[b-greek] GREEK RULES AND LINGUISTIC PRINCIPLES

From: B. Ward Powers (bwpowers@optusnet.com.au)
Date: Sun Aug 12 2001 - 03:23:53 EDT


<x-flowed>
B-greekers all,

GREEK RULES AND LINGUISTIC PRINCIPLES

There has been quite a bit of discussion on b-greek over the years
concerning various rules concerning the language - the last such discussion
just a few months ago. It is my judgement that we have much to gain by
giving more attention than we have hitherto (to state a generalization) to
the application of linguistic principles to the analysis of Greek, and to
deducing rules (especially grammatical and morphological rules) on the
basis of these principles. I believe this will lead to a better
understanding of how the koine Greek language works.

How can we do this, and what will be the result? This is an area which has
been of interest to me for many a long year. The following builds upon and
clarifies some comments I have made before on the b-greek list. In this
post, I state a few linguistic principles, and deduce a rule or two.

If this interests you, keep reading. If not, please move on to the next
post on our b-greek list.

A linguistic principle is a generalized statement about (inter alia) the
behaviour of word forms in a language. A rule exists in a given language,
and is the application of a linguistic principle in a particular situation
in that language. All languages contain both rules which are invariable
(i.e. of universal applicability in that language: e.g., The accusative
form of a neuter word is identical with its nominative), and rules which
have exceptions to them (i.e., irregular word forms). A word form is
regular if that word form is predictable on the basis of such rules; a word
form is irregular if that word form is not predictable on the basis of such
rules.

LINGUISTIC PRINCIPLES AND GREEK RULES

Linguistic principles of a language are patterns of language behaviour
which are noted by observation. That is, they are deduced from what is
observed to happen in the language. Some of them are particularly valid for
koine Greek.

Where a pattern of phonemes (sounds, in speaking) or symbols (letters, in
writing) has a similar meaning in numbers of word forms, this is likely to
indicate that that pattern is an indicator of that meaning. Such a pattern
of sounds or symbols is known as a "morph". A morph is named on the basis
of what it indicates.

A morph may simultaneously indicate two or more linguistic features in a
language. Such a morph is a multimorph. [Example: the suffix endings of
those Greek words which take declension will simultaneously indicate the
number and the case of a word form, and are therefore "numbercase endings";
they may also indicate gender.] Where two or more morphs exist which have
identical meaning, these are known as allomorphs. All the allomorphs with
identical meaning will together comprise the morpheme for that meaning.
[Example: English has more than half a dozen different ways of making a
word plural (cat/cats; ass/asses; ox/oxen; man/men; woman/women;
child/children; mouse/mice; and lots more); all these together comprise the
"English noun plural morpheme".]

Morphs are affected by the phonemes in the morphs to which they are
adjacent; this is a major reason for the existence of allomorphs in a
language. [Example: in Greek, a dental phoneme will drop out of a word form
when it is followed by another consonant.] An explanation of this
occurrence can be formulated as a linguistic rule for that language.
[Example: When a morph beginning with a consonant is added to a morph which
ends in a dental, that dental phoneme drops out of that word form. This is
termed the "Dental Drop-out Rule"; it is a rule of universal applicability:
that is, there are no exceptions. Thus the future active of SPEUDW is
SPEUSW, aorist ESPEUSA.]

Rules are explanations of observed phenomena, deduced on the basis of such
observation. That is, a rule is a statement of what can be expected in a
particular word form. When a word form conforms to the rules, the meaning
of its form can thus be deduced from these rules.

SOME LINGUISTIC ANALYSIS OF LUW

If we apply these principles of linguistic analysis to the forms of the
First Aorist of LUW, we will come to linguistically sound conclusions which
differ to some extent from those currently put forward in numbers of textbooks.

In what follows, I summarize steps of linguistic analysis which in a
linguistics treatment would be set out in more details. I use here the
traditional terms for tenses and aspects.

If we compare similar forms in different tenses, we can identify what
"alterations" make a difference between the tenses, and correlate this with
what that difference is. This also will introduce us to the concept of
morph slots (places within the sequence of a word form where morphs can
occur). For illustration purposes we will examine the second plural active
forms. If we use the term "lexal" for the form of the word's root (the
"meaning bit"), the LU in these forms, and look at the forms LUETE, ELUETE,
LUSETE, ELUSATE, LELUKATE, and ELELUKEITE, we see:
(a) The ending -TE is the part of the form which indicates "second plural
active"; it is the "pronoun ending", "you".
(b) E- prefixed to the lexal indicates "past time".
(c) The durative (Present and Imperfect tenses) and the future forms have
an -E- between the lexal and the ending which has a "neutral" effect on the
meaning of these forms. (Lengthening this neutral morph will indicate the
subjunctive.)
(d) -S- added between the lexal and the neutral morph indicates future tense.
(e) Replacing the neutral morph of the Imperfect (ELUETE) with -SA- creates
the aorist form (ELUSATE); i.e. changes the aspect from durative to punctiliar.
(f) Replacing the neutral morph with -KA-, plus reduplication, creates the
present perfect form (LELUKATE); i.e. changes the aspect to perfective.
(g) Prefixing the past time morph E- at the beginning (though in use this
was sometimes omitted) and the suffix -EI- before the pronoun ending and
removing the -A- of the -KA- morph, gives (E)LELUKEITE, the pluperfect.

THE RULES FOR THE FIRST AORIST MORPH

The above linguistic analysis produces several significant results; the one
on which I wish to focus is the -SA-, which is the punctiliar morph of the
First Aorist. It is a "defining morph": a verb which does not take this
-SA- in the aorist is not a First Aorist verb.

