This book presents a new theory of grammatical categories - the Universal Spine Hypothesis - and reinforces generative notions of Universal Grammar while accommodating insights from linguistic typology.
Kasper Boye’s “Epistemic Meaning” is a book based on the author’s doctoral dissertation of the same name, supervised by Peter Harder at the University of Copenhagen. Boye’s work can informally be divided into two major parts, as indicated by the book’s subtitle -- the first part, which might be termed the “crosslinguistic” section, includes a nearly exhaustive study of the philosophical and linguistic notions of epistemic meaning and provides examples of its manifestations in a wide-range of languages (Chapters 1 and 2). Following this, Boye devotes the next two-thirds of his work to a functional account of epistemic meaning as expressed in terms of semantic mapping, scope issues, and the notion of propositions (Chapters 3, 4, and 5), among other topics. These three chapters may be called the “functional-cognitive” section of the work, and it is here especially that Boye invokes and builds upon the tradition of the same name (citing and adding to the ideas of Aikhenvald 2004, 2010, Bybee 1985, de Haan 1999, Givón 1982, Halliday 1970, 1985, and many others). The author borrows from other (formal) traditions as well, presenting a descriptive account of the problems of epistemic meaning while remaining focused on what conclusions can be drawn from the evidence at hand. Chapter 6 presents a short conclusion.
Beginning in the first chapter, Boye discuss his preliminary assumptions and theoretical and empirical bases, as well as how he plans to elaborate the work that has been done so far on epistemic meaning. Crucially, Boye takes the notion of epistemic meaning to include both evidentiality and epistemic modality. In discussing the history of thought in these areas, that is, how evidentiality and epistemic modality have been considered to be traditionally related, Boye comments that three ideas about the relationship between these two categories are prevalent. These three ideas maintain, namely, that evidentiality and epistemic modality are separate but perhaps related, that the two categories do in fact bleed into one another, despite their separation, and thirdly, that one category is a subset of another (p. 1-2). Boye’s goal is explicit with respect to the research that precedes his: the two categories are distinct but themselves subsets of a larger category of epistemicity (p. 2).
What are the meanings of evidentials and epistemic modals, and why categorize them as part of a larger group? Boye recognizes the complexities of defining these ideas, but writes that his term epistemic/epistemicity “covers not only the notion of degree of certainty or epistemic support, but also the notion of source of information or epistemic justification” (p. 15). Careful not to beg the question, Boye invokes philosophical history dating back to ancient Greece, noting that epistemology as a whole examines “the notion of knowledge”, which includes “notions such as belief, certainty, epistemic necessity and epistemic possibility...[which] maybe be generalized over in terms of the notion of epistemic support”, the very definition of “epistemic modal meaning” (p. 18). Such notions carry over into the category of justification/evidentiality, which takes its place next to epistemic support under a larger category of “justificatory support” (i.e., epistemicity, p. 36).
After Boye establishes the parameters for his research, he begins looking at various data in earnest. Chapter 2 is devoted not only to what characterizes (or would characterize) a general epistemic system, but how that system is instantiated in a number of specific languages (his fifty-language sample is mostly North and South American, p. 12-13). In his words, “not all epistemic expressions are found in notionally coherent systems. In fact, the majority seems to be found in incoherent systems or in no system at all” (p. 49). That is to say, languages do not all deal with epistemicity, evidentiality, or epistemic modality morphologically, but all languages must deal with these problems in some way (p. 114). The tendency here, then, to define groups of phenomena as (or some deviation from) “notionally coherent systems” is a way to provide descriptive adequacy for epistemic systems, and the linguistic examples given (p. 53-120) reflect this methodology.
In his fifty-language sample, Boye finds a number of possibilities with respect to epistemic systems and their subsystems. Many of the languages, if not most, handle epistemic systems morphologically. As for those that don’t, “[i]t may be hypothesized that even if a language does not show systematic coding of epistemic meaning, it still shows ‘scattered coding’, and that even if a language does not show grammaticalized coding, it still shows lexical coding” (p. 114). According to Boye, then, the universality of epistemic systems is coded in some extra-linguistic conceptual system.
