1. Introduction

Two topics are addressed and conflate in this paper, which I will succinctly introduce. On one hand, there is the fact that, despite being mostly used referentially, in argument positions, proper names (PNs) can also appear in non-referential, predicative contexts. An eminent case of predicate PNs is found in naming constructions (NCs), such as (1); which are formed by naming verbs (NVs), such as name, nickname¸ title, baptize, dub, denominate, etc. The complement of these constructions usually features an object DP and a predicate PN, which is not being used referentially.

(1) a. They named the boy John.

b. I nicknamed my roommate Shrek.

c. James titled his novel Ulysses.

On the other hand, constructions where there is a predicate modifying the object DP can be analyzed as cases of secondary predication, mainly resultative (2), object depictive (3), or pseudo-resultative (4) constructions (Levinson 2007). Accordingly, resultatives predicate a state of the object resulting from an event while object depictives denote a property held by the object during the event. Pseudo-resultatives, in turn, don’t modify the object, but rather an object created with it.

(2) a. Mary hammered the metal flat. = The metal is flat.

b. Mary cooked the meat black. = The meat is black.

(3) a. Mary hammered the metal hot. = The metal was hot.

b. Mary cooked the meat raw. = The meat was raw.

(4) a. Mary braided her hair tight. ≠ Mary’s hair is tight

(rather, her braid is tight).

b. Mary piled the cushions high. ≠ The cushions are high

(rather, the pile is high).

I will argue that NCs like (1) can be analyzed as resultative constructions, however, if this is the case, PNs would have to behave as result states. Consequently, not only a syntactic and eventive account is necessary to justify this perspective, but also a compositional account, able to provide PNs with the appropriate meaning to combine with NVs and render these constructions.

Additionally, PNs in NCs have been highlighted by Matushansky (2008) as evidence supporting the general notion that PNs are underlyingly predicates (Frege 1970[1892]; Geurts 1997; Borer 2005; Bach 2015; Fara 2015; among many others), containing, at least, the notion of an entity that bears a name. In contrast, others have regarded PNs as rigid designators (Kripke 1980). Along the predicativist line, the use of determiners, and particularly of articles with argument PNs in various languages is also considered to indicate that a PN is the complement of a DP. Examples in (5) and (6) show that in Māori, as well as in other languages like Northern Norwegian, there is an obligatory proprial article used for PNs that contrasts with the regular definite article. Meanwhile, (7) and (8) illustrate the use of definite articles with PNs, obligatory in Greek, and optional in Spanish, respectively.

(5) Ka whakarongo puku *(a) Ponga.

tam listen silent the-prop Ponga

‘Ponga listened in silence’. (Māori: Bauer 2003: 143)¹²

(6) *(Ho) Siri e hœr.

the-prop Siri is here

‘Siri is here’. (Northern Norwegian: Julien 2005: 176)

(7) *(O) Yanis theorite ilithios.

the-NOM Yanis-NOM consider-PASS.3SG idiot-NOM

‘Yani is considered an idiot.’ (Modern Greek:

Matushansky 2006: 287)

(8) (La) Lidia llegó tarde.

the.NOM Lidia arrive-PST.3SG late

‘Lidia arrived late.’

Extensive literature can be found dealing with proper names in argument positions for all sorts of phenomena, like examples above (5-8), both syntactically and semantically (Böer 1975; Elbourne 2005; Matushansky 2006, 2015; Bach 2015; etc.). Within the predicativist perspective, however, little work has been done on describing actual predicate uses of proper names, like NCs. Therefore, in this paper, I address this issue and show that a careful look into the syntactic and eventive structure of NCs and evidence from Spanish and other languages reveals that their complement should be interpreted as a small clause (SC) that encodes a proposition on its own. Moreover, using the tools of a Distributed Morphology (DM) approach and a minimalist analysis of argument structure, I claim that the structure that NCs display strongly suggests they should be interpreted as resultatives.

Given that a previous semantic account (Matushansky 2008) cannot explain some of the data presented here and is not compatible with the syntactic structure of NCs, this paper, importantly, also provides the necessary elements to successfully reformulate the compositionality of predicate PNs in NCs. Furthermore, it solves the puzzle that the interaction of PNs and naming verbs represents semantically for a predicativist theory while providing an answer on how to interpret PNs as states in a resultative construction.

