Scalar Structure Across Categories

Teamleader:
Gillian Ramchand

Time span:
Project starts:
01.12.2006

Project ends:
01.12.2008

Brief description and goals:
1. Introduction and Summary

In our first grammar classes we are often told that ‘A Noun is a name for a person, place, animal or thing’, or, ‘A Verb is a ‘doing’ word.’, with similar notional definitions for the other basic lexical categories (Adjective and Preposition). Although modern syntactic theory has not managed to improve very much on traditional notional definitions of the major parts of speech, practically every theory of grammar makes use of distinct syntactic categories, justified on the basis of formal morphological and distributional criteria. This is particularly true of the traditional ‘lexical’ categories Noun, Verb, Adjective, and to a lesser extent Preposition which have been argued to exist in most, if not all of the world’s languages (see Baker 2003 for a recent study). While it has long been acknowledged, since the advent of X-bar syntax that there are structural parallels across categories (Chomsky 1970, Abney 1987), it has often been implicitly assumed that the major categories are semantically different from each other. Recent work in semantics, however, is beginning to uncover pervasive compositional similarities across categories in the area of scales and part-whole structuring (Schwarzschild 2002, Kennedy and McNally 1999, Krifka 1992, Zwarts 2005). The leading hypothesis behind the project is that the notion of homomorphism or part/whole matching might be as general and unifying a semantic property as X-bar syntax turned out to be for structural representations. To investigate the generality of the idea, I will examine the phenomenon of scalar or part-whole structure within the reference types of all the major syntactic categories. I seek to demonstrate that all categories show pervasive and linguistically relevant denotational distinctions with respect to reference, systematically contrasting homogenous undifferentiated reference with extended, or linearly ordered scalar structures. The idea here is that the abstract structuring principles of reference are the same across categories, even though the basic ontological domains are different— Substance (N) vs. Space (P) vs. Property (A) vs. Time (V).

The project will study a three basic domains where part-structure or scalar structure has been shown to be linguistically relevant, and where elements of different syntactic categories interact: (i) explicit measure constructions internal to V, N, A and P; (ii) interactions between temporal and nominal structure in Verb-Object collocations; and (iii) interactions with V and secondary predicates of the A and P type in resultatives. Detailed data of semantic judgements and entailments will be gathered for three main languages: English (representing the Germanic group), a Romance language ( where differences within the expression of goal and result have already been noticed), and Slavic (which uses prefixes of measure and result, but where the article system within the nominal domain is severely restricted). The first aim of the project is thus empirical and seeks to establish for a group of different languages exactly in which linguistic constructions we find constraints requiring mapping relationships between distinct scalar paths, especially in languages with quite different grammatical resources. The second major goal is to formalize a set of semantic composition rules that are sufficiently abstract and flexible to yield parallel results for syntactic categories representing semantically distinct ontological categories (Space, Time, Property and Substance). The final goal of the investigation is to provide a basis for arguing whether the different syntactic categories involve either exactly the same, or merely similar and parallel functional structure within the context of a universal grammar of hierarchically ordered elements. To this end, we will examine the extent to which the morphological and derivational relationships among categories support a common core. The project is built around a project leader with skills in formal semantic theory and designed to integrate with the more purely syntactic research issues currently being pursued at the Center for Advanced Study in Theoretical Linguistics (CASTL) at the University of Tromsø (see sections 3.4 and 4). The two postgraduate positions will be advertised, but confined to the language areas of Romance and Slavic respectively and a prerequisite of the positions will be demonstrated research experience within formal semantics.

2. Background
Scalar structure can be found in the meanings of (i) prepositions, in the form of spatial paths;
(ii) adjectives, in the form of gradability and comparison with regard to property degree; (iii) nominals, in the form of the part-whole scales required for measurement and (iv) verbs, in the unfolding of events according to a linearly ordered time line. The literature on these topics has begun to investigate the nature and properties of these scales (e.g., whether they are open-ended, monotonic, cumulative, etc.), and how these properties can be used to explain the properties of the different lexical categories. In addition, recent work has begun to take notice of the fact that these seemingly disparate scales are often constrained to map onto each other in a strictly homomorphic way when elements of different categories are merged in complementation structures. Interesting linguistic effects induced by these complex expressions have been derived from this homomorphic relationship, as we now proceed to show in more detail.

