all took place in the Italic and Hellenic languages before becoming loanwords in English. Sometimes, however, the assimilations took a different form. For example, gemination, i.e. pronouncing two identical consonants occurring together twice, lingering on the consonant, as it were, occurs in Italian on individual words, e.g. avventura, but not in English, e.g. adventure, which contains a very strong stress element.

TOP

CAP Article 8

Roots with IE *EM-, MEM- and NEM and the Problem of Extensions

Introduction

In this article we consider 6 IE roots ending in *em-, although in three cases the final consonant in the putative root is followed by an extension. In only one root is the vowel long *mēmso- (No. 1210-75). The distribution of these roots across the IE languages is summarised in the following table. Explanations of the abbreviations are described in article 5

Tabulating the Distribution of IE Roots

Bird (1993) tabulates the distribution of all the roots and non-roots in Pokorny (1959) and Mann (1984) in Pokorny's 14 categories (cf. article 1 ) as follows:

Consequences in English of the Distribution
Of the Roots Nos. 456-457, 1209-1210, and 1299-1300

TOP

CAP Article 9

Reducing the Number of Entries in Pokorny

There are 2044 entries in Pokorny's etymological dictionary. My own very small contribution to the field of comparative philology or diachronic linguistics has been twofold:

1. To number these entries from 1 – 2044, then using this numbering system in my own work.
2. To attempt to reduce the overall number of entries.

The reasons for doing this are similar to those of the morphologist working in the field of applied synchronic linguistics. There, in the case of English language teaching (ELT), a major problem is to reduce the vocabulary of some 600,000 headwords (lexemes) in the Oxford English Dictionary (OED) to manageable proportions for both testing and teaching purposes. The number has been reduced to a base of 25,000 for testing purposes by such researchers as Goulden et al. described in Nation (2001:363-367).

In my opinion, the work of these two types of linguist overlaps, and what the philologist has achieved can be of real value to the applied morphologist.

As a first stage in my endeavour, I consider Watkins' adaptation of Pokorny's work as presented in Watkins, C. (Ed), (1985) The American Heritage Dictionary of Indo-European Roots , Boston : Houghton Mifflin. A new edition of this work is now available and the useful general introduction to the field is available on the internet at http://www.bartl e by.com/61/8.html .

The reasons for choosing Watkins as my starting point is that Watkins's work (1985) follows so shortly after the publication of Pokorny's Index (1969) and Watkins himself is one of the most conservative of Pokorny's commentators, sometimes adding little more than what Pokorny suggests himself. The changes that Watkins suggests as recorded in Bird (1993), The Roots and Non-Roots of Indo-European: A Lexicostatistical Survey , Wiesbaden : Harrassowitz (passim) are listed below.

1. Non-IE elements in Pokorny

1.   413     DHRIGH    hair
2.   1614   (S)K'EL     to jump

*Mann also includes the following but Mann's analysis appears in later articles

3.   549     GEN EBH    plug, pin
4.   951     KOIS         to worry?

2. Watkins' Suggested Entries to be in Subsumed Pokorny

If we use Watkins findings as displayed in the second table above, we can reduce the overall number of Pokorny's roots from 2044 to 1993; if we include the findings in the first table also, we can reduce the overall number to 1991.

3. Recurring Features in the Subsumed Forms

A study of the forms that are to be subsumed under existing forms also reveals features that can help the philologist reduce the overall number of hypothetical forms still further. A selection of these features appears below.

3.1 Similar Form and/or Meaning

19. 1176    2. MEI    to change
     1177    3. MEI    to wander, go
     1184    MEI-GW  to change

27. 1423    3. PEU    to cut, strike
     1422    2. PEU    to explore

3.2  Metathesis (exchange of places between the sounds/letters of a word)

29. 1556    2 SEK    to cut      1607     SKEI    to cut

Other examples of this phenomenon often with additional features are: 1, 2, 3, 6, 11, 12, 29

3.3. Extension

21. 1213    3. MEN    to think             1218    MENDH  to consider
22. 1214    4. MEN    small, isolated    1222    MENI      kind of fish

3.4 Reduction

25. 1392    PELE    flat, to spread    1435    1. PLA-K    broad, flat
                                                 1437    PLAT         broad, flat
31. 1584    1.SER   to flow              1734   SR-EDH      to spout

This feature may appear in a somewhat confused form as in the above examples where both extension and reduction have taken place.

Although only a selection of possible changes is presented here, it is to be hoped that the important implications of the findings are obvious for the research, that will be reported in future articles.

TOP

CAP Article 10

The Roots in Articles 1-8 and Mann's References to Them

Pokorny's dictionary has served as the standard dictionary in the field of Indo-European studies for all those that can read German since 1959 and especially since 1969, when the index was published. The American Heritage Dictionary of Indo-European Roots , revised and edited by Calvert Watkins, non-German speakers have had the benefit of being able to gain at least some insight, however limited, into the breadth of the field of Indo-European comparative linguistics since 1985.

Mann, S.E. (1985-7), An Indo-European comparative dictionary is the only real attempt since 1985 to produce this kind of dictionary in English. Unfortunately, Mann died while the dictionary was being published in fascicles, viz. the back cover of the 9 th of 11 fascicles. Moreover, no index ever appeared. As a result, Mann's great work appears to have remained virtually ignored since his death. For example, my web search for reviews of his work drew a blank and I noted that it does not even appear in the bibliography of etymological dictionaries in Beekes, R.S.P. (1995), Comparative Indo-European linguistics: An introduction, Amsterdam/Philadelphia, John Benjamins (page 291). Mann remains instead better known for his work for his work on Czech, Albanian, and Armenian.

This, I believe, is a great pity. The following tables, which will appear on a regular basis on my web site, constitute my own small attempt to make good this deficiency. I have written up notes on the whole of Mann's dictionary but my attempts to interest Mann's publisher or the relevant academic community at large in the last ten years have been to no avail.

TOP