After laying out the twofold perspective under which NCs will be analyzed, this paper is organized as follows: §2 will define the basic theoretical assumptions of the neoconstructivist approach to argument structure that I subscribe to. Then, I discuss data from Spanish and other languages that give evidence for a resultative analysis of NCs. Afterward, I propose a syntactic structure and justify its appropriateness. In §3, I turn to the semantics of proper names and the structure of NCs. By pointing out the problematic aspects of a previous approach (Matushansky 2008) regarding the interaction between these two, I offer a solution and a new compositional and eventive account. §4, in turn, provides the concluding remarks.

2. The syntax of naming constructions

Throughout this paper, I will adopt a neoconstructivist approach to argument structure within a minimalist and DM framework. Given the weight that some theoretical aspects will have further ahead, it is convenient that before diving into the actual phenomenon I take a moment to explain the basic assumptions that will inform the rest of the syntactic discussion.

2.1. A neoconstructivist perspective of argument structure

In general terms, this theoretical perspective considers that argument structure occurs in syntax. More specifically, it regards argument structure as syntactically constrained, that is, limited to the available configurations of specifiers and complements (Hale & Keyser 2002; Harley 2005). Furthermore, the DM framework that I follow (Marantz 1997, 2005; Halle & Marantz 2005; Harley 2011) considers argument structure to be pure syntactical, no longer dependent on lexical and conceptual information. To accomplish this, there are two types of components that enforce the argument structure in syntax: functional heads and roots. As such, the event of a proposition is encoded in the well-known functional head of little v, which verbalizes the structure (Chomsky 1995; Harley 2011; Mateu 2015). Following along, lexical verbs are actually roots (√), which carry the conceptual meaning, non-accessible to syntax, and merge with the v head. Beyond the verbal domain, analogous functional heads have been proposed for nouns and adjectives, so that nouns and adjectives would be, respectively, the result of a root moving into a n head, which nominalizes it, or to an a head, which adjectivizes it (Acquaviva 2008).

The particular syntactic arrangement that concerns NCs, is found in causative constructions. The theoretical argument that backs the analysis can be understood as a decomposition of the event structure taking place on syntax, also compatible with neo-Davidsonian event semantics (Parsons 1990). To exemplify, a causative event like break can be separated into two: the cause behind the breaking, and the state of being broken. As it has become standard, little v head encodes a CAUSE, but, unlike other versions (Hale & Keyser 2002) which consider the become event to be encoded in a second v head, I will take Von Stechow’s (1995) and Harley’s (2011) argument for a small clause as the structure for change of state predication.

A small clause (SC) is a structure of non-verbal predication, more exactly, a relation between a subject and a predicate not mediated by a T head. The specific internal structure of a SC that I assume stems from Stowell’s (1981), also featured in Harley’s work. In (9) we have an example of how this structure looks for an unaccusative construction: the DP the coriander is the subject of a SC, whose non-verbal predicate is a root grow, carrying only conceptual meaning. Note that there is no functional head mediating the predication (cf. Bowers 1993), following Stowell’s idea that the subject-predicate relation can be established structurally. In (9) the root then moves out of the SC into the verbalizing head v; and later, the full sentence would be The coriander grows.

The relevance of this SC lies also in how it can account for causative constructions, where there is a change of state event. For example, (10) shows the structure of the causative version of (9): Kyle grows the coriander. As mentioned, the SC encodes the resulting state of being grown, but the DP the coriander remains in its place, and the external argument Kyle is merged as a subject of the main clause, which encodes the event of causing the coriander to be grown.

Now, as I stated, the key constructions are resultatives. These have a similar underlying structure as causatives like (10), with the exception that the predicate of the SC is not verbalized into the v head. Instead, as can be seen in (11), resultative constructions merge another root in a √P that has a SC complement, then, it moves into v, leaving the elements of the resultative SC in its place (Marantz 2005; Harley 2011, 2012).

Resultatives, thus, can be understood as a change of state constructions where a causative event has an effect over the verb’s object, yielding a resulting state that is predicated through a SC structure. One important detail to keep in mind is that verbs featured in these constructions may not require the presence of a resulting state, but it can still be conceptually entailed, i.e., the verb paint doesn’t convey the specific color that something becomes after painting it, as in We painted the wall, yet it can be expressed through a SC as in We painted [_SCthe wall yellow].

