1. Chierchia’s Ph.D.-student years: 1979–1983

Gennaro Chierchia’s application to enter the Ph.D. program in Linguistics at UMass Amherst in Fall 1979 reached us through the Institute for International Education, which handled all Fulbright applications. We had no writing sample to look at and no correspondence with the applicant. We saw that his background was entirely in philosophy at the University of Rome, and I don’t think we even knew the title of his diploma thesis, Da Carnap a Montague: Rilevanza Linguistica della Semantica dei Mondi Possibili, “From Carnap to Montague: Linguistic relevance of possible worlds semantics.” And we had not heard of the professors who wrote letters for him. Altogether the application looked quite “risky.” But he would have a Fulbright fellowship for the first year; so we decided to admit him for the M.A./Ph.D. sequence, with the understanding that there was no guarantee that he could stay beyond the first year. So he came.

He arrived with his wife Isa Orvieto at the end of August, and after they had spent a few days in the Campus Center without much contact, Emmon and I found them and brought them to live with us while they hunted for an apartment and things to furnish it with.¹ That was a week that we and they look back on as the beginning of a lifelong friendship.

And we very quickly understood how very wonderfully suited he was for our Ph.D. program. In his first semester, he was exempted from Linguistics 610, our first-semester introduction to semantics, and took instead the normally third-semester Linguistics 753, Montague Grammar. And he got the only A+ in the class on his term paper, “On Carlson’s treatment of bare plurals.” Already in his second semester, Spring 1980, he started thinking about doing a generals paper on kinds. At that time our Ph.D. program required two generals papers, one in syntax and/or semantics and one in phonology and/or historical linguistics, to be written in the fourth and fifth semesters, so Gennaro was off to an unusually fast start. He was worried about phonology and doing a phonology generals paper – if he had had his way, he would have done only semantics and syntax with some philosophy on the side. But he discovered to his surprise that he enjoyed phonology with Lisa Selkirk and Alan Prince, and he wrote his second generals paper on “raddoppiamento sintatico” in Italian. That paper turned into a NELS paper published in 1982, and later, in 1986, he published an article on length, syllabification, and the phonological cycle in Italian in the Journal of Italian Linguistics. He also found his way into psycholinguistics courses in our program and took a serious interest in psycholinguistics and cognitive science more generally, driven by hard questions about mental models, mental representations, and human competence in the domains of semantics and logic.

Gennaro’s years in our Ph.D. program, from 1979 to 1983, came right at the time of our interdisciplinary Sloan Grants in Cognitive Science. The first was a collaborative project between Linguistics and Computer & Information Science (COINS) from 1978–1980. One of the most successful results of that grant was the hiring of Lyn Frazier in 1978 in a new position in psycholinguistics whose first two years were funded by the grant. Lyn soon became actively engaged in joint research with Charles Clifton and with Keith Rayner, both in Psychology, and those collaborations greatly strengthened our ties with the Psychology Department. Lyn had been at UMass for just a year when Gennaro arrived; his work with Lyn is mentioned below. The second Sloan grant was a five-year grant from 1980 to 1985, with four participating departments, Linguistics, COINS, Psychology, and Philosophy, and at that same time we instituted a Cognitive Science Program, with provisions for specializations in Cognitive Science within the Ph.D. programs in each of the four cooperating departments. In Linguistics it was agreed that the normal four-year Ph.D. program would be extended to a five-year program for students doing a Cognitive Science specialization, implying an additional year of funding. The Sloan grants were a very big help for research activities and especially for the support of students as Research Assistants in relatively ‘poor’ departments like Linguistics and Philosophy in the years that we had them.

In Spring 1980 Gennaro took a very “Sloany”² interdisciplinary Linguistics-Philosophy-COINS seminar on Formal Semantics and Computational Semantics that I co-taught with Elliot Soloway of our COINS department and a Sloan-funded visitor, Douglas Moran, who had worked with Joyce Friedman on computer models of Montague grammar. Topics included computer implementations of model-theoretic semantics (Friedman et al. Reference Friedman, Moran and Warren1978a, Reference Friedman, Moran and Warren1978b), AI approaches to semantics, problems of integrating model-theoretic semantics with linguistic theory and processing models (including issues I had been wrestling with in ParteeReference Partee, Bäuerle, Egli and von Stechow1979), dynamic partial models, non-monotonic logics, and models of belief and inconsistency. Gennaro was one of three students in the seminar for credit, and his participation greatly raised the level of the class. He wrote a nice term paper on the problems of the semantics of belief sentences and hyperintensionality, a topic to which he indirectly returned in his explorations of property theory and alternatives to Montague’s intensional logic.

In his second year, 1980–81, Gennaro was in my very lively Fall seminar on anaphora, cross-listed with COINS and Philosophy but very linguistic in content; others in that seminar included Yasuaki Abe, Dorit Abusch (Philosophy), Dan Finer, James Pustejovsky, and Mats Rooth. The seminar included presentations by Edwin Williams and me on our debates (started the previous spring in an introductory semantics class) on formal semantics versus Chomskyan “logical form,” and presentations by Mats Rooth on VP-deletion, by Emmon and me on our new joint work on pronouns and reflexives in Montague grammar, by Irene Heim on donkey-sentences, by Jonathan Mitchell on Castañeda’s quasi-indexical he*, known since David Lewis’s work (Lewis Reference Lewis1979a) as a de se pronoun, and much more. And in Spring 1981 he participated in a Linguistics and Philosophy seminar that Emmon and I led on tense and aspect and anaphora. Mats Rooth was in that one too; he had started just a year before Gennaro, and they were often in seminars together; other participants included Jonathan Mitchell, Dan Finer, Yasuaki Abe, Peter Sells, James Pustejovsky, and auditors Dorit Abusch, Murray Kiteley, Barry Richards, and Wynn Chao. Toward the end of the semester a major focus of the seminar was trying to better understand the new work of Hans Kamp on Discourse Representation Theory (DRT), particularly Kamp’s work on how DRT might help some problems about aspect and temporal anaphora, and the equally new work of Barwise and Perry (Barwise and Perry Reference Barwise and Perry1981), including Barwise’s work on events (Barwise Reference Barwise and Perry1981).

Spring of 1981 was when Irene Heim was finishing her dissertation, which she defended in the fall (while starting a postdoc at Stanford.) Irene and Gennaro overlapped during Gennaro’s first two years – she had started in 1977, and had nearly stopped taking courses when Gennaro arrived in 1979. My interdisciplinary Sloan-funded workshop on “indefinite reference” which gave Irene her dissertation topic (she found the problems non-trivial and interesting, and realized that no one really had a good semantic account of them) had been in the year before Gennaro arrived, with its first part in Fall 1978 and its second part in Spring 1979. My memories of Gennaro and Irene carrying on long intense debates off in some corner during conferences must come mainly from the years after they had both finished.