2.1 Prepositions (Space)
Important work by Zwarts (Zwarts 1997, Zwarts 2005; and other earlier papers) and Zwarts and Winter 2000 has developed the idea that the proper understanding of the semantics of locative (specially directional) prepositions includes reference to a part-whole structure of paths ordered by the subpath relation and closed under concatenation. This means that the meaning of a prepositional phrase like into the house (as in Alex walked into the house) can be understood as the set of paths that have the inside of the house as their endpoint, and that include a single transition from a not-in-the-house chunk of path to an in-the-house chunk of path. We can thus account for the intuitions about truth-conditions of sentences with such prepositions, but, in addition, properties of their part-whole structure allow us to predict that, while Alex swam is atelic, Alex swam to the beach is telic. That is, we can explain that the nature of the part-whole structure induced by the preposition can affect the aspectual properties of the items it combines with, notably, verbs. Directional prepositions can be divided into two groups, those whose part-whole structure is cumulative (towards, along) and those for which it is not (into, to, from, out of, past) e.g., with towards, concatenating two paths ‘towards the station’ gives rise to a complex path that is also ‘towards the station’, something which doesn’t hold of prepositions such as to. Non-cumulative prepositions introduce a boundedness that cumulative prepositions don’t. When related to the part-whole eventive structure of the verb, those prepositions whose part-whole structures are non-cumulative can change the aspect of the verb to telic, i.e., informally, add boundedness to the verb (cf. Alex swam *in/for an hour vs. Alex swam to the beach in/*for an hour) (see also Krifka 1998).
Studying the properties of the part-whole structures associated with prepositions not only allows us to understand their interaction with verbal aspect, but also properties of the prepositions themselves. For example, once concatenation is available as an operation on paths, it is possible to understand how prepositional phrases can be conjoined with and as in up the stairs and down the stairs. This conjoined prepositional phrase is not the set of paths that are both ‘up the stairs’ and ‘down the stairs’, for that set is empty. Rather, it is the set of paths that are concatenations of up-the-stairs paths and down-the-stairs paths. Concatenation is also useful in drawing an important analogy between prepositions and nouns: prepositions can be pluralized.
For example, in the case of around the house, which denotes the set of paths that enclose the house once, we can concatenate such paths and obtain the set of paths that enclose the house more than once. This gives us the plural version of around, with round and round its overt expression. And Zwarts has also argued (see also Jackendoff 1990) that, just as with count nouns, it is possible to derive mass denotations for prepositions from their count denotations (as in Alex crawled across the desert for hours).

2.2 Adjectives (Properties)
As is well known (see Seuren 1985, von Stechow 1984, Klein 1991, Kennedy 1999, Kennedy 2000, Kennedy and McNally 1999, Svenonius and Kennedy 2005 , Kennedy and McNally 2005, among others), several aspects of the scales of gradable adjectives are linguistically relevant. Adjectival scales can differ from each other in the degrees/measurement values they use, the dimension/kind of measurement they introduce, and in their ordering relation. Antonym pairs like tall/short or expensive/cheap are distinguished because of their ordering relation: tall and short, for example, make use of the same degrees and dimension (height) but their ordering relations are opposite from each other. Non-antonymous adjectives like tall and expensive differ in their dimension parameter (height and price, respectively).

Kennedy and McNally (2005) argue that the structure of the scale itself, i.e., whether it is closed-or open-ended, is relevant in the understanding of degree modification of gradable adjectives. Adjectives like empty or full induce closed scales (i.e., they have minimal and maximal values), whereas the scales of adjectives like long or short are open-ended. This difference can be seen in the fact that, say, a box, can be ‘totally empty’ or ‘totally full’, but it does not make sense to speak of a rope that is ‘totally long’ or ‘totally short’. Modifiers such as half are sensitive to this distinction as well: a box can be ‘half empty/full’, but it is not possible for a rope to be ‘half long/short’ (see e.g., Lehrer 1985, Hay 1998, Kennedy and McNally 1999). Even whether a scale is lower-closed (i.e., with a minimal but no maximal value) or upper-closed (i.e., with a maximal but no minimal value) seems to be relevant: it is possible for a pipe to be ‘fully straight’, but not for it to be ‘fully bent’ (with fully being an upper-endpoint modifier and only the scale of straight having an upper or maximal value). Kennedy and McNally further argue that the homomorphic relation that incremental-theme verbs establish with their objects (see below and Krifka 1989, Krifka 1992, Krifka 1998; also, see Tenny 1994, Jackendoff 1996 and Ramchand 1997) is relevant in determining the properties of the scale of the adjectives derived from these verbs. The generalization is that the scale of such adjectives is totally closed (i.e., it will have minimum and maximum values). For example, eat a cookie is a classic example of a verb with an incremental theme (the homomorphism here is between the subevents of eating and the parts of the cookie: for all of these subevents, we can identify the piece of the cookie that was eaten) and the derived gradable adjective eaten induces a closed scale, as can be seen from the fact that it is possible for a cookie to be ‘half eaten’. The idea here is that, since we can define a beginning point and an endpoint for the event of eating a cookie, we can identify minimal and maximal values for the degree to which the cookie is eaten.
Faller (2000) and Winter (2005) propose a uniform analysis of measure phrase modification in the adjective and prepositional domain and hence take steps towards a cross-categorial analysis of modification phenomena. Measure phrases such as ten meters or five years can modify both adjectives and prepositions, though there are restrictions: ten meters behind/outside/*near/ *on the house, ten meters wide/*narrow/deep/*shallow (cf. Cresswell 1977, Seuren 1985, Bierwisch 1989, Kennedy 2000, among others, for discussion of the adjective cases). That both prepositions and adjectives can take measure phrases can be explained if prepositional phrases, adjectives, and measure phrases denote sets of paths (vectors in Winter 2005; see also Zwarts and Winter 2000). We saw above what such a semantics means for prepositions; for adjectives, it involves treating tall as the set of paths that describe the lengths that are considered tall in a given context. Measure phrase modification of both prepositions and adjectives would then simply be set intersection (but see Svenonius 2004 for a dissenting view concerning the interaction of measure phrases with PPs). Winter explains the restrictions indicated above as cases where the measure phrase is sensitive to whether the scale of the preposition or adjective is open-ended or not (cf. Kennedy and McNally above). Certain adjective-specific phenomena also follow by making reference to their (lack of) minimal and maximal values, such as neutralization or the fact that in sentences such as Johnny is five years old, there is no entailment that Johnny is old (see also Kennedy 2000, who only stipulates neutralization).