So this allows us to formulate the first rule:
1. To form the aorist tense flexions of a verb, add -SA- as the punctiliar
morph to the verb's lexal.

A verb which does this is a First Aorist verb (that is, First Conjugation),
and regular, and (at least in this matter) predictable.

But this is not the whole story. In some aorist forms of the LUW paradigm
the alpha of -SA- is missing. The first of these is the third person
singular active form, ELUSE(N), with movable nu. This indicates the need
for further explanation, i.e., a second (or supplementary) rule. When we
compare the third singular active forms of imperfect, aorist, and perfect
[ELUE(N), ELUSE(N), LELUKE(N)], we find they have in common -E(N),
suggesting that in these forms this is the third person singular pronoun
ending, with the meaning he/she/it. And in each case the vowel preceding
this ending has elided (respectively -E-, -A-, and -A-). Recognizing this
allows us to formulate a second rule:
2. The -A- of the punctiliar morph -SA- elides before a following vowel.

We test this rule in all the other forms of all the other flexions of the
LUW paradigm, and we find it holds up without exception. In the subjunctive
active and middle the punctiliar morph is always followed by the lengthened
neutral morph, and so elides its -A- in all forms. In the second singular
imperative active the -A- elides in front of the -ON ending, and before the
-AI of the second singular imperative middle. So also in front of the -AI
ending of the infinitive, and in the optative.

In one form - the second singular aorist middle - an original ELUSASO form
(compare the perfect) has lost its second sigma and the -A- and -O have
contracted into -W.

So all forms in the active and middle of LUW are explicable on the basis of
these two rules. In terms therefore of their explanatory effectiveness they
are what are called "powerful rules".

LIQUIDS

Where a verb's lexal ends with a liquid, it will be observed that in the
aorist there is no sigma alpha following it, only alpha. This is because of
the incompatibility in Greek of a liquid with a following sigma. (This
conjunction is found only in a few foreign words such as KHNSOS, census,
tax.) This observation allows us to formulate our next rule, which brings
liquid verbs under the umbrella of First Conjugation regular (i.e.,
predictable) verbs:
3. When the punctiliar morph -SA- is suffixed to a verb with a liquid
lexal, the -S- slides off the liquid and is lost, leaving only the
remaining -A- as the punctiliar morph.

Thus -SA- and -A- are allomorphs of the same punctiliar morph.

Two other features will be observed:
(a) When the vowel preceding the liquid is an epsilon or omicron, it takes
compensatory lengthening for the loss of a stem letter (the sigma), -E- to
-EI- and -O- to -OU-. Thus root AGGEL- to aorist HGGEILA, root EGER- to
aorist HGEIRA.
(b) In the third singular aorist indicative of a liquid verb, both the
sigma and alpha will have disappeared: the sigma has slid off the liquid,
and the alpha has elided before the -E(N) ending. The form will be
recognized for what it is by the factors which caused this situation: the
liquid lexal, and the -E(N) pronoun ending.

SO WHAT?

1. I think it's real neat to see the results which are produced by such
linguistic analysis, and I affirm that they assist us in our understznding
of the functioning of the language.

2. Much of what has been thus deduced and formulated into the above rules
applies even more widely in the language. Thus the way in which the
punctiliar morph -SA- elides its vowel before a following vowel is true
also of the other two morphs which can occur in other verb forms in the
same "morph slot" of a verb: the neutral morph and the perfective morph.
Thus the third singular imperfect active form ELUE(N), the third singular
perfect active form LELUKE(N), LELUKWS and all perfect participles,
(E)LELUKEIN and all the pluperfect.

3. Because the punctiliar morph of non-liquid verbs is always -SA- unless
the alpha elides in front of another vowel, it means that a punctiliar
morph -S- can NEVER occur followed by another consonant.

4. This explanation distinguishes the -SA- or -S- of the aorist from the
-S- of the future, which is ALWAYS followed by the neutral morph (-O- or
-E-). Three future and aorist forms overlap (LUSW, LUSON, LUSHi); that is all.

5. When followed through further, this approach enables us to detect
numerous patterns in the morphology of koine Greek. And this in its turn
would simplify teaching, learning, and remembering Greek morphology.

6. We should note those paradigms which are outside the scope of this
analysis. The aorist passive does not take the -SA- morph (and the perfect
middle/passive does not take the -KA- morph). And, by definition, it only
applies to verbs with a First Aorist.

7. If the above linguistic analysis were to be taken into account,
approaches to teaching Greek would be modified: first, more emphasis would
be placed upon seeing related morphological patterns and principles, and
less on independent rote learning of separate paradigms; second, various
charts which show morphs in Greek verbs would need a little modification
(especially those which do not recognize the elision of the -A- of the -SA-
in front of a following vowel).

For the information of those for whom this may be of interest.

Regards,

Ward

                                http://www.netspace.net.au/~bwpowers
Rev Dr B. Ward Powers Phone (International): 61-2-8714-7255
259A Trafalgar Street Phone (Australia): (02) 8714-7255
PETERSHAM NSW 2049 email: bwpowers@optusnet.com.au
AUSTRALIA. Director, Tyndale College


---
B-Greek home page: http://metalab.unc.edu/bgreek
You are currently subscribed to b-greek as: [jwrobie@mindspring.com]
To unsubscribe, forward this message to leave-b-greek-327Q@franklin.oit.unc.edu
To subscribe, send a message to subscribe-b-greek@franklin.oit.unc.edu


</x-flowed>



This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:37:03 EDT