Also in the Spring of 1981, Gennaro took a psycholinguistics seminar with Lyn Frazier. Gennaro was one of the first of many students to work with Lyn on problems of semantic processing, and he has gone on to become one of the leaders in the growing fields of semantic processing and semantic acquisition, as discussed in other chapters in this volume. In that seminar, he wrote a paper suggesting potential ways of testing Greg Carlson’s hypothesis about bare plurals not being quantifier phrases by means of experiments related to Philip Johnson-Laird’s work on mental models. And he participated in some of our Thursday evening Cognitive Science Program interdisciplinary meetings, including one in February featuring mostly philosophers discussing issues in the foundations of cognitive science (Gary Matthews on children and philosophy, plus short talks by Jay Garfield, Elliot Soloway, and others), and one in March on semantics, which featured a presentation by Gennaro on Hans Kamp’s new Discourse Representation Theory and its relation to issues of logical form, compositionality, and model-theoretic semantics, as well as presentations by Emmon Bach and by Irene Heim. I have a copy of Gennaro’s handout from his presentation, where he managed to pack into one page the basic issue of Discourse Representation Structure (DRS) as an intermediate level of representation, the DRS handling of possible and impossible donkey-sentences, and something that looks like it might be the beginnings of the argument that he and Mats Rooth developed in their paper for the 1983 NELS at UMass (Chierchia and Rooth Reference Chierchia, Rooth, Jones and Sells1984) against some of Kamp’s claims about the indispensability of the intermediate DRS level in accounting for anaphora.

And that spring he wrote his semantics generals paper. I haven’t found a record of the title of Gennaro’s semantics generals paper, but I know that it further developed his work on an alternative approach to Carlson’s theory of bare plurals, using a Cocchiarella-influenced foundation instead of Montague’s intensional type theory. So I think it is probably the paper he refers to in several places as his 1981 manuscript “English bare plurals, mass nouns, and the structure of the world,” a version of which was later published (Chierchia Reference Chierchia, Borowski and Finer1983a). And very soon after that he had two papers published developing those themes, his 1982 WCCFL paper, “Bare plurals, mass nouns, and nominalization” (Chierchia Reference Chierchia, Flickinger, Macken and Wiegand1982a), and his 1982 paper in Linguistics and Philosophy, “Nominalization and Montague Grammar. A semantics without types for natural languages” (Chierchia Reference Chierchia1982b). Those two papers brought his work to the wider attention of the semantics and philosophy of language communities. That work, and Heim’s dissertation, and Rooth’s work on a two-dimensional semantics for focus constructions not long after, were a major step forward in foundational semantic research. They represented some of the first work anywhere by linguists with enough formal facility to propose serious changes to the metatheory we worked in.³ Up until then, “we linguists” had been largely dependent on philosophers and logicians to provide the theoretical tools for us to work with. When philosophers asked me why I was working with a typed rather than an untyped logic, I could only reply that that’s what Montague had provided, and I had no competence to try to judge or change that logic. Linguists have always been open to theoretical innovations from philosophers and logicians – David Kaplan once remarked that linguists seemed like vacuum cleaners, though Emmon Bach noted wryly that he hoped we weren’t just taking in dust and dirt. But in semantics, linguists had been more or less relying on first-order predicate logic until Montague’s work, and through the 1970s most of the Montague-influenced work by linguists adopted Montague’s formal framework without serious modification and focused on extending empirical coverage, exploring alternative approaches to the syntax–semantic interface, looking for possible constraints to make the theory more “linguistic,” and sometimes worrying about how to integrate a Chomskyan perspective on language “in the head” and the Frege–Montague approach to semantics in terms of mind-external truth-conditions. Gennaro was the first linguist, I am pretty sure, to challenge both Montague’s type theory and his possible-worlds-based account of intensionality. The transition from “Montague grammar” as the name of the field to the now-standard “formal semantics” had several causes; one of the most important was the kinds of challenges to the foundations that Gennaro, Irene, and Mats (and Angelika) raised.

Gennaro’s background in philosophy and logic was clearly part of what made him able to propose foundational changes, but his argumentation was strongly grounded in empirical linguistic phenomena. His acquaintance with Nino Cocchiarella’s work played a big role. He writes in the introduction to his dissertation (Chierchia Reference Chierchia1984a), as a preamble to his acknowledgment to Cocchiarella:

Starting with his first-semester work on Carlson’s treatment of bare plurals, he came to view that construction as one instance of a process of nominalizing predicates, other instances being the formation of mass terms, infinitives, and gerunds, as well as the nominal use of color terms like blue, and to see Montague’s rigid adherence to type theory as an obstacle to achieving an explanatory account of such processes. In his L&P paper, he showed how one could adapt one of Cocchiarella’s non-standard second-order logics (HST*) to construct a non-typed variant of Montague’s intensional logic (IL*) and use it to obtain an improved account of bare plurals as part of a general account of nominalizations extending also to infinitives and gerunds. The paper is quite long and contains difficult material (I depended on my philosophy colleagues to help evaluate technical parts that were over my head); but it is lucidly and beautifully written. It includes along the way the most interesting arguments I had seen up to that point within the model-theoretic tradition for the inclusion of a separate level of logical form within a grammar, arguments related to those of Cresswell (Reference Cresswell1973) for structured meanings, building on earlier arguments by Carnap for his notion of “intensional isomorphism” and David Lewis’s proposals for structured meanings in Lewis (Reference Lewis, Davidson, Harman and Partee1970). There were two big changes from Montague’s intensional logic. One of the most radical was the abandonment of the type theory, so that a function could apply to an argument ‘of the same type’; this was motivated by appeal to an important pattern of behavior of predicate expressions and their nominalizations, illustrated by examples like the following from Chierchia (Reference Chierchia1982b).

Chierchia argued that the subjects of the b sentences and of (2c) should be regarded as individual-denoting nominalizations of the predicates seen in the a sentences. He showed how an elegant account could be gotten by seeing property expressions as having two ‘faces’: they can be used as predicates, as in the a sentences, but they can be nominalized when we want to be able to predicate something of those properties in turn. Cresswell (Reference Cresswell1973) had argued that the e-type domain should contain correlates of the kinds of things found in all the other semantic domains – propositions, properties, etc. Within a type theory this cannot be done uniformly without paradox. Cresswell avoided paradox by letting the mapping from other domains into the entity domain be partial. Chierchia followed Cocchiarella in doing away with the typing and moving to a property theory instead, with basically just two types – individuals and relations. This is not the place to try to fill in the details, but these issues were central in his dissertation and continued to recur in Gennaro’s work over the next years, resulting in some of his most well-known early contributions to the field.

By the end of the Spring of 1981, it was known that Emmon and I would be away at the Max Planck Institute in Nijmegen during 1982–83, and since I would be his dissertation chair and Emmon would be on the committee, it seemed that it might be difficult for Gennaro to write his dissertation that year, his fourth year in our program, as would be normal. And since he had already completed his Generals Papers, he strongly considered trying to finish in just three years, something that had been done only once in our Ph.D. program, by Greg Carlson. But I have a copy of a letter I wrote to him in June to Italy, where he and Isa had returned for the summer, in which I report that the faculty strongly advised that he take advantage of being a Ph.D. student for the full four years for which financial support was more or less guaranteed, with the suggestion that a fifth year of support could very well be approved if necessary. (He agreed, and in the end he did finish in four years, defending his dissertation in August 1983 shortly after our return, filing it in November and getting his degree in February 1984.)