2.3 Verbs (Time) and Nouns (Substance)
A long tradition of work has related verbal part-whole event structures homomorphically to the part-whole structures of their nominal objects, in an attempt to explain, e.g., the change in telicity between eat an apple, with a singular count noun as object, and eat apples (see Krifka 1989, Krifka 1992, Krifka 1998, Dowty 1979, Verkuyl 1993, Jackendoff 1996, Wechsler 2005). Crucial in these attempts is the algebraic notion ‘part of’, but this is not the only notion that has been found to be relevant. Krifka (1998) uses the notions of path and temporal precedence in an attempt at a more general understanding of telicity, e.g., with prepositional phrases (see above) and with predicates that express changes of properties, such as bake the lobster (see also Verkuyl 1993, Pustejovsky 1991, Krifka 1998). Krifka also explains the distribution of adverbials like for an hour and in an hour in terms of the compatibility of their part-whole structures with the algebra of events induced by the verb, in a way that is similar to the measure-function restrictions in the nominal domain. Another area where verbs’ event structures and their homomorphic interaction with the structures of other categories have linguistic relevance is in the explanation of phenomena related to secondary predication (as in John pounded the metal flat ). For example, Wechsler 2005, van den Wyngaerd 2001 and Beavers 2002 argue that the contrast between John pounded the metal flat and *John pounded the metal long has to do with the properties of the scale denoted by the adjectives in question: only those adjectives that denote an appropriately bounded scale, like flat, can be result phrases. Beavers (2005) and Wechsler (2005) have argued that further contrasts in this area (The outlaw knocked/*beat the sheriff dead, The rabbits had been battered to death/*dead vs. He shot the miller dead/to death) result from additional scalar constraints on the homomorphism (e.g., durative events require gradable and/or durative prepositional/adjectival scales).

Measure phrase modifiers and their scalar properties are also relevant in understanding the behaviour of nouns themselves. Consider first the fact that measure phrases such as 2 liters can combine with nouns in the pseudopartitive construction (2 liters of oil), but not all measure phrases can do this (*90 degrees of oil). Conversely, 2 liters cannot be compounded (*2-liter oil), but 90 degrees can (90-degree oil). Schwarzschild (2002) derives these contrasts from the monotonicity properties of the scales associated with the measure phrases and, importantly, from the part-whole structure of the nouns in question. Volume is monotonic, whereas temperature is not: if a quantity of oil has a certain volume, then every proper subpart will have lower volumes, and proper superparts will have larger volumes; but, if a quantity of oil has a certain temperature, proper subparts will not have a lower temperature. Pseudopartitive constructions presuppose a monotonic part-whole structure for the stuff being measured; for compounds, non-monotonic structures are required. Interestingly, the part-whole structure of the noun is also relevant: the pseudopartitive construction requires its nouns to be mass (two hours of work vs. *two hours of job), but compounding with count nouns is possible (a 2-hour job). Singular count nouns don’t induce (non-trivial) part-whole structures (i.e., their denotation contains only atomic parts) and are impossible in the pseudopartitive construction because this kind of construction requires monotonicity, a property that is lacking if the noun in question has no proper subparts. On the other hand, mass nouns, such as work or water, do induce a monotonic part-whole structure (see Link 1983 and much ensuing work) (and hence they are impossible in compounds: *2-hour work). On matters relating to measure of the substance introduced by nouns, see also Krifka 1992.