In the following, I’ll argue in favor of a resultative analysis for naming constructions, which follows (11), but also considers how would names fit into this predicative structure according to a preexisting theory of the meaning of proper names (Matushansky 2008).

2.2. A resultative account of NCs

Naming verbs, such as name, nickname, title, dub, baptize, denominate, etc., may be found in constructions like (1), where the name that is given appears as a predicate of its recipient. As we saw above, this type of predication can be analyzed as a SC. However, this idea implies that NCs exhibit both a causative naming event, and a change of state; otherwise, a SC would be pointless. Furthermore, the fact that naming someone or something does not involve any physical change or obvious change of state, and the implication that names should be somehow interpreted as states could be discouraging in pursuing the resultative hypothesis; nevertheless, there are good reasons to consider it.

First, let’s take a look at Spanish naming verbs, which will turn out to be highly relevant in the next section. The pair of sentences in (12) show an expected behavior for resultatives: as shown in (12b) a NV may appear with only a direct object complement, meaning that there is no indication as to what name was given, therefore, a change of state is not explicitly stated.

(12) a. Eva nombró a su gato Félix.

Eva name-PST.3SG ACC POSS.3 cat Félix

‘Eva named her cat Felix.’

b. Eva nombró a su gato.

Eva name-PST.3SG ACC POSS.3 cat

‘Eva gave her cat a name.’

It can be easily proven that names in the complements of NCs are not arguments, but predicates, so we are not dealing with a ditransitive construction. With the test in (13) for (12a) we can discard that the structure of a NV consists of a double object. In (13b) it can be seen that the recipient of the name behaves as a direct object, as expected, because it can be pronominalized with the accusative pronoun lo used for direct objects in Spanish. Example (13c), though, shows that the PN can’t undergo the same process as the actual argument of the verb can. Consequently, this test indicates that PNs may, indeed, appear as predicative elements.

(13) a. Eva nombró a su gatoi Félixj.

b. Eva loi nombró Félixj.

Eva it.ACC name-PST.3SG Félix.

‘Eva named it Félix.’

c. *Eva loj nombró a su gatoi.

Eva it.ACC name-PST.3SG ACC poss cat

‘*Eva named it her cat.’

Now, as for English NCs, Matushansky (2008) argues that they can only appear with a SC complement, meaning that something like (14) would be ungrammatical. Despite her attempt to make this a typological claim, Spanish NCs like (12) prove this to be incorrect, for a complement with a direct object can be accepted. Even more, it is not the case that every English speaker, or most of them, would reject (14); in fact, it is safe to say that the commonly shared judgment points towards its grammaticality.

(14) The boy named his new pet.

A case for a resultative analysis of NCs can be further pursued with the comparison made by Levin (2019) between lexical causatives and resultative constructions, in the sense that they both lack an explicit causative element. Lexical causatives are sentences like The sound woke me up, or The janitor cleaned the floor. These constructions allow for a causative periphrasis like ‘The sound caused me to be awake’ or ‘The janitor caused the floor to be clean’. The difference between lexical causatives and resultatives is that the former doesn’t include explicit information about the causing event, they rather focus on the caused event. That is, we don’t know what the janitor did to the floor to make it clean. Resultatives, on the other hand, do include information of the causing event.

In contrast, NCs seem to align more with resultatives as described by Levin (2019) because they include information about the causing event through the verb itself; whereas the information about the caused event lies in what she calls result phrase, which in the terms of this paper would be the SC complement of a NV. A typical resultative construction like (15) shows that the verb provides details on how we get to the state in which the tulips are flat; that is, we have more lexical resources to understand the directly causal relation between the event of watering and the state of being flat.

(15) The gardener watered the tulips flat.

This same reasoning is appliable to NVs, as represented in (16a). If we consider that the the boy Elmer is a SC that expresses that its subject (the boy) experiences the state of being Elmer, it can be attested that the NV offers information about the causing event, i.e., the naming event. In other words, the causing event of naming caused the boy to BECOME (or to be) Elmer. Example (16b) is equally subject to this analysis: a causing event of titling caused the play to BECOME (or to BE) Romeo and Juliet.

(16) a. The parents named the boy Elmer.

b. William titled her play Romeo and Juliet.

Further syntactic evidence of the similarity between NCs and resultatives is found in Finnish. Matushansky (2008) even comments on data from Fong (2003) showing that, in this language, the translative case is associated with a BECOME value and is used for change of state constructions, specifically with resultatives and NVs. As (17) shows, the color yellow, which is the predication of the object wall, is declined in translative case, which conveys a resultative change of state. Likewise, in (18) the same pattern occurs with a NV: the proper name Elmer also carries translative case.