Our second Sloan grant, 1980–85, had added the Psychology and Philosophy Departments to the collaboration with Linguistics and COINS; Psychology had become a strong partner through Lyn Frazier’s collaborations, and we already had strong connections with Philosophy through semantics and philosophy of language. That second Sloan grant gave us funds to bring in three postdocs each year, aiming for people whose research spanned two or more of our fields. My own interests continued to most strongly include philosophy, as did Gennaro’s. And by lucky coincidence, Hans Kamp had just introduced me to the work of Ray Turner, a theoretical computer scientist at the University of Essex, whose first Ph.D. had been in mathematics and whose second Ph.D. was in philosophy with Kamp, with a dissertation on how to use finite approximations to model the semantics of counterfactual conditionals without possible worlds. Turner’s main research interests were the mathematical foundations of denotational semantics for programming languages and in varieties of logics. I got in contact with Turner, and told him I would love to have a chance to learn enough algebra and model theory sometime to be able to understand his dissertation, since I was quite excited about what it seemed to be about. I asked him if he would like to apply for one of our postdocs. He did, and his work was deemed of interest to the computer scientists, the philosophers, and the linguists, and we got him here for a year in 1981–82, Gennaro’s third year.

So in 1981–82 Ray Turner was here for the year, and Gennaro had finished his Generals Papers, had some publications in progress, and was starting to plan his dissertation. In both 1980–81 and 1981–82 Gennaro was my Research Assistant on the Sloan grant, with his own research as his main task. I happened to have a “Conti Fellowship” in 1981–82 from the Graduate Dean, which reduced my teaching load and left me more time for graduate teaching and research. So with Gennaro’s help, Ray and I organized an informal “Monday evening model theory seminar” within the Sloan-supported Cognitive Science program, with a couple dozen participants from Linguistics, Philosophy, COINS, Mathematics, and Psychology – faculty, graduate students, postdocs, and visitors, including our graduate students Gennaro, Mats Rooth, Craige Roberts, Nirit Kadmon; Charles Clifton from Psychology, Frank Wattenberg from Mathematics, Michael Arbib from COINS (he and I were co-directors of our Cognitive Science program in those years), Michael Jubien and Ed Gettier from Philosophy, and several Linguistics faculty. One of my goals was to learn a little bit of classical model theory, but I mostly worked on that on my own. In the seminar we focused more, with Ray’s help, on the kinds of tools that allow for finite (approximate) modeling of various sorts of infinite phenomena, which I hoped could help in the search for theories of human semantic competence, where we are up against the problem of finite brains and the plausibly non-denumerable cardinality of the sets of possible semantic values in possible-worlds semantics.

In the Fall 1981 Monday evening model theory seminar, one of the first topics Ray proceeded to teach us about was Dana Scott’s semantics for the lambda-calculus, related to Scott-Strachey semantics for programming languages (StoyReference Stoy1977), and how notions like continuity of functions could be defined on finite domains with Scott’s techniques. When Ray gave his first lecture, a couple of our linguistics graduate students told me afterwards that they hadn’t understood a word. One problem was that Ray was not accustomed to lecturing to people who knew as little mathematics as many in the audience did; another was that he had a constitutional aversion to giving examples.⁴ So since I had some freed-up time that semester and knew that the best way to learn something is to teach it, I took it upon myself to spend the week before the next Monday to prepare to retell what Ray had just taught us, with examples; and I had a chance to check it all with Ray in advance. We went through about half the semester that way, with it often taking me two or three hours to retell what Ray had told in one. Later our mathematician colleague Frank Wattenberg explained the basics of non-standard calculus with infinitesimals, and showed how it has a similar motivation, though different techniques.

By the end of the Fall semester of 1981, Gennaro’s dissertation prospectus had been approved, with the title “Syntax and semantics of infinitives and gerunds in English and Italian,” and a committee consisting of me as chair, Emmon Bach, Roger Higgins, and Ed Gettier from Philosophy. In the Spring of 1982 the dissertation was in progress, and Gennaro and Ray also audited the Tense and Aspect seminar that Emmon and I taught as a cross-listed Linguistics/Philosophy seminar. Students in the seminar for credit included Nirit Kadmon, Dorit Abusch, Charlie Jones, Bipin Indurkhya (from Computer Science), and David Lebeaux. Auditors besides Gennaro and Ray included Lyn Frazier, Mats Rooth, Murray Kiteley, Toshi Nishigauchi, and Dan Finer. Ray gave a talk on counterfactuals in that seminar, based on his dissertation. He related possible-worlds semantics approaches like Stalnaker’s and Lewis’s to metalinguistic approaches like Goodman’s and Rescher’s, and wanted to find a way to make use of the best features of each; one of his main ideas concerned ways to approach the idea of comparative similarity between “worlds” via successively refined approximations.

And the Monday night model theory seminar continued through Spring 1982, with Ray and Gennaro both playing an active role. One topic for the model theory seminar was property theory and what it could be good for in linguistic semantics. Gennaro made some presentations in which he introduced the mathematicians to the problems posed for Montague’s type theory by the semantics of nominalizations. He presented the basic ideas of property theory, and explained Cocchiarella’s approach and how he was using it to solve the problems of nominalization in his dissertation work. Michael Jubien of the Philosophy Department was working on his own approach to property theory (JubienReference Jubien, Chierchia, Partee and Turner1989), and he gave some presentations about that. We also had more discussion of Barwise and Perry’s newly developed Situation Semantics, particularly the role that properties played in that approach.

Ray and Gennaro got well acquainted that year, and Ray was one of the few people around who could discuss Cocchiarella’s work with Gennaro; Ray had some objections to Cocchiarella’s approach. Gennaro and Ray had many discussions of Scott’s type-free semantics for the lambda-calculus and its potential application to natural language syntax and semantics. Gennaro mentions that and some other alternatives to Cochiarella’s system in his papers and dissertation, but stayed with his IL* based on Cochiarella’s HST* for the purposes of the dissertation, and the two of them didn’t start to work seriously together on alternatives until after the dissertation had been completed.

One big event of the Spring 1982 semester was a Sloan-grant-funded semantics conference that I organized featuring Hans Kamp presenting his new Discourse Representation Theory (Kamp Reference Kamp, Groenendijk, Janssen, Stokhof, Groenendijk, Janssen, Stokhof, Portner and Partee1981) and Barwise and Perry presenting their new Situation Semantics. The official title was “Propositional Attitudes, Situation Semantics, and Mental Models,” but I might as well have called it “Discourse Representation Theory and Situation Semantics,” since giving those two new approaches a critical hearing was the main purpose of the conference.⁵ Other participants from outside the Five College community (UMass Amherst plus Amherst, Hampshire, Mt. Holyoke, and Smith Colleges) included Manfred Bierwisch, Tyler Burge, Ewan Klein, Michael Dummett, Bas van Fraassen, David Lewis, David Dowty, Aravind Joshi, Dana Scott, Sylvain Bromberger, Joseph Almog, Richmond Thomason, Philip Johnson-Laird, W. V. O. Quine, Daniel Dennett, William Lycan, Jerry Fodor, Janet Fodor, and Robert Stalnaker. One nice fringe benefit of those Sloan grant conferences for the hosting university was how they gave students like Gennaro a chance to hear and meet such a stellar array of scholars.