This research is outstanding in its level of detail, depth of analysis and hence in the solid basis it provides for our understanding of the properties of scales and their role in natural language. We believe the time has come to put these insights together with the goal of arriving at a common core for the four lexical categories, an enterprise that, while attempted to some degree in the literature described above, has never been attempted systematically in a formal way.

3. The Project: Aims and Methodologies
As established in the previous section, recent research done on English has opened up a new and intriguing line of inquiry— that of assessing the parallels and differences in the ways that the different lexical categories in natural language encode linearly incremental substructure. The evidence so far is that these connections are real, and pervasive, and are implicated whenever phrasal categories of different types come together in syntax. Moreover, the research in question has also provided a solid foundation of diagnostics, entailment tests and generalizations from English that can be systematically tested against the patterns found in other languages. On the one hand, we aim to discover if a formal unification of these different phenomena within a particular language is possible and/or necessary. On the other, we wish to investigate whether the patterns of scalar homomorphism we find are idiosyncratic to English or whether they are a more general property of the combinatory semantics of natural languages. Preliminary investigations suggest that this is so (Filip 2000 for Czech and Russian; Folli 2003 for Italian), and that the correspondences are general, but a systematic study has so far not been undertaken. This is partly because the field work required within semantics is both complicated and subtle, requiring the investigator to have a close knowledge and feeling for the language she is investigating in order to discover the right (truly parallel) tests and control for contextual effects and ambiguities. For these reasons, the majority of work in formal semantics has used English as the language being investigated. The further assumption has often been that languages do not vary very much in their abstract semantic structure (as opposed to the ways in which they can vary in morphology or word order). While we believe that this is true, in the sense that the human cognitive system underlies all natural language and will hence share many abstract properties and categories, we also believe that there are potential differences between languages precisely at the interface between syntax and semantics. For this reason, the project will focus on three distinct languages, using the foundational work on the semantics of English as a pivot. Given the detailed nature of the semantic database we intend to construct, and the wide range of constructions to be investigated, we do not believe it is feasible for a group of three researchers attempt this kind of parallel corpus for more than three languages within the time span. However, the set of data questions and patterns from each language will be installed on the web as a template for others to use and add to the crosslinguistic picture over time, after the completion of the project.

3.1 The Languages
The second language we will look at, other than English, will be drawn from the Romance group of languages. The reasons for this are twofold. Firstly, languages of the Romance family of have already been noticed to depart considerably from English and the Germanic group in general, in the way that they express spatial prepositional paths in conjunction with motion verbs (see Talmy 1985, Folli 2003, Zubizarretta and Oh 2004). Thus, while the English sentence in (1) can get a reading whereby ‘the boat’ comes pass ‘under the bridge’, the corresponding Spanish sentence in (2) cannot get the path interpretation at all, and only refers to a static floating event.

(1) The boat floated under the bridge.
(2) La barca flotaba debajo del puente
the-FEM boat floated-IMPF under of.the-MASC bridge
’The boat was floating under the bridge’


Moreover, while resultative secondary predicational constructions involving adjectives (adjectival resultatives) are possible in English, they are very restricted in Spanish (3).

(3) El lago se congel´o (* s´olido)
the-masc lake REFL froze-PERF (*solid)
‘The lake froze solid.’

Interestingly, with a derived, and further modified participial adjective, secondary predication does seem to be possible. (Parallel facts can be demonstrated for Italian (see Folli 20030).

(4) El lago se congel´o bien congeladito
the-masc lake REFL froze-PERF well frozen-DIM
‘The lake froze well frozen.’

Given these differences, it is important to know whether they are correlated with the different semantic referential properties of prepositions and adjectives, either separately or with respect to their collocation with verbs. This is therefore an ideal language group with which to set up a systematic set of parallel data with English.

The third language in our data set will be taken from the Slavic family of languages, which once again demonstrates interesting differences from both Romance and Germanic. In particular, the Slavic family (with the exception of Bulgarian) has no nominal articles, and quantization effects interact with the presence of prefixes on the verb. A certain class of prefixes in Slavic has been argued to contribute measure functions to the event and/or object of the verb (Filip to appear, Romanova 2005, as in examples (5) below.

(5) (a) po-iskat’ -look for a while
Petja po-iskal knigu
Peter po-looked book
‘Peter looked for a book for a little while.’

(b) na-brat’ – pick a lot of
Olja na-brala gribov
Olga na-picked mushrooms
‘Olga picked a lot of mushrooms.’

