DOI Page / Citation link:
https://doi.org/10.11588/diglit.1167#0086
Examples of minimal pairs:
• beCOpi /'thori/ "willow" < Ir.t — bTu>pi /'tori/ "hand" < dr.t (Sahidic has TCOpe in
both cases).
• b0COM /'chom/ "garden" < k>nw — bX(0A /'com/ "roll of papyrus, book" < dm'
(Sahidic has ©CDA /'kiom/ and A0XDA6 /'co:mg/ respectively)
• bdOCC0 /a'kV "magician" < hkiw — bdvK(u /a'ko/ "destruction" < Iqy.t (Sahidic has
9>KO and &K(o respectively)
While this rule holds true for by far most cases, there are a few irregular developments,
e.g. ghs.t "female gazelle" > bS^2CI, dsr "to be read" > beG0pUJ, dhr.w "to be bitter"
(stative) > b©02- On these cf. Fecht (i960: note 505 on p. i8of.), Osing (1976a: notes 611
and 614 on p. 612-614), and Vergote (1945: 44-46).
The distinction is neutralized if no stressed vowel follows. Both signs then occur in
mutually exclusive environments: 1) The aspirate sign is used before sonorants (A I ml, N
/n/, ^ HI, p hi, B /p/, (1)1, OV /vff). Thus, the aspirate sign may not only be a representative
of an Egyptian stop of class 1, but also of a stop of class 2 in this position (e.g.: dbc
"10000" > b©6cV STB&; qniui "sheaf' > bXN&3\ sKI-ta^ff). 2) The non-aspirate sign is
used in all other cases.
Hintze (1947b: 204f.) explains that the aspirate writing indicates a partial desonorization
of the following sonorant (thus XN&.'B' would be spoken something like knnaw). Vycichl
(1960a and 1990: 2i8f.) assumes that the stop was strenghthened since it was felt in the
syllable offset (assuming that XNA/if was spoken something like ak-navi). I prefer Hintze's
solution.
The rules outlined above may lead to morphological alternations, e.g.:
• p}i (demonstrative pronoun) > b<j>dJ /'phaj/ in autonomous position but bn<M- /paj-/
(proclitic) in attributive position (Sahidic has n&I and n€l- respectively)
• p> (definite article) > b<§- before sonorants, but bn- elsewhere (Sahidic has IT
throughout).
Whereas according to the general rule the stop opposition is only preserved directly
preceding a stressed vowel, there are cases in which the distinction is also observed in
other environments:
• The constrast of aspirates and non-aspirates is mostly preserved in Greek loan words
without regard to word stress.
• In reduplicated verbs, sometimes the expected form appears (thth "to confuse" >
b0O£)T£2) /'th3xt3x/), sometimes aspiration is generalized to both syllables by anal-
ogy (ptpt "to tread down" > HOTt^eT /'photph3t/ instead of *chOTn£T), sometimes
the non-aspirate stop is generalized {w'dwid "to be green" > bO'8'OTO'8'eT /'wotwot/
instead of *0?TOeOtfeT).
• The status nominalis of verbs and nouns almost always retains an etymological
aspirate (Kasser 1994b: note 24 on p. 291), although the mutation of its stressed
vowel to -£- indicates that it was unstressed.
86
• beCOpi /'thori/ "willow" < Ir.t — bTu>pi /'tori/ "hand" < dr.t (Sahidic has TCOpe in
both cases).
• b0COM /'chom/ "garden" < k>nw — bX(0A /'com/ "roll of papyrus, book" < dm'
(Sahidic has ©CDA /'kiom/ and A0XDA6 /'co:mg/ respectively)
• bdOCC0 /a'kV "magician" < hkiw — bdvK(u /a'ko/ "destruction" < Iqy.t (Sahidic has
9>KO and &K(o respectively)
While this rule holds true for by far most cases, there are a few irregular developments,
e.g. ghs.t "female gazelle" > bS^2CI, dsr "to be read" > beG0pUJ, dhr.w "to be bitter"
(stative) > b©02- On these cf. Fecht (i960: note 505 on p. i8of.), Osing (1976a: notes 611
and 614 on p. 612-614), and Vergote (1945: 44-46).
The distinction is neutralized if no stressed vowel follows. Both signs then occur in
mutually exclusive environments: 1) The aspirate sign is used before sonorants (A I ml, N
/n/, ^ HI, p hi, B /p/, (1)1, OV /vff). Thus, the aspirate sign may not only be a representative
of an Egyptian stop of class 1, but also of a stop of class 2 in this position (e.g.: dbc
"10000" > b©6cV STB&; qniui "sheaf' > bXN&3\ sKI-ta^ff). 2) The non-aspirate sign is
used in all other cases.
Hintze (1947b: 204f.) explains that the aspirate writing indicates a partial desonorization
of the following sonorant (thus XN&.'B' would be spoken something like knnaw). Vycichl
(1960a and 1990: 2i8f.) assumes that the stop was strenghthened since it was felt in the
syllable offset (assuming that XNA/if was spoken something like ak-navi). I prefer Hintze's
solution.
The rules outlined above may lead to morphological alternations, e.g.:
• p}i (demonstrative pronoun) > b<j>dJ /'phaj/ in autonomous position but bn<M- /paj-/
(proclitic) in attributive position (Sahidic has n&I and n€l- respectively)
• p> (definite article) > b<§- before sonorants, but bn- elsewhere (Sahidic has IT
throughout).
Whereas according to the general rule the stop opposition is only preserved directly
preceding a stressed vowel, there are cases in which the distinction is also observed in
other environments:
• The constrast of aspirates and non-aspirates is mostly preserved in Greek loan words
without regard to word stress.
• In reduplicated verbs, sometimes the expected form appears (thth "to confuse" >
b0O£)T£2) /'th3xt3x/), sometimes aspiration is generalized to both syllables by anal-
ogy (ptpt "to tread down" > HOTt^eT /'photph3t/ instead of *chOTn£T), sometimes
the non-aspirate stop is generalized {w'dwid "to be green" > bO'8'OTO'8'eT /'wotwot/
instead of *0?TOeOtfeT).
• The status nominalis of verbs and nouns almost always retains an etymological
aspirate (Kasser 1994b: note 24 on p. 291), although the mutation of its stressed
vowel to -£- indicates that it was unstressed.
86