Gennaro’s research was well underway by the time Emmon and I left for the Netherlands in the summer of 1982. And Gennaro and Isa’s first son Gabriele was born in June 1982 – something else to keep them both busy. Gennaro was supported by another semester of Sloan grant Research Assistantship in Fall 1982, and a Teaching Assistantship in Spring 1983. We corresponded and he sent me drafts of chapters, and he even managed to come visit us in Nijmegen from November 22 to December 4. In addition to intense discussion of his dissertation, he came with us to semantics seminars in Amsterdam and Groningen (which we attended regularly all year), and got acquainted with many of the scholars who made the Dutch semantics scene so lively and productive. We went together to Groningen to an all-day workshop on interrogative quantifiers November 26, and on December 2 Gennaro was one of two invited speakers (the other was Henk Zeevat) at the regular Groningen semantics seminar, where he presented his revision of Montague’s Intensional Logic. In the Amsterdam seminar we heard Johan van Benthem one week, and Fred Landman the next week. And around the New Year our family (including my three sons and Emmon’s son and daughter) visited him and four generations of his family in Rome, where they graciously found room for all seven of us to stay in one of the family apartments, and where we played some wonderful Italian card game together on New Year’s Eve.⁶

During the spring semester Gennaro mailed chapters from Amherst to Emmon and me in Nijmegen and we continued to correspond back and forth. He accomplished a great amount of what he set out to do; he only regretted not being able to include as much about Italian in the dissertation as he had hoped to. The title he had proposed in his prospectus had explicitly mentioned “English and Italian”; since he couldn’t include very much about Italian, but did have some interesting sections on Italian and didn’t want it to seem to be just about English, he changed the title to “Topics in the syntax and semantics of gerunds and infinitives.”

Back in Amherst, Gennaro was having a busy Spring 1983 semester, teaching an introductory linguistics course, applying for jobs, and finishing his dissertation. He gave a talk at WCCFL II in February on the theory of control that he worked out in Chapter 4 of the dissertation, published as Chierchia (Reference Chierchia, Barlow, Flickinger and Wescoat1983b).

Since we communicated by writing letters to each other the year I was in Nijmegen, there are records of some of our conversations. A letter from Gennaro June 1, 1983 mentions getting good comments from Hans Kamp, which makes me remember that we had succeeded in getting Hans Kamp to join our Philosophy Department starting in Fall 1982, although he stayed only two years before moving to the University of Texas, so we were there together only one year, 1983–84. But he was there long enough for a number of our students to benefit from his teaching and advising, including Gennaro and especially Nirit Kadmon and Craige Roberts.

When Gennaro wrote from Amherst on June 1, he sent a draft of Chapter 5, the fourth complete chapter draft that I had received. Chapter 1, “Theories of Properties and Natural Language Semantics,” laid out his preferred logic IL* based on Cocchiarella’s HST*, with motivations and discussion, and comparison with four other approaches including Cresswell’s, Parsons’s ‘floating types approach,’ and Scott’s semantics for the untyped λ-calculus. Chapter 2, “Locally Configurational Grammar,” spelled out his approach to syntax, the syntax-semantics map, sortal distinctions to replace the missing “types,” and a substantial theory of features and morphological operations. Chapter 4, “Control and Semantic Structure,” gave his theory of control as semantic entailment and an account of the differences among obligatory control, “semi-obligatory control,” and prominence control, plus a novel account of control vs. raising. Chapter 5, “Infinitives, Gerunds, and Anaphora,” included accounts of PRO_ARB, strict and sloppy anaphora, reflexives, and control. All that remained to draft was Chapter 3, on infinitives and gerunds in English and Italian, and in the end that chapter was mostly about English.

An excerpt from the accompanying letter:

During the summer Gennaro was working intensely on finishing his dissertation. He had gotten a job at Brown to start in the fall, but getting a proper visa was a problem involving a painful amount of uncertainty and bureaucratic hassle, since his Fulbright scholarship had come with a stipulation that he return to his home country for two years before reapplying for a US visa. Isa and baby Gabriele had gone back to Italy for the summer; Isa was not totally overjoyed with the decision to stay in the US after Gennaro completed his Ph.D., and took the summer to be back in Italy while she could. Gennaro reports that he was working intensely but it was lonely, and the uncertainties and difficulties about whether he was going to be able to get the visa added to his preoccupations.

On June 20, he sent a complete draft of Chapter 3. Chapter 3, “Distributional Properties of Predicative Expressions and the Semantics of Nominalization,” contains a substantial theory of infinitivals and gerunds as a species of nominalization, arguments about the constituency of infinitives and gerunds, and a typology of clausal constructions and the notion of finiteness. Instead of a complete treatment of Italian infinitivals and gerunds, he included a substantial case study of lexical subjects in Italian infinitives, contrasting his semantic account with Rizzi’s syntactic account. Early in the chapter he presents his account of Chomsky’s so-called “PRO theorem”: the “PRO theorem” had purported to derive the fact that PRO must be ungoverned, but Bresnan had shown that since one of the premises on which it was based was false, so it actually ended up being a stipulation. One of Chierchia’s basic claims is that a finite VP like runs denotes a propositional function, while to run denotes an individual, the result of nominalizing the propositional function. Since only a propositional function can take an argument, it follows that *John to run as a sentence is semantically ill-formed. And conversely, to run but not runs can be an argument of tries (tries to run vs. *tries runs). Using such basic aspects of his proposed semantics, Gennaro offered a semantic explanation of the fact that verbs may take both finite (that S) and non-finite (for NP to VP) sentential complements, but only non-finite VP complements, never finite ones (*tries runs). His account went farther in explaining a range of facts about the distribution of “PRO” and about control in general on the basis of general semantic principles plus some plausible assumptions about English syntax and morphology. (Needless to say, it was not the last word – his later work on attitudes and de se anaphora (Chierchia Reference Chierchia, Bartsch, van Benthem and van Emde Boas1989) was one of the subsequent major advances in work on control and anaphora.)

On July 13, shortly before our return, I sent Gennaro several pages of comments and questions, also some comments from Ray Turner, who had visited for a few days the week before and read Chapter 1, and had some general observations concerning the relations among Scott’s techniques and Cocchiarella’s property theory. His advice was that Gennaro might say that probably all the issues he’s concerned with are independent of the choice of how you avoid paradox while allowing self-application of properties; Cocchiarella’s “homogeneous stratification technique” is one way, Scott’s construction is another.

We returned a little later that July. He and I had many meetings in August, and he successfully defended his dissertation August 26. Then he went to Brown to start his first job. I still have a note from him from September 8, 1983, written in haste and beginning with “I got my visa!!” So he could be legal for the rest of that year, and “even can get paid.” At Brown he had Polly Jacobson as a valued colleague, and they soon were at work designing joint research projects at the intersection of their interests.

The specifically linguistic interest of Gennaro’s dissertation lies in the rich payoff that his innovations in the treatment of model-theoretic semantics produced for the description of such central linguistic concerns as control, thematic relations, and predication. He also took a major step forward in bringing empirical linguistic argumentation to bear on the choice among alternatives to Montague’s type theory, arguing that a second-order theory offers greater explanatory power for natural language semantics than either a first-order theory or a full type theory like Montague’s. One of many lasting contributions from the dissertation is his contention that the only semantic types for which there is a full range of anaphora and quantification in natural language – i.e., for which there are something like real variables and variable-binding phenomena – are first-order or second-order: entities or properties. This interesting generalization, which has held up well over the decades, has had important consequences of several sorts. An example concerning the analysis of particular phenomena is the treatment of degree modification, where his hypothesis suggests that “degrees” are to be seen as a subclass of entities. A more general consequence comes in language acquisition, where he predicts a big difference between the acquisition of first- or second-order vocabulary, where the whole logical space is available for potential meanings, and higher-order vocabulary, which is generally restricted to a small number of lexical items. There his prediction is that the only higher-order lexical items that languages exploit come from a small set of potentially universal and innate possible meanings, so that the learner does not have to make arbitrary discriminations within higher-order meaning spaces.

