SLS Indiana Sept 06 1 The /a/→/aj/ Shift in Russian Verbs Hierarchies and Markedness Tore Nesset University of Tromsø
Feb 02, 2016
SLS Indiana Sept 06 1
The /a/→/aj/ Shift in Russian Verbs
Hierarchies and Markedness
Tore NessetUniversity of Tromsø
SLS Indiana Sept 06 2
(1) The /a/→/aj/ Shift
Pre-C stem:Pre-V stem:
igra-l igraj-ut
SLS Indiana Sept 06 3
(1) The /a/→/aj/ Shift
Pre-C stem:Pre-V stem:
igra-l igraj-ut
SLS Indiana Sept 06 4
(1) The /a/→/aj/ Shift
Pre-C stem:Pre-V stem:
igra-l igraj-ut
murlika-l murličØ-ut
SLS Indiana Sept 06 5
(1) The /a/→/aj/ Shift
Pre-C stem: Pre-V stem:
igra-l igraj-ut
murlika-l murličØ-utmurlikaj-ut
Non-productive /a/ suffix replaced by productive /aj/
SLS Indiana Sept 06 6
(7) Two questions
• Which verbs show /a/→/aj/ shift?
• Which inflected forms show /a/→/aj/ shift?
SLS Indiana Sept 06 7
(8) Hypothesis
Marked/non-prototypical verbs/verb forms show /a/→/aj/ shift.
SLS Indiana Sept 06 8
(9) Root-Final Consonant
• Place:
– Labial – Dental – Velar
• Manner:
– Plosive – Fricative – (Sonorant)
• Consonant mutation
– Unmarked – Marked
SLS Indiana Sept 06 9
(10) Root-Final Place
DentalVelar
Labial
76
20
4
15
75
10
22
33
44
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
50-100% -aj
0-49% -aj
-aj not attested
SLS Indiana Sept 06 10
(11) Root-Final Manner
Velarplosive Velar
fricative Dentalplosive Dental
fricative
13
73
13
20
80
0
68
26
6
100
00
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
50-100% -aj
0-49% -aj
-aj not attested
SLS Indiana Sept 06 11
(12) Consonant Mutation
Unmarked: Marked:
Velar:k~č
g~ž
sk~š’:
zg~ž’:
Dental:t~č
d~ž
st~š’:
d~žd:
SLS Indiana Sept 06 12
(12) Consonant Mutation
Unmarked: Marked:
Velar:k~č
g~ž
sk~š’:
zg~ž’:
Dental:t~č
d~ž
st~š’:
d~žd:
SLS Indiana Sept 06 13
(12) Consonant Mutation
Unmarked: Marked:
Velar:k~č
g~ž
sk~š’:
zg~ž’:
Dental:t~č
d~ž
st~š’:
d~žd:
SLS Indiana Sept 06 14
(12) Consonant Mutation
Unmarked: Marked:
Velar:k~č
g~ž
sk~š’:
zg~ž’:
Dental:t~č
d~ž
st~š’:
d~žd:
• Unmarked:Target is single C
• Marked: Target is long C or 2 Cs
SLS Indiana Sept 06 15
(13) Impact of Consonant Mutation
Velarunmarked Velar
marked Dentalunmarked Dental
marked
17
75
8
17
67
17
80
20
0
40
20
40
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
50-100% -aj
0-49% -aj
-aj not attested
SLS Indiana Sept 06 16
(14) Root Shape HierarchyRoot
Dental Velar Labial
Fricative Plosive
Markedalternation
Unmarkedalternation
Suffix: /a/ Suffix: /aj/
SLS Indiana Sept 06 17
(14) Root Shape and MarkednessRoot
Dental Velar Labial
Fricative Plosive
Markedalternation
Unmarkedalternation
Suffix: /a/ Suffix: /aj/
SLS Indiana Sept 06 18
(14) Root Shape and MarkednessRoot
Dental Velar Labial
Fricative Plosive
Markedalternation
Unmarkedalternation
Suffix: /a/ Suffix: /aj/
SLS Indiana Sept 06 19
(14) Root Shape and MarkednessRoot
Dental Velar Labial
Fricative Plosive
Markedalternation
Unmarkedalternation
Suffix: /a/ Suffix: /aj/
SLS Indiana Sept 06 20
(14) Root Shape and MarkednessRoot
Dental Velar Labial
Fricative Plosive??
Markedalternation
Unmarkedalternation
Suffix: /a/ Suffix: /aj/
SLS Indiana Sept 06 21
(15) Root Shape and Hypothesis
• Confirming evidence:
– Marked/non-prototypical roots tend to show
/a/→/aj/ shift.
• Problem:
– Are dental fricatives less marked than
dental plosives?
SLS Indiana Sept 06 22
(16) Inside the paradigm
0
10
20
30
40
50
60
70
80
90
100
3 sg 3 pl 1 pl 1 sg Imp sg Gerund
SLS Indiana Sept 06 23
(17) The /aj/% Hierarchy
Gerund ≥
Imperative ≥
1 singular ≥
1 plural ≥
3 plural ≥
3 singular
SLS Indiana Sept 06 24
(18) Hierarchy and markedness
• Hierarchy corresponds roughly to markedness if– Non-finite — Finite
– Imperative — Indicative
– 1st person — 3rd person
– Plural — Singular
SLS Indiana Sept 06 25
(19) Problem: 1 sg and 1 pl
Gerund ≥
Imperative ≥
1 singular ≥
1 plural ≥
3 plural ≥
3 singular
SLS Indiana Sept 06 28
(22-23) Conclusions
• The /a/ /aj/ shift depends on ➝systematic interaction of hierarchically ordered factors:– The Root Shape Hierarchy– The Inflectional Paradigm Hierarchy
• The /a/ /aj/ correlates with ➝markedness/prototypicality(although the correlation isn’t perfect)
SLS Indiana Sept 06 29
(24) Theoretical implication
Verb classes and inflectional paradigms
behave like structured networks,
not like unordered lists.
SLS Indiana Sept 06 30
(25) More questions
• Is frequency relevant?
• Do verbs with “postfix” -sja behave differently?
• What about prefixed verbs?
• And development over time?
SLS Indiana Sept 06 31
Thanks to
Hyugh Anh John Korba
Chris Wiesen