Slavic is therefore important to the cross categorial leading ideas of the project because it represents a language which uses verbal quantification to achieve the same effects as nominal quantification in other languages (see also Partee, Bach, and Kratzer 1987 for work on this crosslinguistic ‘parameter’ ).

3.2 The Tests
The first task of the research group will be to devise a set of constructions to be tested and diagnostics to be used. The challenge here is to start out with a broad ranging and completely parallel set of questions and diagnostics, so that the results will be directly comparable across a wide range of domains. Our domains of inquiry are given in the following three subgroupings.

I. The internal structure of NPs, APs, PPs and VPs with respect to explicit measure constructions Here we will be investigating constructions such as two gallons of milk, three years older than, two miles into the woods, ran for two hours. We will construct tests to probe the cooccurence restrictions on measure phrases and the types of lexical items they are felicitous with. In particular, we will check to see whether the type of measure phrase is sensitive to part-whole structure or the mass/count distinction in the case of nouns, to gradability or boundedness in the case of adjectives, to the cumulativity of the path structure in the case of verbs and prepositions, in the different languages.

II. The relationship between Verbal denotations and their direct complement PPs and NP
Here the question we will ask is whether the classification of a particular verb with respect to its own internal ‘path structure’ imposes restrictions on the semantic classification of either the PP or the NP that it takes as its complement. In English, for example the atelic activity verb run combines with into to give a path reading, but gives a purely locational reading with in (6). With the punctual semelfactive verb jump the pattern is different: path of motion interpretations are possible with both into and in but in the latter, only the punctual (semelfactive) reading of jump is possible on the directional reading (6d) (cf. Ramchand 2005).

(6) (a) run into the room (directional; durative) ’
(b) run on the beach (only locational; durative)
(c) jump into the room (directional; punctual or durativel)
(d) jump in the lake (directional; punctual).

Verb classes for the individual languages will have to be established, together with specific tests to determine the interactions with PP denotations. Similarly, the interaction between mass nominal objects and bounded nominal objects with creation/consumption verbs will be examined in this category.

III. The relationship of the V to a small clause secondary predicate that it combines with to introduce a result subevent or an extra argument
Here the relationship between the verb and the other category phrase is necessarily more indirect— it is embedded within the ‘small clause’ complement of the main V. We will investigate whether the path structure of the verb imposes interpretational restrictions on the adjectives or PPs found in small clause subordinated predicational structures when complex event structures are formed. In English for example, the iterative reading of shoot is compatible with to death, while the punctual reading of shoot is only good with the adjectival non-gradable adjective dead (cf. Beavers 2005) (7a,b). In (7c), we see a similar pattern where the punctual telic verb break can combine with both point-denoting (in) and durational Ps (into) to give a result 7d); while durative manner verb pound may only combine with the durative P into to give a result.

(7) (a) shoot the cowboy dead (punctual; non-gradable A)
(b) shoot the cowboy to death (iterative; durational P)
(c) break the stick in pieces/into pieces.
(d) pound the metal *in pieces/into pieces

The possibility of resultative secondary predications will be investigated in each of the two non-English languages, and the correlations with secondary predicate internal structure established. The project members will work closely together in a team, having weekly meetings and jointly discussing and devising the relevant tests which will work for each particular language. This initial period will be intensely collaborative because it is important that the set of tests administered be as parallel as possible, and cover the major phenomena types in the different languages. A set of tests that is too parochial could well end up being inapplicable to one of the languages, and/or exclude important patterns worth testing. Once the parallel tests are devised, the individual researchers will pursue the empirical side for their own languages, but we will continue to have monthly meetings in the data gathering period to trouble-shoot any problems with the tests, share results, and continue to maintain parallelism. As the data is gathered, it will be entered on a crossclassified database.

3.3 Formalization and Analysis
The second year of the project will involve analysis of the data and an attempt to arrive at a set of generalizations in terms of a unified formalism. Where there are differences among the languages, we will also describe and attempt to formalize the parameters that distinguish them.

At the midpoint of the project, we intend to have a small workshop with invited international participants who are acknowledged experts in the various empirical domains. At this workshop we will present and share our own results and promote discussion on possible ways to formally distinguish and/or unify the phenomena across the domains of Noun, Verb, Adjective and Preposition. The invited guests will include: Manfred Krifka (Berlin), Roger Schwarzschild (Rutgers), Chris Kennedy (Northwestern) and Joost Zwarts (Utrecht). We believe that this formalization stage is important because it will allow us to couch generalizations in a clear and objective language that is understandable and replicable from researcher to researcher. In this way, we hope to contribute to cumulative scientific understanding in this area.
The final aspect of the project lies in assessing the relationship of these patterns to theories of syntax. In particular, if strong patterns concerning semantic denotational structure are found among the categories, it opens the question of whether each lexical category should be dominated by parallel or identical functional structure. To this end, we will try to correlate the semantic parametric differences among our languages to the morphology and the syntactic relationships within those languages.
In addition to the database of test questions and answers that we will produce, we also intend to publish the results of our analysis in journal articles, both in syntax and formal semantics. We would also plan to publish the proceedings of the workshop. We anticipate that Linguistics and Philosophy, Natural Languages Semantics and Natural Language and Linguistic Theory would be ideal places to publish our more theoretical results.