That work required a combined depth of understanding and creative intellectual power in all three areas of linguistics, logic, and philosophy. Already then he combined in one person to a remarkable degree the interests and abilities of the ideal linguist and the ideal philosopher–logician, including being highly independent and undogmatic. With his willingness to consider and draw upon work from highly diverse frameworks and his ability to integrate and modify, he was able to have a strong positive influence in helping to see beyond battles between “schools” in working toward larger shared goals, as well as being a major contributor to research of importance for both linguistics and philosophy. In the dissertation, his treatment of non-overt subjects and control structures was novel and elegant: without positing null subjects at either a syntactic or semantic level of representation, he treated control relations as semantic entailments of predicates, ultimately as restrictions on admissible model structures. The formal innovations in the treatment included the beginnings of a formal theory of thematic relations in which they are viewed as higher-order relations between n-place predicates and individuals, and play a role in the formulation of constraints on the possible interpretations of natural language predicates. (In a sense part of this work could be viewed as the beginnings of a linguistically motivated theory of “meaning postulates,” or of constraints on lexical meaning and lexical entailments.) Using this “directly semantic” approach, he achieved an account of Visser’s and Bach’s generalizations that built on but improved upon both Bresnan’s and Bach’s treatments, and provided an integrated account of the distribution of intensional argument positions, non-thematic argument positions, and the distinction between Equi and Raising.

2. The assistant professor years: 1983–1988

Gennaro’s first position was at Brown University, starting in Fall 1983, where Polly Jacobson was his closest colleague. Gennaro was busy revising his dissertation to submit in late fall, and settling in to teaching.

In November Gennaro and Mats Rooth, who was still at UMass, presented a joint paper, “Configurational notions in Discourse Representation Theory” at NELS 14 at UMass,⁷ published in 1984 (Chierchia and Rooth Reference Chierchia, Rooth, Jones and Sells1984). This was an interesting and carefully argued paper that challenged a fundamental anti-compositionality claim in Kamp’s seminal DRT paper (Kamp Reference Kamp, Groenendijk, Janssen, Stokhof, Groenendijk, Janssen, Stokhof, Portner and Partee1981). It was well-known that in a theory that adheres to the principles of Montague (Reference Montague and Thomason1970), the use of an intermediate logical representation language (such as Montague’s IL) must strictly be dispensable: there must be a direct compositional model-theoretic interpretation, obtainable as the composition of the translation function from the object language to the logical representation language and the interpretation function that gives a model-theoretic interpretation of the logical representation language. Kamp claimed that his intermediate representation language, the language of “DRS boxes,” was not dispensable, because a geometrical notion of “accessibility” was said to be needed for characterizing possible anaphoric relations. As Gennaro and Mats write, if this is the best possible theory of anaphora, then the internal geometry of DRSs would have to be regarded as a crucial and indispensable component of natural language semantics, contrary to Montague’s general view.

What they show in their paper is that accessibility can be dispensed with in Kamp’s theory, and that binding possibilities (i.e. bound anaphora, the only anaphora treated in Kamp (Reference Kamp, Groenendijk, Janssen, Stokhof, Groenendijk, Janssen, Stokhof, Portner and Partee1981) or in Montague’s work) follow from the properties of the model-theoretic interpretation – in Kamp’s case, from embedding conditions for embedding a DRS into a model. As a byproduct, they show that there is no theoretical or empirical difference between Kamp’s notion of discourse referents and the notion of variable familiar from Tarskian satisfaction-based semantics. This was an important result, and the first of several works that showed how one can relate DRT notation and interpretation to more familiar kinds of logics, especially Zeevat (Reference Zeevat1989) and Muskens (Reference Muskens, Dekker and Stokhof1994).

Gennaro and Polly soon applied for a joint NSF grant for research on syntax and semantics, which they eventually got, and which resulted among other things in their exciting joint paper about two kinds of control (Chierchia and JacobsonReference Chierchia, Jacobson, Berman, Choe and McDonough1986), which I return to below.

In Spring 1984 he was one of just three or four invited speakers at a Tarski Memorial Symposium at Ohio State University in March. His talk, based on his dissertation, was “A Fregean approach to predication in English,” an early version of Chierchia (Reference Chierchia1985). Another part of his dissertation was developed into a talk on anaphoric properties of infinitives and gerunds at WCCFL 3 in April, published as Chierchia (Reference Chierchia, Cobler, MacKaye and Wescoat1984b).

In the meantime, the System Development Foundation had come into existence as a deliberately short-lived entity, created as a non-profit entity in 1969 when the non-profit System Development Corporation became a for-profit corporation, and functioning as a grant-giving foundation from 1980, when SDC was sold to Burroughs, until it had disbursed all of its funds in 1988.⁸ By far the biggest grant from SDF was the $15 million grant in 1983 to Stanford University together with SRI International and Xerox PARC to found CSLI, the Center for the Study of Language and Information, a research center involving linguists, philosophers, logicians, mathematicians, and later also psychologists. I was asked to serve on the Advisory Board, and at one of the first meetings I attended, in November 1983, it was strongly hinted that I should get acquainted with the chief program officer, Charlie Smith, and put in an application for a grant of our own for semantic research at UMass, ideally in some sort of collaborative arrangement with CSLI, and perhaps getting and using some of the same kind of “Dandelion” computers that CSLI had just acquired. (The Dandelions were among the first machines with ‘windows’ and ‘mice’.⁹ Exciting and powerful but reportedly very temperamental.) So I invited Charlie Smith to come pay a visit to UMass in February 1984. I took him around to meet colleagues in linguistics, philosophy, and computer science and to give him an overview of the kinds of research we wanted to pursue as a follow-up to our Sloan grants. Charlie Smith was a strange character; he managed to look as though he was sleeping through a lot of the brief presentations we were giving for him, but would occasionally ask some sharp question or say something almost hostile; he seemed to enjoy making people uncomfortable. When he was meeting with a group from the Philosophy department, Michael Jubien described the work he was doing on property theory, and his hopes to get a book written on it. Charlie Smith asked how long it would take to write the book. Michael estimated that with his teaching load it would probably take a few years. Charlie asked how much money it would take to buy out enough teaching time to get it done in a year. Michael, surprised, managed to make an estimate (I no longer remember the numbers involved; it was basically a semester’s teaching time plus a summer), and Charlie basically offered him a grant for that much on the spot, telling him to please write up the proposal and send it in and get to work writing. By the end of his visit he also encouraged me to submit a proposal; those were the only two he encouraged from UMass. I mention all this because what I was most anxious to do with the grant was to get Ray Turner back to UMass again and try to continue promoting interaction between him and Gennaro on property theory and related topics.

