Evidentiality in Uzbek and Kazakh

0.4 
Outline 
The structure of this dissertation is designed to roughly parallel the development of the 
morphemes in question, beginning with background information on the Turkic languages and an 
overview of the Turkic verbal system, then moving onto the use and meaning of the bound past 
tense morphemes discussed above, and finishing with two chapters on the meanings of the 
copular forms. 
In the first chapter, “Evidentiality: Historic, Areal, and Genetic Considerations,” I 
provide a historical basis for the content of the following chapters. This involves a discussion of 
the history of the study of evidentiality (and related phenomena) and the situation of Uzbek and 
Kazakh within this broad field of study. The connection between the study of evidentiality and 
the Turkic languages is a natural one, as the first known reference to 
EVIDENTIAL
MEANING
was 
made in an early of grammar of Turkic languages – Mahmoud Al-Kāšğarī’s 11
th
Century Dīwān 
Luγāt at-Turk (1982, as noted in Friedman 2003) - and because the Turkic languages appear to 
be responsible in part for the development of an “evidentiality belt” in Central Eurasia. 
Chapter 2, “Predication in Uzbek and Kazakh,” provides an outline of the verbal system 
in Uzbek and Kazakh and provides a template for the analysis the verbal systems of other Turkic 
languages. Many grammars of Turkic languages simply provide long lists of possible 
combination of morphemes and disregard the compositionality of complex verb forms. As 
agglutinating languages, the Turkic languages are particularly ill-suited for this type of treatment, 
as any such analysis necessarily misses the opportunity to make broad generalizations among the 
various forms of the verb, and, furthermore, ignores the similarities between verbal and non-
verbal predicates. By dividing verbal morphology into two types, finite and non-finite, we are 
able to draw parallels between non-finite verbal forms and non-verbal predicates and thereby 

---

10 
predict the distribution of copular forms, including ekan/eken. The finite/non-finite distinction 
made in Chapter 2 enables us to explain why copular forms typically only follow non-finite 
forms of the verb and non-verbal predicates. The uses of ekan/eken that are unrelated to the non-
confirmativity paradigm are discussed in this chapter as well. 
In “The Past and Confirmativity,” the third chapter, I describe the three bound past tense 
morphemes, -di/-DI, -gan/-GAn, and -(i)b/-(I)p, focusing on their markedness values for 
confirmativity and for other relevant features such as 
TEMPORAL DISTANCE
and 
DEFINITENESS
.
As described above, the primary differences between these morphemes is that the simple past 
-di/-DI is marked as [+
CONFIRMATIVE
], the so-called perfect -gan/-GAn is unmarked for 
confirmativity [Ø 
CONFIRMATIVE
],
and the converbial past -(i)b/-(I)p is marked as 
[-
CONFIRMATIVE
]. 
Chapter 4, “Evidential Meanings of Ekan/Eken” discusses the morphemes most often 
described as “evidential” in Uzbek and Kazakh: ekan and eken, which are marked as non-
confirmative [-
CONFIRMATIVE
]. The expression of non-confirmativity by these morphemes has 
two possible interpretations: non-firsthand information source (a type of evidential meaning) and 
admirativity. This chapter examines the evidential component of these morphemes’ meanings 
within the complex of meanings encompassed by the sub-category of non-confirmativity and 
shows the range of evidential meaning that they express. Although these forms have their basis 
in the copula, their distribution differs from other copular forms, and this distribution is outlined 
in Chapter 4 as well. 
The second interpretation of ekan/eken - 
ADMIRATIVITY
- is covered in Chapter 5, 
“Ekan/Eken and the Expression of Emotivity.” It is well-known that forms that express 
evidential meaning frequently also express surprise, doubt, or the reception of unexpected 

---

11 
information, all types of admirative meaning. Although some authors consider admirativity a 
distinct verbal category (DeLancey 1997; 2001), the connection between admirative and 
evidential meaning can be accounted for by employing the non-confirmative approach 
mentioned previously. A purely non-confirmative approach, however, presents some difficulties, 
because when ekan/eken is found in questions, one possible result is a rhetorical question. This 
result rhetorical question interpretation is scarcely attested in other languages, and I argue that 
the admirative and rhetorical question meanings expressed by ekan/eken can be unified by 
invoking the concept of 
EMOTIVITY
, which, according to Jakobson (1960), is the use of language 
to express the speaker’s state of mind. As noted in 0.3, the subjective type of 
MODALITY
/
STATUS
is often incompatible with questions, and the incompatibility of non-confirmative meaning in 
certain types of questions may have resulted in the development of strong emotive meaning, 
thereby producing rhetorical questions. 
The final chapter summarizes the major claims of the previous chapters and suggests 
further directions for research. Because Uzbek and Kazakh lie at the heart of a Eurasian 
evidentiality belt, the claims that follow may apply to other Turkic languages and to other 
languages of the region. 

---

12