140
forms are still in use). The vowel occurs in a context that favours rounded
vowels (Paul 1989: 78); presumably, such a context may encourage
o-ablaut and may discourage a-ablaut (in particular when h i- and /a/-
forms compete). Similar remarks may apply to most of the other stems
that have lost a-ablaut; e.g., those where the vowel appears side by side
with /[/. The largest group
o f
apostates comprises
SCHMELZl ,
q u e l l
\
SCHWELLl , GLIMMl ,
and
KLIMMl ,
the bases being schmelz /jmelts/, quel!
/kvel/, schwell /Jvel/, glimm /glim/, and klimm /klim/, respectively; cf.
also
m e l k
l ,
melk /melk/. In these bases, the vowels occur between a
voiced labial (/v/ or /m/) on the one hand and a lateral (/l/) on the other;
no such stem allows a-ablaut in Contemporary German. Bases with initial
labial+lateral-cluster
(
f l e c h t
l ;
p f l e g
l ,
strong forms now obsolete) be-
long here, too.54
Summing up, the unmarked option for 1-base stems is
NO-FULL-ABL.
However,
stems with e-bases or short /-bases have as a rule a-ablaut, though a number of
stems that do not fit neatly into
a-ABL
have moved into
NO-FULL-ABL,
adopting
the unmarked option. I-base stems that have a-ablaut belong either to
CLASS
Hi/iv (a-ablaut and o-ablaut) or to
CLASS V
(only a-ablaut); membership is
determined in terms of the quantity types and the consonantal structures of the
bases.
4.5
Non-I-base stems and ablaut classes
As for non-I-base stems, /-ablaut is the unmarked option. A-base stems belong
to
CLASS VII
(/-ablaut) or
CLASS VI
(a-ablaut):
— A-bases,
c l a s s
vil:
(
e m p
)
f a n g
l ,
h ä n g
l ,
h a l t
l ,
f a l l
\
s c h l a f
l ,
b r a t
l , (
g e
)
r a t
l ,
b l a s
l ,
l a s s l
— A-bases,
c l a s s
v i
:
w a c h s
l ,
w a s c h
l ,
b a c k
l ,
s c h a f f
l ,
s c h l a g
l ,
t r a g
l ,
g r a b
l ,
l a d
l ,
f a h r l
For a stem to be in
CLASS VI,
its base must meet a severe condition of ‘nor-
malcy’: short vowels must be followed by voiceless consonants, long vowels
54 Cf. Segeral & Scheer (1998: 50), who, however, maintain that the verbs in question do
have (underlying) ö-forms in the finite past (which surface as o-forms) — an untenable
position that mixes up synchrony and diachrony. For an overview of the diachronic devel
opment see Solms & Wegera (1993: 270-278), for details Solms (1984) and Chirita (1988).
141
must be followed by voiced consonants. (Consequently, ablaut forms due to
«-ablaut are invariably long.) Moreover, as in the case of
CLASS
Ill/iv, 5-bases
are not admitted. Stems are in
CLASS VI
if, and only if, their bases conform to
one of the following two patterns:
/a/ + voiceless consonant (# /s/)
/a:/ + voiced consonant ( f Izl)
The remaining A-base stems are in
CLASS VII.
(The bases are
fa n g
/far)/,
hang
/hag/,
halt
/halt/,
fa ll
/fal/,
sch la f
/Jla:f7,
hr at
/bra:t/,
rat
/ra:t/,
bias
/bla:z/,
and
lass
/las/, respectively.)
As regards the small set of stems that have monophthongal U-bases, ab-
laut class membership is straightforward given the above observations on
‘conspicuous’ vocalism.
RUFL
and
STOSSL
show marked base vocalism: the
vowels are long although followed by voiceless obstruents; they are in
CLASS
VII
(the unmarked option for non-I-base stems). There is only one more U-base
stem, namely
KOMML,
which is in
CLASS V.
Likewise, stems with diphthongal
U-bases would not qualify for the marked option u-ablaut. Two such stems are
in
CLASS II (SAUFl , SAUGl ),
which agrees with the disposition of diphthongal
bases not to show change of quality type;55
LAUF
l
and
HAUL
show /-ablaut as
expected in the case of U-base stems. The distribution apparently does not
correlate with synchronically determinable properties of the stems.56
5
Ablaut: an integrated view
Any adequate analysis of ablaut should uncover how ablaut contributes to the
categorisation of verb forms. Consider, e.g.,
v e r s p r e c h e n w
with its forms
versprechen
(infinitive or lst/3rd person plural of the present),
versprachen
(lst/3rd person plural past indicative), and in particular
versprochen
(past
participle). Obviously, it is attributable to the occurrence of the stem form
versproch
that the verb form
versprochen
functions as a past participle. To
grasp this fact, one might be inclined to resort to functional types of stem
forms:
versproch
is a ‘participle stem’ and t h e r e f o r e
versprochen
is a
participle form. However, as noted in Section 1.3, supra, an account like this
would be liable to circularity. After all, a stem form would qualify as a ‘ parti-
55 Sometimes
sc h n a u b e n w
is mentioned as having a ‘poetic’ form schnob.
56 Diachronically, the distribution corresponds to the distinction between ‘old diphthongs’
and ‘new diphthongs’ (diphthongs going back to the so-called New High German diph-
thongisation). Stems with bases in ‘new diphthongs’ show simple ablaut, stems with bases
in ‘old diphthongs’ show full ablaut (but SCHEIDL has changed classes, cf. note 30, supra).