One sticking point was that Charlie Smith seemed to very strongly suggest that if I wanted a grant, it should include a request for some Dandelion computers. My colleagues and I were not so sure. So with helpful advice from my colleagues and close consultation with Ray Turner, I wrote up a proposal for a 3-year grant asking for funds for several Research Assistants, a visiting faculty position for Ray for three years and a post-doc or two, some summer salary, and a visit out to CSLI in the summer of 1984 accompanied by Ray, Gennaro, Emmon, three linguistics Research Assistants, and a Computer Science Research Assistant, for the purpose of acquaintance by immersion with the research going on at CSLI and with the computer environment there, promising that I would submit a separate equipment grant after I better understood our computational needs and what a Dandelion environment had to offer.

I got the grant, with most of what I wanted – tiny in comparison to CSLI’s funding, but very large for me; I had asked for a million dollars and got about $750,000, and the project did begin with just such a visit to CSLI that summer, from late June to early August, with Emmon (only for two weeks, after his trip to China for the lecture series which led to his book (Bach Reference Bach1989)), Ray, and Gennaro, and Computer Science student Bipin Indurkhya, and three linguistics graduate students, Craige Robers, Karina Wilkinson, and Nina Dabek. A lot of quick planning was involved; the students and I would arrive just in time for a COLING conference that included some tutorials the week before, starting June 25, and the linguistics students decided to take advantage of Brian Smith’s LISP course and Bipin and I took Fernando Pereira’s very wonderful PROLOG course – both courses designed specifically for linguistic and computational linguistic applications. (The PROLOG course turned out to use as examples the parsing and interpretation of Montague grammar constructions augmented by Robin Cooper’s quantifier storage mechanisms, so with that and Pereira’s great clarity, I was able to follow well and could spend some of the homework time on the local computers experimenting with things like possible quantifier scopes for NPs embedded in PPs inside other NPs.) Ray and Gennaro arrived July 2; Emmon came later for just two weeks. I had some good email conversations in advance with Jose Meseguer about formal properties of PROLOG, unification, and the relation between algebraic semantics and model-theoretic semantics. Email was brand new to me then; Stanford and SRI were on the ARPAnet and UMass was on cs-net, and at UMass I composed messages via a terminal connecting me to a VAX in the COINS department, with the EMACS editing program as a primitive “word processor” (designed for editing programs, not text; it was great for making sure that parentheses matched), and then over in the COINS department I could print out long folded pages of files of saved mail. One of Jose Meseguer’s replies to a big pile of questions in April begins, “Barbara, Isn’t it neat, the way that we can communicate with each other using these electronic networks?” And as primitive as they look now, it really was.

We rented two houses and rented a car; CSLI helped with some of our expenses, and gave us a computer terminal for one of the houses and shared offices at CSLI. The graduate students explored the computational environment, learned LISP and/or PROLOG, got acquainted with some of the work that was going on at CSLI on anaphora and other topics, and worked on their current research projects. Ray and Bipin especially helped us figure out that there would be no way we could have Dandelions in the Linguistics Department without a full-time expert to take care of them. We did see some impressive Dandelion demos at Xerox PARC, where Chris Halvorsen showed us how syntactic analyses of LFG derivations in Bresnan and Kaplan’s system could be displayed – you could see a whole tree on a screen, but “click” on a node of the tree and you would get a window showing the full feature analysis at that node, and I think the nesting of displays of layers of structure was recursive. Really nice, and potentially very useful. But the machines needed constant expert maintenance. So when I did later put in an equipment request, it was for some much more mundane desktop computers that would be valuable for the department’s phonological and phonetic research. The original grant included a simple desktop computer for me – that was my first.

Ray, Gennaro, and Bipin lived in one of the houses and the rest of us lived in another. We frequently invited two or three Stanford (or visiting) colleagues over to dinners with all of us at the bigger house; on those evenings one or two or a team of us would cook for the whole gang; our guests included Joan Bresnan and Marilyn Ford and their daughter Alexandra; Fernando Pereira and Jose Meseguer; Ewan Klein and Mary Tait; Robin Cooper and Elisabet Engdahl; Joyce Friedman, Jane Robinson, and Eva Hajičová; Peter Sells and Gerald Gazdar; and Jon Barwise and Brian Smith.

All of that interaction and talking with people was good but much of it was slightly superficial. The main work involved a few key projects. I had a chance to work a bit with Hans Kamp on our project on prototype theory and compositionality, which took years of infrequent interaction to finish (Kamp and ParteeReference Kamp and Partee1995). And I made some breakthroughs in my work on type shifting, thanks in part to very valuable feedback, especially from Joe Goguen and Jose Meseguer,¹⁰ when I presented it at CSLI (Partee Reference Partee, Berman, Choe and McDonough1986a, Reference Partee, Groenendijk, Janssen, Stokhof, Portner and Partee1986b). Ray and Gennaro were interacting intensely all the time they were there, and they also had some meetings with the mathematician Solomon Feferman, a student of Tarski’s and a collaborator with Montague on a never-completed book on topics in the foundations of set theory. Feferman’s work, like Ray’s, concerns the relation between logic and computation, and he has done a lot of work on constructive mathematics, and “proofs as programs, programs as proofs” (Feferman Reference Feferman1985).

Ray and Gennaro gave a well-received CSLI colloquium talk in July, “Property theory and the foundations of semantics.” Their collaboration had the nature of a long-running argument that became steadily richer.¹¹ Ray thought a non-constructive consistency proof that didn’t show you how to build a model was not good enough for a logic that was supposed to be an underpinning of natural language semantics; that was his complaint against Cocchiarella’s system, if I understand it correctly. Ray and Gennaro both understood that there were many challenges to developing a theory that had the properties they considered necessary for a successful account of intensionality and nominalizations, and many different choices that could be made at various choice points. The lasting value of much of their work, I believe, lies in laying out so clearly what the desiderata are, what the theoretical and descriptive choices are, what challenges face each choice, which choices may be independent of which others and which are crucially interconnected, and what kinds of motivations may lead researchers to prefer one line of development over another. And in the joint paper which grew out of their interaction (Chierchia and Turner Reference Chierchia and Turner1988), they arrived at a fully worked out account of property theory which they claimed could support the semantics of natural language, driven by Gennaro’s guiding intuition that properties are the semantic counterparts of natural language predicative expressions. On their joint theory, English is a multi-sorted first-order language, rather than the second-order language argued for in Chierchia’s earlier Cocchiarella-inspired work; they show how everything captured with the help of Montague’s type theory can be captured in their system as well, sometimes even better, as in their variant of the treatment of generalized conjunction and disjunction of Partee and Rooth (Reference Partee, Rooth, Bäuerle, Schwarze, von Stechow and Partee1983). And they illustrate the kinds of natural language phenomena for which property theory seems most necessary with examples that involve reference to and quantification over properties, property anaphora, predicates like is fun which can take both individual-denoting DPs and property-denoting infinitival expressions as subject.

Gennaro and Ray continued their collaborative work into the 1986–87 year, partly through visits to UMass supported by the SDF grant. Ray could not stay for the three full years 1984–87 as we had hoped, because he obtained a chair at Essex that required him to spend most of his time there; good news overall, though a disappointment for my plans, and I did not learn as much about property theory and alternative logics as I had hoped; by 1987 my own energy had turned instead to collaboration with Emmon and Angelika Kratzer on our big NSF-sponsored cross-linguistic study of the typology of quantification – possibly my first-ever non-interdisciplinary grant!

