DOI Page / Citation link:
https://doi.org/10.11588/diglit.1167#0153
mtrw (mVt'rVwV) "witness" > sANTpe /m[n]t're/, bA6ep€
htp ('hatpV) "to be content" (since MK) > ^(DTn /'hotp/
ir.t CjirtV) "to do" > seipe /W, blpl
Itrw (jatVrwV) "river, Nile" > seiOOp /p:r/, bIOp
They were lost, however, as the second component of a cluster:
ir.t (jirtV) "to do" > ^ipe /W, bIpI
mwt (mawtV) "to die" > s^Otf /'mil/
rrot framtV) "man" > rmt > sp(lMve /'ronW, bpOUU
zfcj (zabtV) "to laugh" > zbt > HLIHAe /'sopV, bCu)6l
shn.t (sahVntV) "to command" (since MK) > H^fte /'sahna/, bC&2NI
nfr ('nafrV) "good" > sNCW<je /'nufo/, bNO'irqi
sretr ('sanVtrV) "incense" > sCONTE /'sonta/, hCONf
A possible explanation for this loss would be that at a certain stage of the language
progressive assimilation took place in consonant clusters, e.g.: (jirtV) = /'jirt(h)V/ >
/'jirrV/ > /W.
3.14.3.4 <t> I <r> in unstressed word-initial position
<t> is always preserved in unstressed word-initial position:
• twtw (t[V]'watwV) "image" > TOVOOT /Wot/, beOff(uT
• tmfy.t (tVm'JVytV) "mat" (since MK) > STAH /'tme/, ^AH
The same is also true for <r> in most words:
• r-prj.t (r[V]-'pVrVjtV) "temple" > spn€ /r'pe/, bep«^ei
• rmy.t "tear" (since MK) > sp^eiH /rm'je/ (or /rm'je/?), b6pAH
But in three very common words <r> was lost:
• r (r[V]-, proclitic) "to, from, than" > s>b€- hi- (some scholars assume that the prepo-
sition should be read ir rather than r, e.g. Edel 1955/64: §760)
• rh (r[V]h[V]-,proclitic) "can" > S.>>U|- ///-
• rdi.t "to give" > s-bf /'ti/
3.14.3.$ Morphophonological alternations and analogical levelling
Conditioned phonetic developments as described in the preceding sections could lead to
morphophonological alternations if a consonant was lost in some forms of the verb and
preserved in other forms of the same verb. Cf.:
• hqr (h[V]'qarV) "to become hungry" > '■o.KO /'hka/
• hqr.w ('haqVrwV) (stative of the same verb) > b20K€p /'hakar/
However, alternations like these were eliminated by analogical levelling in most verbs.
<t> and <r> were either irregularly conserved in all forms:
153
htp ('hatpV) "to be content" (since MK) > ^(DTn /'hotp/
ir.t CjirtV) "to do" > seipe /W, blpl
Itrw (jatVrwV) "river, Nile" > seiOOp /p:r/, bIOp
They were lost, however, as the second component of a cluster:
ir.t (jirtV) "to do" > ^ipe /W, bIpI
mwt (mawtV) "to die" > s^Otf /'mil/
rrot framtV) "man" > rmt > sp(lMve /'ronW, bpOUU
zfcj (zabtV) "to laugh" > zbt > HLIHAe /'sopV, bCu)6l
shn.t (sahVntV) "to command" (since MK) > H^fte /'sahna/, bC&2NI
nfr ('nafrV) "good" > sNCW<je /'nufo/, bNO'irqi
sretr ('sanVtrV) "incense" > sCONTE /'sonta/, hCONf
A possible explanation for this loss would be that at a certain stage of the language
progressive assimilation took place in consonant clusters, e.g.: (jirtV) = /'jirt(h)V/ >
/'jirrV/ > /W.
3.14.3.4 <t> I <r> in unstressed word-initial position
<t> is always preserved in unstressed word-initial position:
• twtw (t[V]'watwV) "image" > TOVOOT /Wot/, beOff(uT
• tmfy.t (tVm'JVytV) "mat" (since MK) > STAH /'tme/, ^AH
The same is also true for <r> in most words:
• r-prj.t (r[V]-'pVrVjtV) "temple" > spn€ /r'pe/, bep«^ei
• rmy.t "tear" (since MK) > sp^eiH /rm'je/ (or /rm'je/?), b6pAH
But in three very common words <r> was lost:
• r (r[V]-, proclitic) "to, from, than" > s>b€- hi- (some scholars assume that the prepo-
sition should be read ir rather than r, e.g. Edel 1955/64: §760)
• rh (r[V]h[V]-,proclitic) "can" > S.>>U|- ///-
• rdi.t "to give" > s-bf /'ti/
3.14.3.$ Morphophonological alternations and analogical levelling
Conditioned phonetic developments as described in the preceding sections could lead to
morphophonological alternations if a consonant was lost in some forms of the verb and
preserved in other forms of the same verb. Cf.:
• hqr (h[V]'qarV) "to become hungry" > '■o.KO /'hka/
• hqr.w ('haqVrwV) (stative of the same verb) > b20K€p /'hakar/
However, alternations like these were eliminated by analogical levelling in most verbs.
<t> and <r> were either irregularly conserved in all forms:
153