3.4 Connections to Subprojects at CASTL, University of Tromsø
One of the major intellectual partners in this project is the Centre for the Advanced Study of Theoretical Linguistics (CASTL) at the University of Tromsø. CASTL’s remit is to explore the scope and limits of human language variation in the major areas of syntax and phonology. The subprojects in that group include ‘The Structure of Nominals’, ‘Clausal Structure and the VP’, and ‘Nanosyntax’. Peter Svenonius is also the manager of an Yngre Fremragende Forsker (YFF) project on the category Preposition, called ‘Moving Right Along’. What all of these projects have in common is a preoccupation with the notion of lexical category, with decomposing it structurally and morphologically to discover parallels and differences (both within, and selectively across languages). One currently live hypothesis in the research group is the idea that lexical categories are not actually discrete, but merely involve more or less of a universally available hierarchically ordered set of functional projections (Starke, lecture presentations, UiTø). These are important and interesting projects, but no subgroup connected to CASTL is currently looking at those issues from a formal semantic perspective. If we are to believe (as many generativists now do) that syntax and semantics are strongly correlated, then an answer to the syntactic crosslinguistic questions should be attacked from the semantic side as well. At the very least the facts concerning semantic commonalities across lexical categories will need to be taken into account in any final theory of universal functional structure and the syntax-semantics interface.

Thus, in addition to the project’s own working group, the three researchers on this project will attend group meetings from other CASTL subprojects and present ongoing results as they feed into those subprojects’ research questions. It is for this reason that the last phase of the project is explicitly concerned with making connections to syntactic theory. There is currently a lot of research momentum and focus at the University of Tromsø, and this project has been designed to be distinct from, but complementary to the questions being asked by our colleagues.

4. Why These Researchers, in This Place, Now
I have had a long history of research in the area of event semantics and its connections to argument structure and nominal denotation, culminating in a monograph in 1997 on Aspect and Argument Structure. Since then, I have continued to work on the syntax and semantics of events, and I am in the process of writing a monograph on event structure and verbal meaning, under contract with Cambridge University Press (Ramchand 2005). I have also been working within the area of prepositions and prefixes and their relation to event structure. I recently supervised a doctoral thesis at Oxford University called Constructing Telicity in English and Italian, and that work culminated in a joint paper on path of motion structures in those two languages (Folli and Ramchand 2001). I have also co-authored with Peter Svenonius on the syntax and semantics of the verb-particle construction (Ramchand and Svenonius 2002), and have written a recent paper in conjunction with the subproject on Slavic prefixes here at the University of Tromsø (Ramchand 2004). I have already supervised one PhD project on aspect and telicity in a Romance language, Italian (Folli 2003), and am in the process of supervising a PhD project on Slavic prefixes (Romanova 2005). For this reason, I believe I am in a good position to lead and guide a postdoctoral research team working in this area on Romance and Slavic languages respectively. The University of Tromsø has also recently acquired an independent postdoctoral researcher, Luisa Mart´ý, who has just completed her doctoral thesis in formal semantics at the University of Connecticut. She has strengths in the area of semantic formalism and is a native speaker of Spanish. Luisa Mart´ý is an important addition to the research environment here in Tromsø in the area of semantics, and I anticipate that she will be a strong influence on the Romance side of the project.

The University of Utrecht has also recently launched a joint reading group and blog with a number of researchers at the University of Tromsø , precisely in the area of scalar structure, where many of our interests converge. I hope that a serious funded project on this end will allow this interaction to continue to flourish, and I anticipate that research visits from one institution to the other will be fruitful for both sides. Joost Zwarts is a member of the Utrecht group, and Øystein Nilsen, a joint CASTL and Utrecht postdoctoral fellow is also a member of the group.

As leader of the project, I am also affiliated with the Center for the Advanced Study of Theoretical Linguistics (CASTL), a Norwegian University centre of excellence. This close relationship to advanced syntactic research, and to my colleagues involved in the syntax of lexical categories, and prepositions in particular, makes for a productive and stimulating environment for this kind of semantic project. It means that we will be in an ideal position to take advantage of current syntactic research, and to create a stimulating collaborative environment where we can make real progress on the scientific questions that concern us all. I believe that research in this area is equally important for CASTL, since it forms a crucial foundation for any syntactic theorizing regarding the decomposition of syntactic categories and the number and inventory of functional heads in grammar.