But in 1984–85 the SDF project had a good first full year. Ray visited for two months in the fall and a month in the spring, the mathematician Bill Marsh from Hampshire College was able to spend two months of the fall semester with us, and Fred Landman came from Amsterdam for the full fall semester. Hans Kamp visited in September, and Gennaro made several short visits up from Brown. We resumed an informal evening seminar on Model Theory and Foundations of Semantics, with linguists, philosophers, mathematicians, and computer scientists. The principal participants included Emmon and me, Fred Landman, and Ray Turner, and Gennaro when he could come. Central issues included the semantics of variables (a topic that Fred and I both had a long-standing interest in, and on which Fred was doing important original work (Landman Reference Landman and Halpern1986)), type theory, and theories of partial information (Fred and Ray).¹²

In December of 1984 Gennaro and Polly Jacobson were awarded their NSF grant for joint research on syntax and semantics. Plans were then made for Gennaro to spend the first half of summer 1985 working with Polly at Brown on their grant and the second half working with Ray and me here on mine. Gennaro was in fact at Brown just from 1983 to 1985, and then accepted an offer from Cornell. But Gennaro and Polly continued their collaborative project, which ran through 1987 and resulted in their valuable paper (Chierchia and Jacobson Reference Chierchia, Jacobson, Berman, Choe and McDonough1986). Up until then, semanticists had been debating three different approaches to controlled subjects of infinitivals like to leave in John tried to leave or Mary persuaded John to leave: (i) many followed traditional transformational grammar analyses and posited a null PRO subject in the syntax, interpreted as a bound variable in the semantics; (ii) some followed Montague in treating infinitivals simply as to plus a VP, which required finding a way to derive passive VPs like to be fed properly, for which there were various proposals like my early “Derived VP rule” (Partee Reference Partee and Partee1973) or the meta-rules of GPSG (GazdarReference Gazdar, Jacobson and Pullum1982, Gazdar et al. Reference Gazdar, Klein, Pullum and Sag1985); Chierchia in his dissertation and related articles argued explicitly for such an approach, with no null subject in either the syntax or the semantics. And (iii) some argued for having no null subject in the syntax, but introducing a (bound variable) subject in the semantics (LFG, some Montague grammar analyses such as Bach and Partee (Reference Bach, Partee, Kreiman and Ojeda1980)). What was new in Chierchia and Jacobson’s work was the marshaling of evidence to show that the best analysis was not the same in all cases: that “local” control works as Chierchia proposed, via semantic entailments that determine what, if anything, the property denoted by an infinitival VP is to be predicated of, whereas “long distance” control involves determining the interpretation of a null PRO subject. Their work enriched the control literature by providing a rich new set of diagnostics for discriminating the different kinds of control.

So in the summer of 1985, as planned, Gennaro spent the first part of the summer working with Polly at Brown on that grant project, and the last part at UMass on the SDF project.

During the summer he made two trips to give invited talks. The first was “Aspects of a categorial theory of Binding” at a Categorial Grammar conference in Tucson in June organized by Emmon Bach, Dick Oehrle, and Deirdre Wheeler (Chierchia Reference Chierchia, Oehrle, Bach and Wheeler1988a), a new topic not at all connected with his dissertation. There he argued against the purely semantic approach to binding argued for in Bach and Partee (Reference Bach, Partee, Kreiman and Ojeda1980) and in favor of a theory combining the Bach and Partee semantics with a syntactic counterpart, involving coindexing on a categorial structure and using a theory of predication inspired by work of Edwin Williams (Williams Reference Williams1980). He made use of Gazdar’s approach to binding, arguing that all binding phenomena should be treated in terms of syntactic binding features analogous to Gazdar’s slashes. This gave him a way to handle case-marking phenomena that purely semantic approaches could not; and he resolved problems that Bach and Partee had left open about the status of indices or the individuation of variables. He also drew interesting consequences for across-the-board phenomena, a topic he had not ventured into earlier, and for control, which he had already studied deeply, and where he was able to maintain his position that (local) control does not involve any covert subject, and where control could be viewed as an instance of subject-predicate agreement, a single assumption that offered a great simplification over the then-current GB assumptions about control.

And in July 1985, Gennaro, Ray Turner, and I spent two weeks at CSLI. Gennaro and Ray gave an invited and well-received presentation of their ongoing research at a CSLI-hosted meeting of the Association for Symbolic Logic, “Semantics and Property Theory,” an early version of the work they published together in 1988. I also gave an invited talk there, “Syntactic Categories and Semantic Types” on types and type-shifting. The meeting was preceded by an interesting set of one-week courses, of which I particularly remember Yiannis Moschovakis’s course on the theory of algorithms, in which he reported on his efforts to find a suitably ‘intensional’ notion of “function,” more discriminating than the standard set-theoretic extensional notion, but not as excessively fine-grained as some notions of algorithm. There was a great density of logicians with an interest in language, in keeping with the mission of CSLI; other short courses were offered by Barwise, Thomason, Plotkin, Meseguer, Moss, and Maarten van Emden, and it was a very lively atmosphere.

Then in August we had an informal but intensive research gathering of participants in my SDF grant in Amherst, including Partee, Bach, Chierchia, Turner, Hans Kamp, seven research assistants, and other informal participants. Topics included Ray and Gennaro’s work, the work that Craige Roberts and Nirit Kadmon had begun with Hans Kamp while he was on our faculty concerning modal subordination (Craige) and distributivity (Nirit), and I recall a great deal of discussion of alternative approaches to the semantics of plurality and distributivity.

At the end of the summer of 1985, Gennaro and his family moved to Ithaca, where Gennaro began teaching at Cornell. That fall he and Polly Jacobson presented their joint paper at the NELS conference at McGill in November, and in January he gave an invited talk on “The Logic of Control” at the University of Geneva.

During 1985–86 Gennaro was at Cornell, but made several visits to UMass. Ray Turner was here part of the time in 1985–86 as well.

Our main collective SDF-project event in 1985–86 was a conference in March 1986 that Gennaro, Ray, and I organized on Properties, Types, and Meaning, bringing logicians, philosophers, and linguists together to address some of the foundational issues central to our grant project and linguistic problems involving them. The conference took place at Hampshire College through the kindness of our Hampshire colleagues, especially Jay Garfield, and many of the participants were housed in a quaint dormitory there which also held our meeting room. The conference is described in the two-volume collection based on it that the three of us edited (Chierchia et al. Reference Chierchia, Partee and Turner1989b). The presenters were Ray, Gennaro, Peter Aczel, Rich Thomason, Nick Asher and Hans Kamp, Michael Jubien, George Bealer, Johan van Benthem, Jeroen Groenendijk and Martin Stokhof, David Dowty, Greg Carlson, Lenhart Schubert and Jeff Pelletier, Henk Zeevat, Solomon Feferman, Ewan Klein, and Gordon Plotkin. There was a good group of invited discussants as well, including Emmon, Janet Fodor, Erhard Hinrichs, Angelika, Fred Landman, Richard Larson, Godehard Link, Chris Menzel, Uwe Mönnich, and Carl Pollard. Gennaro’s paper in that volume argues in favor of structured meanings, on the basis of arguments concerning thematic roles and control, independent of propositional attitudes, and staying neutral concerning property theory. Ray’s paper (Turner Reference Turner, Chierchia, Partee and Turner1989) succinctly lays out the difficult problems of intensionality and nominalization, and how the problems require a theoretical framework permitting properties to be regarded as individuals, as spelled out in his joint work with Gennaro. Using work further described in Chierchia and Turner (Reference Chierchia and Turner1988), he shows a way of achieving this with a rather weak logic and property theory using “close variants of models of the λ-calculus” (Chierchia et al. Reference Chierchia, Partee, Turner, Chierchia, Partee and Turer1989a: 7) and contrasts it with other approaches, including that of Gennaro’s dissertation.

