The Arabic Language

faʿala
, 
faʿila
, 
faʿula
. These 
correlate with three patterns of the imperfect verb: 
yafʿalu
, 
yafʿilu
, 
yafʿulu
. The 
imperfect patterns are derived from the perfect; they consist of a prefix, the 
verbal stem and an ending. The verbal stem in the imperfect has the form CCvC, 
in which the first radical is vowel-less and the second radical has a vowel that 

---

118
The Arabic Language
correlates with the vowel of the perfect. The underlying form is CvCvC like the 
perfect, but in combination with the prefix and the ending this would produce 
a non-allowed form with a sequence of four Cv’s, hence, the deletion of the first 
vowel (|ya-ḍa-ri-bu| → 
yaḍribu
). According to the grammarians, the derivation of 
the imperfect from 
faʿila
and 
faʿula
perfects is rule-bound: they always have an 
imperfect 
yafʿalu
and 
yafʿulu
, respectively. In the case of 
faʿala
, the vowel of the 
imperfect is either 
u
or 
i
: here the correlation is based on 
samāʿ
‘hearing’, in other 
words, it has to be learned from the speakers and cannot be predicted. Those 
faʿala
verbs that have a velar or pharyngal as second or third radical, for example, 
zaraʿa
‘to sow’, have a regular (
qiyāsī
) imperfect with 
a
, 
yazraʿu
. The endings 
-u
/
-a
/-0/ of 
the imperfect verb are case endings (cf. above, pp. 111f.). Finally, the prefixes of 
the imperfect cannot be pronouns, since they co-occur with nominal agents, as in 
yaḍribu zaydun
‘Zayd hits’, and, as we have seen above, a sentence cannot contain 
two agents. Consequently, they are to be regarded as grammatical markers (
dalāʾil
) 
that have no independent status like the pronouns.
When a verb contains one of the glides 
w
or 
y
, it is commonly known as a 
weak verb (
muʿtall
). For this class of verbs, the grammarians set up a number of 
morphological and phonological rules to explain the various changes that affect 
them. Morphologically, they patterned weak verbs on the sound verbs, and by 
reconstructing underlying forms they managed to achieve greater symmetry. For 
the hollow verbs (verbs in which the medial radical is 
w
or 
y
), for instance, they 
reconstructed the medial radical by referring to morphologically related words: 
qāla
‘to say’ is related to 
qawl
‘speech’, 
sāra
‘to go’ to 
sayr
‘journey’, and 
ḫāfa
‘to fear’ 
to 
ḫawf
‘fear’, therefore, their radicals are 
q-w-l
, 
s-y-r
and 
ḫ-w-f
, respectively. Their 
next step was to determine to which perfect pattern these verbs belonged on the 
basis of the underlying forms of the imperfect 
*yaqwulu
, 
*yasyiru
and 
*yaḫwafu
. 
The first two must come from a perfect 
faʿala
(
yaqwulu
cannot come from 
faʿila
, 
for then it would have an imperfect 
yafʿalu
, and it cannot come from 
faʿula
, since 
that pattern is reserved for intransitive verbs; for 
*yasyiru
the only possibility 
is 
faʿala
). But 
*yaḫwafu
must come from 
*ḫawifa
, since otherwise the imperfect 
pattern 
yafʿalu
would be inexplicable.
The next step was the explanation of the phonological changes: how do we get 
from 
*yaqwulu
and so on to 
yaqūlu
and so on? We cannot go into all the details of 
the intricate system of rules set up by the grammarians, so we will have to limit 
ourselves to some examples (an extensive analysis of all weak forms, including 
the perfect forms such as 
qultu
, 
sirtu
, 
ḫiftu
, can be found in Bohas and Guillaume 
1984). One of the first rules states that a combination of /a/ followed by /w/ or 
/y/ plus a vowel is changed into 
ʾalif
/”/, an abstract element on the underlying 
level, thus /qawama/ → /qa”ma/, which is realised phonetically as [qɑːma] (cf. 
below, p. 121); likewise /sayara/ → /sa”ra/ and /ḫawifa/ → /ḫa”fa/.
A second rule changes the order of the vowels and the glides for ease of pronun
-
ciation, so that, for instance, /yaqwulu/ becomes /yaquwlu/, and /yasyiru/ 

---

The Arabic Linguistic Tradition 
119
becomes /yasiyru/. A third rule states that there is a special relationship between 
/i/ and /y/, /u/ and /w/, and /a/ and /”/; therefore, after a vowel, glides often 
change into the related glide of the vowel, for example, /iw/ → /iy/, /uy/ → /
uw/, as in the words /miwqa”t/ → /miyqa”t/ and /muysir/ → /muwsir/. In the 
case of the weak verbs, this rule is invoked to explain derivations such as the 
imperfect of 
ḫāfa
: /yaḫwafu/ → /yaḫawfu/ → /yaḫa”fu/.
In the system of phonological rules with which the Arabic grammarians 
operated, one of the most important principles was that of the relative weight 
of the phonemes of the language. They set up a hierarchy that went from the 
lightest elements, the vowels, via the glides to the consonants. Within the class of 
the vowels, they determined the following order from the lightest to the heaviest 
vowel: /a/ > /i/ > /u/. In the explanation of phonological changes, this relative 
order of the phonemes played an important role, since speakers of the language 
were credited with an aversion to combinations that were too heavy. Thus, for 
instance, the combination /-iya-/ is possible, because it goes from heavy to light, 
whereas the combination /-iyu-/ is regarded as too heavy and therefore imper-
missible because it contains a passage from a heavy to a heavier element (e.g., in 
/qa”ḍiyu/ ‘judge’, which therefore becomes /qa”ḍiy/).
Among the derivational alterations of the verb are the so-called verbal measures 
(see above, Chapter 6, pp. 95–7). Arabic grammarians regarded these as part of the 
derivational morphology. They distinguished between three primary augmented 
verbal measures, 
ʾafʿala
, 
faʿʿala
and 
fāʿala
, each with its own concomitant meaning, 
but all expressing the syntactic process of 
taʿdiya
‘transitivity’, that is, increasing 
the valency of the verb with one, for example, 
ʿalima 
‘to know something’ versus 
ʿallama 
‘to instruct someone in something’ and 
ʾaʿlama 
‘to inform someone about 
something’. These measures, just as the base stem, could receive an additional 
augment 
-t-
to express the process of 
muṭāwaʿa
, lit. ‘obedience, compliance’: 
faʿʿala
> 
tafaʿʿala
; 
fāʿala
→ 
tafāʿala
; 
faʿala
→ 
iftaʿala
; 
ʾafʿala
→ 
istafʿala
. The augment 
-n- 
in 
infaʿala 
expressed this same notion. 
Muṭāwaʿa
was regarded as the opposite of 
taʿdiya
, that is, decreasing the valency of the verb with one, for example, 
kasara 
‘to break [transitive]’ versus 
inkasara 
‘to break [intransitive]’; 
ʿallama 
‘to instruct 
someone about something’ versus 
taʿallama 
‘to be instructed, to become learned 
in something’ (Larcher 2012: 75–7). What mattered to the grammarians was the 
fact that the augment (
ziyāda
) correlated with an additional meaning.