126
convenience, I have put the usual names of functional types below the end-
points of the classification system.4"
strong
tense
non-past
fmitude
I------------------
non-fin-past
mood
I----------------------1
fin-past-non-subj
fin-past-subj
(‘present stems’)
(‘participle
(‘primary past
(‘secondary past
stems’)
stems’)
stems’)
Figure 1. Functional types o f fonns of stems of strong verbs
As usual, the marked terms of the classifications are put on the right-hand side.
The classification system establishes an order of markedness among the
classes it provides — from the unmarked (primary) present tense forms to the
most highly marked secondary past tense forms.
3.2 Syncretisms
Among the three ‘primary stems’ of a strong verb, two may be identical in
expression form. Three types of strong verbs may be distinguished in terms of
the identity or non-identity of the vowels of their ‘principal parts’ (in the order
of diminishing number of pertinent verbs):40 41
fin-past
40 In a more explicit account categories would have to be relativised to idiolect systems;
category names such as ‘strong’ would have to be replaced by ‘strong!-,.S')’; see Lieb
(1983). The presentation in the present paper is simplified in a number of ways.
41 This classificatory criterion is found already in grammarians of the 17th and 18th century,
in particular in Adeiung (see Adelung 1782: 803). More recently, it has been used by Grebe
127
Type A. ‘Participle stem’ and ‘primary past stem’ are identical.
Type B. Three distinct ‘primary stems’.
Type C. ‘Primary present stem’ and ‘participle stem’ are identical.
Very often this classification is regarded as basic; it is less often analysed or
discussed in any detail. However, a closer investigation of the foundation of
syncretism patterns turns out to be revealing. The number of patterns of syn-
cretism is multiplied if syncretisms of primary and secondary forms are also
taken into account; still, the rationale of syncretisms is simple. Any marked
stem form (i. e., any stem form apart from the base) may be ‘missing’; if so,
the next less specific (i.e., the next less marked) form takes over its office. As
regards secondary forms this is obvious from traditional treatments and even
from the terminology. Secondary stem forms may be missing, in particular
since the vocalism of the relevant primary forms may exclude the possibility of
forming them, and it is understood that the primary forms will substitute for
them: if a verb stem has no secondary present form the primary present form
applies throughout the present tense; likewise, if a verb stem has no secondary
past form the primary past form applies throughout the past tense.
It turns out that the same ‘logic’ applies in case of syncretism between
primary stem forms. Consider the stems
FECHTL, SPRECHL
and
MESSL:
Table 4
(next page) lists their primary forms. These are ordered from left to right
according to increasing markedness (as established before, cf. Figure 1). Note
that the ‘participle stems’ are put into the middle for this reason (different from
the usual listing of principal parts).42 The three rows that are preceded by the
lexemes’ names show the distribution of forms among functional types. At the
bottom the expression types are indicated (/e/-form, /o/-form, or /a:/-form).
Table 4 also points out the functional values of the alternations that are in-
volved; this will be explained immediately. What the table says may be de-
tailed as follows.
et al. (1959), Jung (1966), Ross (1967), and Wurzel (1970), among others. Cf. also Ul-
vestad (1956). And, of course, the syncretisms at issue have not been neglected in historical
grammars. Wilmanns (1906) for one uses formal differentiation as his basic classificatory
criterion opposing “Verba mit voll entwickeltem Ablaut” (25) and “Verba mit schwächer
entwickeltem oder fehlendem Ablaut” (32).
42 But Adelung (1782: 803) used this order in presenting his classification of strong verbs
(and in particular in introducing mnemonic names for the classes of strong verbs), certainly
because (non-)identity of the vowels of principal parts is his superordinate criterion. It is
also adopted (and again in order to take care of syncretisms) in Johnston (1997).
128
functional type:
‘primary
‘participle
‘primary
present stem’
stem'
past stem’
A. FECHT^
B.
SPRECH
1
C.
me ss l
expression type:
functional value:
fecht
focht
sprech
sproch
sprach
maß
/e/-form
/o/-form
/a:/-form
—
past
finite past
Table 4. Syncretisms among primary stem forms
The three verbal stems
FECHTL, SPRECHL
and
MESSL
each have a base in Id
(thus an I-base). In addition,
FECHTL
has an a-ablaut form (i.e., an ablaut form
formed by o-ablaut, vide supra, Section 2.6). The o-form appears in the past
participle and in the finite past (the shaded box indicates the range of applica-
tion). o-ablaut thus functions as a general marker of past tense; put differently,
the functional value of o-ablaut is past.
MESSL
has (in addition to its base) an
o-ablaut form (i.e., an ablaut form formed by a-ablaut, vide supra, Section
2.6). The o-form appears only in the past indicative; again the box indicates
the range of application. The functional value of the a-ablaut form is finite
past, that is, o-ablaut serves to mark finite past forms only.
SPRECH
l
has (in addition to its base) an o-ablaut form and an a-ablaut
form. The a-form appears in the past indicative; again, a-ablaut marks finite
past. The o-form appears in the remaining past tense verb forms, viz. in the
past participle; again, o-ablaut marks past tense. In this case the o-form is
restricted to the participle since
SPRECHL
has an a-form that functions as a
finite past form. The existence of this more specific form excludes the appear-
ance of the o-form of
SPRECHL
in finite past forms of the verb
SPRECHENW.43
The bases are found in the domains that are not covered by ablaut forms.
Secondary forms further limit the range of application of primary forms; e.g.,
43 See Kiparsky (1973) on the ‘Elsewhere-principle’ or ‘Paninian principle’; for discussion
of its role in morphology see, among others, Anderson (1992) and Halle & Marantz (1993).