That conference was in a sense the peak of our joint collaboration. Ray and Gennaro both visited UMass some more in the following year, and there was a lot of work to do on the editing of the two volumes, most of which fell on Gennaro. By earlier agreement, I did very little work on the editing. In fact almost all of the editing work fell on Gennaro, and he was the sole author of the introduction to Volume 2. Ray helped with some specific advice on some of the technical papers in Volume 1. Gennaro read every article with care, and corresponded extensively with the authors about suggested revisions, and endured all the editorial headaches. The introductions to both volumes are important works in themselves, one solely by Gennaro and the other largely by Gennaro; they not only give very good summaries of the main ideas of the papers, but they insightfully place them in the context of the larger linguistic, logical, and philosophical issues to which the conference was dedicated. The conference and the book represented a rare confluence of ideas and approaches from logic, philosophy, foundations of computer science, and linguistics, and I believe the book has been a useful resource for anyone interested in approaches to and linguistic uses of property theory or type theory.

In the summer of 1986, Gennaro and I both gave talks (at different times) in July at the Workshop on Logical Form organized by Irene Heim, Howard Lasnik, and Bob May at the LSA Linguistic Institute at CUNY. My summer 1986 in Amherst involved working with a number of students on their research, several of them supported on the SDF grant while working on their dissertations; Craige Roberts and Arnold Chien defended their dissertations that August. Emmon and Angelika and I spent a lot of time reworking the NSF grant proposal on cross-linguistic quantification that we had submitted unsuccessfully the year before. (This time it was successful.) I don’t have any record of any visits to CSLI that summer or any summer visits from Ray or Gennaro.

But then during 1986–87 Gennaro spent much of the time in the spring semester, from January to June, at UMass as a Research Associate on the SDF project. That spring Nirit Kadmon, Jae-Woong Choe, Sandro Zucchi, Karina Wilkinson, and Mark Aronszajn (Philosophy) were all working on their dissertations – there were a record number of semantics dissertations right around then, perhaps partly because of the unusually high amount of funding and semantics research activity the grants of that period made possible. During that spring while Gennaro was here, Emmon, Angelika, and I gave our first joint seminar, on the semantics of events. Gennaro had much to contribute from his work on infinitives and gerunds, and Sandro Zucchi followed up on some of the seminar topics for his later dissertation work. Gennaro was still managing to be active back at Cornell at the same time, and in May was co-organizer with Fred Landman of a conference on “Events and Thematic Structure” at Cornell.

In April 1987 Gennaro gave a talk “Anaphora and attitudes de se” at the Amsterdam Colloquium (Chierchia Reference Chierchia, Bartsch, van Benthem and van Emde Boas1989). This paper, a new direction for Gennaro then, was another great combination of philosophical and linguistic insight, bringing badly needed clarity to the old and vexing questions of how to treat Castañeda’s “quasi-indicator” pronouns, how they relate to the understood subjects of infinitives, and how they relate to ordinary bound variable pronouns. These were puzzles that linguists of many different schools had found themselves embroiled in willy-nilly, and Gennaro is one of the few linguists (Fred Landman is another) who managed to turn these questions into an area of fruitful research rather than a pit of quicksand. David Lewis had laid a good bit of groundwork (Lewis Reference Lewis1979a), but I believe that Gennaro was the first to find a successful way to make use of Lewis’s ideas for linguistic analysis.

In the summer of 1987 Gennaro and I were both at Stanford, at the 1987 LSA Linguistic Institute. Gennaro was invited to give the basic Semantics course, and it was one of the half-dozen lectures chosen to be videotaped and offered for sale. He also “taught” in Linguistics 230, the Workshop on Generics, sponsored by Ivan Sag’s NSF grant, where many of the contributors to the long-in-gestation Generic Book (Carlson and PelletierReference Carlson and Pelletier1995) started the project with much lively discussion and debate. I met Gennaro’s first Ph.D. student Veneeta Srivastav (Veneeta Dayal) in my semantics seminar there, and she was one of the top students in my big class, already showing many of the qualities that have made her one of the leaders in the field. The heavy teaching schedule kept us very busy there, but Isa has some nice photos of a poolside afternoon when Solomon and Anita Feferman had invited Gennaro and Isa and their boys (two then, I believe – Gabriele and Tommaso) and me for lunch at the faculty club swimming pool on August 1.

By the fall of 1988, when he was up for tenure, the very nice semantics textbook that he and Sally McConnell-Ginet published in 1990 was well under way. That textbook is still an ideal choice to use with students who are not necessarily going to become formal semanticists themselves but to whom you want to convey some of the most important and beautiful ideas of contemporary pragmatics and formal semantics, and it furthermore gives a very sound foundation for any who do decide to pursue semantics or pragmatics further.

Also by the fall of 1988, Gennaro had been appointed to the editorial boards of Linguistics and Philosophy and of Linguistic Inquiry, was co-editor of the Linguistic Inquiry Squibs and Discussion Department, and had taken on the managing editorship of Kluwer’s book series Studies in Linguistics and Philosophy. His joint paper with Ray Turner was about to appear, and he had a paper in press in the proceedings of Krifka’s 1988 conference on Genericity (Chierchia Reference Chierchia and Krifka1988b).

Gennaro was by then a recognized star in semantics. His dissertation was published in the Garland Outstanding Dissertations series in 1988, the same year our co-edited conference volumes came out. His work on his own and with Ray Turner on property theory and its linguistic applications was understood to be extremely important foundational work with important consequences for both the philosophy of language and the linguistic practice of semantics. I have always regarded that work, like much of his work, as important steps in “vindicating” natural language by showing that one can make good sense of the observable syntax of natural languages if one can find the right sort of semantics (and pragmatics) to use in compositionally interpreting it. That has always been Gennaro’s strength; it’s nice to see that he has found and uses a phrase that encapsulates that vindication, “the spontaneous logicality of language.” Perhaps we can characterize Gennaro himself as manifesting in the highest degree a similar property, “the spontaneous insightfulness of the ideal philosopher-linguist.”

Acknowledgments

My thanks to Ivano Caponigro for valuable advice, feedback, and editorial assistance with this chapter. Special thanks to Kathleen Adamczyk for unearthing a copy of the 1984 play that Raymond Turner and she wrote about the property-theory debates that consumed Ray and Gennaro during that summer, which is now being published as an appendix to the present volume; and to Raymond Turner and Bipin Indurkhya for feedback on my description of that summer. All errors or lapses of memory are my own.