In order to get at what this conceptual system is, or, at the very least, to describe how this conceptual system interacts with the linguistic system, Boye turns towards a functional-cognitive examination of epistemicity in the latter two thirds of his book. Chapter 3 discusses semantic maps, which are a way to visually depict and describe the phenomena outlined and discussed so far. Semantic maps include “a number of notions[,] each of which generalizes over a number of specific meanings” and between the notions themselves “‘connecting lines’, each of which symbolizes a relation between notions” (p. 126).
Boye shows that epistemic systems can be described by selectively highlighting or foregrounding certain aspects of the semantic map. For instance, in a semantic map detailing epistemic notions of Direct and Indirect Justification, and Full Support, Partial Support, and Neutral Support, the Egyptian Spoken Arabic ‘jimkin’ and Finnish ‘-ne’ (both glossed as ‘probable/possible’) particles can be said to fall within the Partial Support/Neutral Support section of the map (p. 144), while “Turkish, Jacaltec, and perhaps Ladakhi” have “synchronically polyfunctional expressions” that fall under the domain of Full Support/Partial Support (p. 140). Crucially, as the map makes clear, no language will have epistemic expressions that fall under two unconnected nodes on the semantic map, as in Full Support/Neutral Support, without including the middle node of Partial Support (p. 155).
Boye shows that in languages like Slave and West Greendlandic, there are expressions that highlight the map vertically in that they indicate Indirect Justification/Partial Support. Overall, then, the semantic map makes the case for a larger category of epistemicity (i.e., the map as a whole) and the related subcategories of justification (evidentiality) and support (epistemic modality), with any individual language having the ability to connect the notions in very specific and limited ways (p. 155-156).
Chapter 4 seeks to argue that all three types of epistemic meaning (epistemicity, evidentiality, and epistemic modality) share certain “scope” properties and thus support the descriptive account given so far (p. 183). Boye defines scope operationally: “[t]he scope of a given meaning is defined as the meaning to which the meaning at hand applies...a property of linguistic meanings rather than of linguistic expressions” (p. 183). That is, Boye is arguing that the meaning properties of epistemic expressions scope over certain parts of sentences, akin in spirit to quantifiers, operators, and negatives. Likewise, as we would assume that the latter three categories would take their scopal behavior as part of their definition, so should we assume that we could similarly define and describe epistemic scope.
For example, taking the notion of speech acts, Boye argues that they are, or can be, epistemic in nature: “at least one type of such communicative interactions [i.e., speech acts] must be understood as having a meaning correlate: ‘Illocutionary acts’ depend on ‘illocutionary forces’, and illocutionary forces must ultimately be understood as a kind of meanings [sic]...these meanings may be coded as semantic (conventional) meanings” (p. 187). In other words, meanings such as epistemicity “scope over” the domain of the entire sentence. Other phenomena that share similar scope properties include state-of-affairs complements and propositional complements, which, according to Boye, can be distinguished in certain English examples (p. 192, Boye’s example #4.14):
I saw [him write a letter]. (State-of-affairs complement) I saw [(that) he was writing a letter]. (Propositional complement)
Such differences are handled morphologically in Turkish, for instance, and most of the rest of Chapter 4 is dedicated to giving examples of how epistemic expressions fit into different clause types (declaratives, interrogatives, and imperatives) crosslinguistically, whether morphologically or periphrastically.
The notions of propositions and states-of-affairs return in Chapter 5 for a functional-cognitive analysis in earnest, which is the culmination of Boye’s work so far. The questions examined here include: What is the purpose of an epistemic system?; and Why are such phenomena linguistically ubiquitous? For Boye, the answer to these lies in part with the idea that “evidential meanings are linguistic prompts…to evoke an epistemic-justification structure” and that “epistemic modal meanings are linguistic prompts to evoke an epistemic-support structure” (p. 293-294). Each structure differs from itself by evoking structures of differing strengths (p. 293-294). Tied to this is the idea that “the justificatory-support structure represents a general cognitive capacity for anchoring pieces of conceptual information about the world in other pieces of conceptual information, or in other types of concepts” (p. 294). Epistemic systems, then, are a way to articulate claims of certainty about propositions, or from the other direction, linguistic cues to build a structure that allows a listener to comprehend claims of certainty about a proposition. This is, as Boye points out, an important survival skill (p. 294-297).
