Post by Admin on Dec 13, 2021 4:27:04 GMT
Some Indo-Uralic Aspects of Hittite
Alwin Kloekhorst
University of Leiden
Since the Anatolian branch has been shown to have been the first
one to have split off from PIE, in cases where the Anatolian
evidence diverges from the evidence from the other IE
languages it is sometimes problematic to determine which
linguistic situation is more original. In this article it is claimed
that in some cases evidence from the Uralic language family,
which is hypothesized to have been a sister to the IE language
family, may be decisive in solving such problems. Two case studies
are presented that support this claim.
The Anatolian branch can be shown to have been the
first to split off from Proto-Indo-European because several
instances can be identified in which Hittite shows an original
situation where all other Indo-European languages have
undergone a common innovation (cf. Kloekhorst 2008: 7-11).
This means that the Indo-Hittite hypothesis as postulated
already in the 1920s is cogent in the sense that the ancestors
of the speakers of the non-Anatolian Indo-European
languages shared a period of common innovations that no
longer reached the ancestors of the speakers of Proto Anatolian.
This brings about a methodological problem: if a
certain feature is attested in all non-Anatolian Indo-European
languages but not in Anatolian, to what extent can this
feature be reconstructed for Proto-Indo-European? On the
one hand one could argue that the feature was lost in
Anatolian and that its presence in PIE can safely be assumed
(as is usually done); on the other hand, however, one could
now also argue that the feature is due to a common innovation
of the non-Anatolian languages and that its absence in
Anatolian shows that it was not present in PIE. The same is
true for the inverse: if a certain feature is attested in Anatolian
but in none of the other IE languages, it is usually assumed
that we are dealing with an Anatolian innovation.
Nevertheless, it has now become just as possible that we are in
fact dealing with an archaic feature and that its absence in the
other IE languages is due to a common loss that they
underwent during their period of shared innovations.
Of course, there are sometimes internal arguments to
decide the matter. For instance, the verbal root *mer- is
attested in Anatolian with the meaning ‘to disappear’, whereas
in all other IE languages it has the meaning ‘to die’. In this
case it is less probable that an original meaning ‘to die’ would
develop into ‘to disappear’, whereas the other way around is
perfectly cogent (cf. euphemistic expressions like to pass away,
to be gone, French disparaître). We can therefore safely assume
that Anatolian has preserved the original situation, which
means that the semantic shift of ‘to disappear’ > ‘to die’ is a
common innovation of the non-Anatolian IE languages (cf.
Kloekhorst 2008: 8).
Whenever such arguments are absent or not as strong as
in the case mentioned here, it becomes problematic to decide
whether an Anatolian peculiarity is to be regarded as an
innovation or an archaism.
The Proto-Indo-European language did not come about
out of nowhere. Like any other language it must have had its
precursor and relatives. Although the literature about a further
relationship between Indo-European and other language
families is vast and many candidates have been suggested, in
my opinion the most promising view is the theory that
connects Indo-European with the Uralic language family.1
Most recently, Kortlandt (2001: 1) argued that “we may think
of Indo-European as a branch of Indo-Uralic which was
transformed under the influence of a Caucasian substratum.”
Moreover, he was able to give a meaningful interpretation of
the PIE verbal system as built up of morphemes attested in
Proto-Uralic as well (o.c.).
This new approach to the pre-history of Proto-Indo-European may
also be helpful in breaking the methodological
impasse we are sometimes faced with when linguistic features
attested in the Anatolian languages conflict with the material
of the non-Anatolian Indo-European languages. It is my
intention to present here two cases with regard to personal
pronouns where an Indo-Uralic point of view can elucidate the
position of the Hittite material vis-à-vis the non-Anatolian IE
languages.
Alwin Kloekhorst
University of Leiden
Since the Anatolian branch has been shown to have been the first
one to have split off from PIE, in cases where the Anatolian
evidence diverges from the evidence from the other IE
languages it is sometimes problematic to determine which
linguistic situation is more original. In this article it is claimed
that in some cases evidence from the Uralic language family,
which is hypothesized to have been a sister to the IE language
family, may be decisive in solving such problems. Two case studies
are presented that support this claim.
The Anatolian branch can be shown to have been the
first to split off from Proto-Indo-European because several
instances can be identified in which Hittite shows an original
situation where all other Indo-European languages have
undergone a common innovation (cf. Kloekhorst 2008: 7-11).
This means that the Indo-Hittite hypothesis as postulated
already in the 1920s is cogent in the sense that the ancestors
of the speakers of the non-Anatolian Indo-European
languages shared a period of common innovations that no
longer reached the ancestors of the speakers of Proto Anatolian.
This brings about a methodological problem: if a
certain feature is attested in all non-Anatolian Indo-European
languages but not in Anatolian, to what extent can this
feature be reconstructed for Proto-Indo-European? On the
one hand one could argue that the feature was lost in
Anatolian and that its presence in PIE can safely be assumed
(as is usually done); on the other hand, however, one could
now also argue that the feature is due to a common innovation
of the non-Anatolian languages and that its absence in
Anatolian shows that it was not present in PIE. The same is
true for the inverse: if a certain feature is attested in Anatolian
but in none of the other IE languages, it is usually assumed
that we are dealing with an Anatolian innovation.
Nevertheless, it has now become just as possible that we are in
fact dealing with an archaic feature and that its absence in the
other IE languages is due to a common loss that they
underwent during their period of shared innovations.
Of course, there are sometimes internal arguments to
decide the matter. For instance, the verbal root *mer- is
attested in Anatolian with the meaning ‘to disappear’, whereas
in all other IE languages it has the meaning ‘to die’. In this
case it is less probable that an original meaning ‘to die’ would
develop into ‘to disappear’, whereas the other way around is
perfectly cogent (cf. euphemistic expressions like to pass away,
to be gone, French disparaître). We can therefore safely assume
that Anatolian has preserved the original situation, which
means that the semantic shift of ‘to disappear’ > ‘to die’ is a
common innovation of the non-Anatolian IE languages (cf.
Kloekhorst 2008: 8).
Whenever such arguments are absent or not as strong as
in the case mentioned here, it becomes problematic to decide
whether an Anatolian peculiarity is to be regarded as an
innovation or an archaism.
The Proto-Indo-European language did not come about
out of nowhere. Like any other language it must have had its
precursor and relatives. Although the literature about a further
relationship between Indo-European and other language
families is vast and many candidates have been suggested, in
my opinion the most promising view is the theory that
connects Indo-European with the Uralic language family.1
Most recently, Kortlandt (2001: 1) argued that “we may think
of Indo-European as a branch of Indo-Uralic which was
transformed under the influence of a Caucasian substratum.”
Moreover, he was able to give a meaningful interpretation of
the PIE verbal system as built up of morphemes attested in
Proto-Uralic as well (o.c.).
This new approach to the pre-history of Proto-Indo-European may
also be helpful in breaking the methodological
impasse we are sometimes faced with when linguistic features
attested in the Anatolian languages conflict with the material
of the non-Anatolian Indo-European languages. It is my
intention to present here two cases with regard to personal
pronouns where an Indo-Uralic point of view can elucidate the
position of the Hittite material vis-à-vis the non-Anatolian IE
languages.