Re: confession: roots

At 5:28 pm -0500 4/5/01, Eric Christopherson wrote:
[snip]
>
>I always labor under the assumption that at one time, such "roots" were
>actually words to themselves, perhaps in an isolating language, and that over
>time they merged with words around them. I know this is probably
>linguistically naive,
It is.
----------------------------------------------------------------

At 4:55 pm -0600 4/5/01, Tom Tadfor Little wrote:
[snip]
>
>Someone may correct me if I'm wrong, but my understanding is that roots are
>something of an abstraction created by linguists to help analyze word
>forms.
That's my understanding also.

David Crystal (A Dictionary of Linguistics and Phonetics) writes:
"A root is the BASE FORM of a WORD which cannot be further analysed without
total loss of identity.  Putting it another way, it is that part of the
word left when all AFFIXES are removed."

I.e. that part of the word when all affixes have been abstracted.  Affixes
include not only prefixes & suffixes but also infixes; since in the Semitic
langs vowels act as infixes, the root is the (usually triliteral)
consonants, e.g. KTB (write).  The fact that /ktb/ is unpronounceable is,
as Nik points out, irrelevant.  The root (or 'radical' as some prefer to
call them) is an abstraction.

R.L. Trask (A Dictionary of Grammatical Terms in Linguistics) writes:
"In morphology, the simplest possible form of a lexical morpheme, upon
which all other bound and free forms involving that morpheme are based.
For example, the Latin verb meaning 'love' has the root _am-_, from which
are formed the various stems, such as the present _ama-_ and perfect
_amav-_, which in turn serve as bases for the construction of inflected
forms such as _amat_ 'he loves' and _amavi_ 'I have loved'.  Similarly, the
Arabic verb meaning 'write' has the triconsonantal rrot _ktb_ from which
all other forms are derived by various layers of affixation.

Both Crystal and Trask give basical the same meanings to stem & verb when
used as morphological tems.

Trask's use of _ama-_ and _amav-_ is IMO a little unfortunate in that
_amav-_ could be (and diachronically probably was) derived from the stem
_ama-_ itself, rather than the root _am-_, which is also the base of the
noun _amor_ 'love' and adjective _amicus_ 'friendly' (also commonly used
substantivally to mean 'friend').

So what is a _stem_ and a _base_?  A stem is root plus formative affix from
which other forms are derived; a base may be either the root or a stem from
which other forms are derived.

Let me use the Latin verb as an example. Well over 90% of Latin verb forms
can be derived by regular rules once the three basic stems are know (these
are not always formed regularly from one another), e.g.

The verb 'break' is derived from the root *rup

All the forms of the Latin verb are regularly derived from one of three
stems formed with the root *rup as its base:
1. rump- - the 'present stem' - from from the root *rup with the infix -n-
here assimilated before /p/;
2. ru:p- the 'perfect stem' - from the root *rup with lengthening of the
root vowel;
3. rupt- the supine - from root *rup with formative suffix -t.

Verbs whose present stem end in a consonant or -i, use this stem also for
the future:
PRESENT       FUTURE
rumpo:        rumpam     I.......
rumpis        rumpe:s    you (s).......
rumpit        rumpet     he/she/it.......

Verbs, like _ama-_, however, form a separate future stem with the present
stem as its base, e.g. ama:-b-, thus:
ama:bo:
ama:bis
ama:bit

The present stem forms the base in all verbs for the imperfect stem, thus:
rump-e:ba:-, ama:-ba:-
rumpe:bam          ama:bam
rumpe:ba:s         ama:ba:s
rumpe:bat          ama:bat

Similarly the perfect tense (active only) was formed by adding subject
suffixes directly to the perfect stem, i.e. had the perfect stem as its
base.

But the perfect stem itself formed other stems to serve as bases for other
tenses, e.g.
pluperfect (past perfect): ru:pera:-, ama:u-era:-
future perfect: ru:p-er-, ama:u-er-

The supine stem was the base of the supine itself, i.e. ruptum 'to break' -
indicating purpose, used only in association with verbs of motion, e.g.
hanc fenestram ruptum eo = I'm going to beak this window.

The supine stem also served as the base for the perfect passive participle
by adding 1st (fem.) & 2nd (masc. & neuter) declension case endings, thus:
ruptus, rupta, ruptum etc. 'broken'

It also served as the base for forming the future active participle stem
_rupt-u:r-_ to which were suffixed the 1st & 2nd declension case endings,
thus: ruptu:rus, ruptu:ra, ruptu:rum.

This is by no means exhaustive (I've not mentioned subjunctives, for example :)
But I hope helps with the notions of root, stem and base.

Ray.



=========================================
A mind which thinks at its own expense
will always interfere with language.
                   [J.G. Hamann 1760]
=========================================

Replies