5. Implications for the Science of Language and Cognition
While the project as set out above has very clearly defined goals and outcomes, the particular research results also have important implications for other scientific areas, in particular cognition. The concept of Universal Grammar, ushered in by Chomsky in the late 50s, has generated much controversy over the past few decades, but nobody seriously doubts now that a certain aspect of human intelligence is genetically given and underpins the possibility of natural language. It is still an open question, however, exactly how much of that natively given endowment is actually specific to language, and how much is part of our general cognitive capacity. While it is beyond the scope of this project to do directly psychological work ourselves, the members of the project intend to interact with groups working on cognitive linguistics of categories (within Norway and Europe more generally) and teams doing psycholinguistic work on the conception of space and time in order to keep abreast of these results. A project investigating the semantic commonalities across grammatically given categories is important precisely because it potentially taps into our general cognitive structuring of the environment, independent of the symbolic/grammatical primes required for the syntactic system of language. Conversely, if we find that syntactic structure systematically grammaticalizes the concepts under investigation in a way not immediately derivable from functionalist or cognitive constraints, then we have important evidence for the role of a universal set of semantic combinatoric principles, derived from, but not exhaustively underwritten by our general cognitive faculties.
(references appended)


References
Abney, Steven. 1987. The English noun phrase in its sentential aspect. Ph.D. thesis, MIT, Cambridge, Ma.

Baker, Mark C. 2003. Verbs, Nouns, and Adjectives: Their Universal Grammar. Number 102 in Cambridge Studies in Lingusitics, Cambridge: Cambridge University Press.

Beavers, John. 2002. Aspect and the distribution of
prepositional resultative phrases in English. LinGO
Working Papers, Stanford University 2002-7.

Beavers, John. 2005. Scalar complexity and the structure of events, ms., Stanford University.

Bierwisch, Manfred. 1989. The semantics of gradation. In Manfred Bierwisch and Ewald Lang, eds., Di
mensional Adjectives, Berlin: Springer-Verlag.

Chomsky, Noam. 1970. Remarks on nominalization. In

Roderick A. Jacobs and Peter S. Rosenbaum, eds., Readings in English Transformational Grammar, 184–221, Washington, D.C.: Georgetown University Press.
Cresswell, Max. 1977. The semantics of degree. In

Barbara Partee, ed., Montague Grammar, 261–292, New York: Academic Press.

Dowty, David R. 1979. Word Meaning and Montague Grammar: The Semantics of Verbs and Times in Generative Semantics and in Montague’s PTQ. Dordrecht: Reidel.

Faller, Martina. 2000. Dimensional adjectives and measure phrases in vector space semantics. In Stefan

Kaufmann Faller, Martina and Marc Pauly, eds., Formalizing the Dynamics of Information, Stanford, CA: CSLI Publications.

Filip, Hana. 2000. The quantization puzzle. In Carol Tenny and James Pustejovsky, eds., Events as Gram
matical Objects, 39–96, Stanford, Ca.: CSLI.

Filip, Hana. to appear. Prefixes and the delimitation of events. Journal of Slavic Linguistics 12.

Folli, Rafaella. 2003. Deriving Telicity in English and Italian. Ph.D. thesis, Oxford University, Oxford.

Folli, Raffaella and Ramchand, Gillian. 2001. Getting results: Motion constructions in Italian and Scottish

Gaelic. In K. Megerdoomian and L.A. Bar-el, eds., WCCFL 20 Proceedings, 101–114, Somerville, Ma.: Cascadilla.

Hay, Jennifer. 1998. The non-uniformity of Degree Achievements, a talk given at 72th Annual Meeting of the LSA, New York.

Jackendoff, Ray. 1990. Semantic Structures. Number 18 in Current Studies in Linguistics, Cambridge, Ma.: MIT Press.

Jackendoff, Ray. 1996. The proper treatment of measuring out, telicity and perhaps even quantification in English. Natural Language and Linguistic Theory 14:305–354.

Kennedy, Chris. 1999. Gradable adjectives denote measure functions, not partial functions. Studies in the Linguistic Sciences 29.1.

Kennedy, Christopher. 2000. Polar opposition and the ontology of ‘degrees’. Linguistics and Philosophy 24:33–77.

Kennedy, Christopher and McNally, Louise. 1999. From event structure to scale structure: Degree modification in deverbal adjectives. In Tanya Matthews and Devon Strolovitch, eds., Proceedings from SALT IX, 163–180,
Ithaca, NY: CLC Publications.