Returning to the semantic map discussed in Chapter 3, Boye remarks that “[t]he continuity of the three regions of the map [i.e., epistemicity, evidentiality, and epistemic modality] reflects the coherence of the three conceptual structures evoked by the meanings found in the relevant regions” (p. 301). These are anchored to propositions via reference structure, which accounts for the properties of scope discussed previously (p. 316). Boye closes the chapter discussing these ideas in detail with respect to the data given. Chapter 6 presents a short conclusion, reaffirming his goals and general conclusion, that evidentiality and epistemic modality are indeed subcategories of epistemicity as a whole.
Boye’s work is a valuable addition to the literature on epistemic meaning -- he is very clear about his thesis and goals and is careful to show how each is demonstrated and met. In an important way, Boye not only builds upon previous work, but settles a question posed by his predecessors: How is the epistemic system characterized with respect to evidentiality and epistemic modality? His solution of putting the latter two under the domain of the former allows for descriptive adequacy to be reached in this area of inquiry.
In his attempt to explain much of the data and the concepts he was dealing with, I was surprised and impressed to see that Boye invokes not only the major players in his own discipline, but many in formal linguistics (Werner Abraham, Lisa Matthewson, Guglielmo Cinque, and others) and philosophy (Donald Davidson, Carnap, Grice, Searle, Strawson) as well, which is a move that more ideologically committed researchers would tend to avoid. In this broader sense, there is certainly something of interest here for any researcher of epistemicity (or for that matter, epistemology).
As I mentioned above, the work can be informally delineated into two sections: one on crosslinguistic data and epistemic systems and subsystems, and the other on the functional-cognitive account of these data. Any typologist working in the vein of Aikhenvald (2004) and others who is interested in epistemic meaning would do well to investigate Boye’s work, if not but for the bulk of crosslinguistic data in the first “section” alone. These first two chapters, comprising a total of 125 pages, are an incredibly valuable resource in and of themselves to any linguist interested in a crosslinguistic breakdown of epistemic meaning and its instantiations.
Naturally, of course, the data presented here are of use to those in related disciplines, such as syntacticians, semanticists, and others, especially those interested in American languages. However, the vast majority of data given is synchronic -- linguists who specialize in grammaticalization may not find much in terms of comparative data in this text alone.
As for the second section, I found myself struggling at times with some of the conceptual material, but this is likely because my own training as a syntactician has necessitated a paucity of study into functional-cognitive linguistics. Despite this, however, Boye’s solutions and ideas seem innovative and plausible, although I am unsure if his conclusions achieve explanatory adequacy, as the author is careful, on several accounts, to maintain that his in-depth descriptive account of the data may preclude such an effort.
Nevertheless, in characterizing the epistemic system in such a way, Boye has hardly closed the door on the topic -- in fact, his work is likely to promote new and further interesting investigations into the problems discussed. Seeing an author make use of a wide array of data and theories ought to be an inspiring call for other like-minded linguists to do the same in order to bridge disciplines and ideas, and to promote using similar research as a springboard for the creative investigation of linguistic problems.
Aikhenvald, Alexandra. 2004. Evidentiality. Oxford: Oxford University Press.
Aikhenvald, Alexandra. 2010. Imperatives and Commands. Oxford: Oxford University Press.
Bybee, Joan. 1985. Morphology: A Study of the Relation between Meaning and Form. Amsterdam: John Benjamins.
de Haan, Ferdinand. 1999. Evidentiality and epistemic modality: Setting boundaries. Southwest Journal of Linguistics 18: 83-101.
Givón, Talmy. 1982. Evidentiality and epistemic space. Studies in Language 6: 23-49.
Halliday, M. A. K. 1970. Functional diversity in language as seen from a consideration of modality and mood in English. Foundations of Language 6: 322-365.
Halliday, M. A. K. 1985. An Introduction to Functional Grammar. London: Edward Arnold.
ABOUT THE REVIEWER:
R. E. Santana-LaBarge is a student of Elly van Gelderen at Arizona State University in Tempe, Arizona. His fields of interest include Minimalist syntax, formal semantics, the philosophy of language, and Cartesian rationalism.