Kennedy, Christopher and McNally, Louse. 2005. Scale structure and the semantic typology of gradable predicates. Language 81:2.

Klein, Ewan. 1991. Comparatives. In A. von Stechow and D. Wunderlich, eds., Semantik: Ein internationales Handbuch der zeitgenossichen forschung, Berlin: Walter de Gruyter. Krifka, Manfred. 1989. Nominal reference and temporal constitution and quantification in event semantics. In P. van Emde Boas R. Bartsch, J. van
Bentham, ed., Semantics and Contextual Expression, 75–115, Dordrecht: Forsi. Krifka, Manfred. 1992. Thematic relations and links between nominal reference and temporal constitution. In Ivan A. Sag and Anna Szabolcsi, eds., Lexical Matters, 29–53, Stanford, Ca.: CSLI.

Krifka, Manfred. 1998. The origins of telicity. In Susan Rothstein, ed., Events and Grammar, 197–235, Dordrecht: Kluwer.

Lehrer, A. 1985. Markedness and antonymy. Journal of Linguistics 21:397–429.

Link, G. 1983. The logical analysis of plurals and mass terms: A lattice-theoretical approach. In R. B¨auerle, C. Schwarze, and A. von Stechow, eds., Meaning, Use and Interpretation of Language, 302–323, Berlin:
Walter de Gruyter. Partee, Barbara, Bach, Emmon, and Kratzer, Angelika. 1987. Quantification: A crosslinguistic perspective. Ms., NSF proposal, University of Massachusetts, Amherst. Pustejovsky,

James. 1991. The syntax of event structure. Cognition 41:47–81.

Ramchand, Gillian. 1997. Aspect and Predication: The Semantics of Argument Structure. Oxford: Oxford University Press.

Ramchand, Gillian. 2004. Time and the event: The semantics of Russian prefixes. In Peter Svenonius, ed., Nordlyd, University of Tromsø, www.ub.uit.no/munin/nordlyd/.

Ramchand, Gillian. 2005. Verb meaning and the lexicon. Ms. University of Tromsø.

Ramchand, Gillian and Svenonius, Peter. 2002. The lexical syntax and lexical semantics of the verb-particle construction. In Line Mikkelsen and Chris Potts, eds., Proceedings of WCCFL 21, Somerville, Ma.: Cascadilla Press.

Romanova, Eugenia. 2005. Perfectivity in russian, ms. University of Tromsø.

Schwarzschild, Roger. 2002. The grammar of measurement. In Brendan Jackson, ed., Proceedings of SALT 12, Ithaca, NY: CLC Publications.

Seuren, Peter. 1985. The structure and selection of positive and negative gradable adjectives. Papers from the Parasession on the Lexicon 336–346.

Svenonius, Peter. 2004. Spatial prepositions in English, ms. University of Tromsø; available at ling.auf.net.

Svenonius, Peter and Kennedy, Chris. 2005. Northern norwegian degree questions and the syntax of measurement, ms. University of Tromsø and University of Chicago.

Talmy, Leonard. 1985. Lexicalization patterns: Semantic structure in lexical forms. In Timothy Shopen, ed., Language Typology and Syntactic Description, I: Clause Structure, 57–149, Cambridge: Cambridge University Press.

Tenny, Carol. 1994. Aspectual Roles and the Syntax-Semantics Interface. Dordrecht: Kluwer.
van den Wyngaerd, Guido. 2001. Measuring events. Language 77:61–90.

Verkuyl, Henk J. 1993. A Theory of Aspectuality: The Interaction between Temporal and Atemporal Structure. Cambridge: Cambridge University Press. von Stechow, Arnim. 1984. Comparing semantic theories of comparison. Journal of Semantics:3 1–77. Wechsler,
Stephen. 2005. Resultatives under the ‘event-argument homomorphism’ model of telicity. In N. Erteschik-Shir and T. Rappoport, eds., The Syntax of Aspect, Oxford: Oxford University Press.

Winter, Yoad. 2005. Cross-categorial restrictions on measure phrase modification. Linguistics and Philosophy 28.

Zubizarretta, Maria-Luisa and Oh, Eunjeong. 2004. The lexicon syntax interface: The case of motion verbs, unpublished ms. Zwarts, Joost. 1997. Vectors as relative positions: A compositional semantics of modified PPs. Journal of Semantics 14:57–86.

Zwarts, Joost. 2005. Prepositional aspect and the algebra of paths, to appear, Linguistics and Philosophy.

Zwarts, Joost and Winter, Yoad. 2000. Vector space semantics: A model-theoretic analysis of locative prepositions. Journal of Logic, Language, and Information 9:169–211.