(17) Me maalas-i-mme seinä-n keltaise-ksi.

we paint-PST-1PL wall-ACC yellow-TRS

‘We painted the/a wall yellow.’

(18) Me kutsu-mme poika-a Elmeri-ksi.

we name-1PL boy-PART Elmer-TRS

‘We named the boy Elmer.’

Despite this similarity, Matushansky (2008) argues that the fact that the presence of a PN is not optional in English NCs means they can’t be resultatives. Thus, she proposes that NCs have the structure in (19). Besides the obvious framework differences, (i.e., the lack of root and the inclusion of a VP instead of a √P) the main point of interest is that she uses two heads of v to account for the two events: CAUSE and BECOME, yet the structure also has a SC that coordinates the predicate proper name with the object of the verb.

It should be remembered that a SC is meant to be a predicational structure that should express an event or state on its own (Munakata 2006). In that sense, the simultaneous presence of a v BECOME and a SC seems redundant. That is why, I suggest a structure aligned with Harley’s (2011) proposal, as explained above, and exemplified in (20) below, with only a SC complement. In this perspective of argument structure, it is more theoretically suitable to express the optionality of a complement through structural differences of the actual complement, rather than attributing the differences to additional functional heads. This is also more consistent with the fact that NCs don’t need to have a SC complement, like the case of resultatives.

Last, it should also be noted in (20) that my approach considers the NV to be the verbalization of a category-less root, which apart from being a basic notion for word formation in DM, is also an appropriate proposal for verbs usually called denominal. In this sense, several of the NVs across languages have nominal counterparts, that should be analyzed as nominalizations of roots in a n head. (21) shows just a few examples for the basic NV name:

(21) English: name_n , name_v / Spanish: nombre_n , nombrar_v / Nahuatl: tocaitl_n , tocayotia_v / Finnish: nimi_n , nimetä_v / Greek: ónoma_n , onomázo_v

Derived from this, some NVs across languages could exhibit a more literal ‘denominal’ behavior than others. Consider, for example, that a verb like baptize lacks a naming nominal counterpart; that is, there is not a kind of proper name that is a ‘baptism’, whereas there is a ‘nickname’ for the verb nickname. While the morphological discussion regarding how each of type of NV is formed has no implications for the current analysis, it could have an impact on further developments. For instance, In Sánchez Sánchez (2022), I discuss that an actual denominal approach (following Arad (2003) and Acquaviva (2008)) would imply that verbs like baptize, dub, christen and others, could be interpreted as verbalizations directly form a root, whereas verbs like name, nickname, title, etc. would be first nominalized in a n head and then verbalized into v. Nevertheless, Gallego (2015) points out that according to Chomsky (2001) feature valuation prevents nominal heads from selecting arguments, so the SC would have to be the complement of v, which raises a discussion regarding the compositional role of the verb that would exceed the scope of this paper. For this reason, this issue is not fully addressed here.

Once the syntactic structure of NCs has been set and framed as a resultative construction, its semantic account is still not straightforwardly solved. The fact that their SC complement features a PN is in itself a major issue, since predicate PNs are commonly thought of as determiners, not easily compatible with non-referential uses. Finding out how does the meaning of proper names interacts with that of NCs is therefore substantial to properly account for these constructions.

3. The semantics of naming constructions and proper names

This section will deal with the compositional and event structure of NCs and with its interaction with the syntactic structure described in §2. The first step is to explain what is the compositional meaning of proper names that I use for this analysis, and why I do so. As will be seen, this meaning regards PNs as complex entities, that’s why a thorough account is in order. Next, I will problematize Matushansky’s (2008) proposal of the compositionality of this meaning of PN in NCs, in the light of data presented so far. Finally, I will offer a solution and a more insightful explanation of NCs.

3.1. The meaning of proper names

As mentioned in the Introduction, I regard proper names as definite descriptions, which may also appear in non-referential, predicative, positions, like NCs. Semantically, however, it is not clear how could argumental and predicative uses of proper names be derived from the same meaning. Therefore, I will turn to the meaning of proper names as outlined by Matushansky (2008), which attributes not only syntactic but semantic complexity to these entities.

Following this view, proper names exhibit what has been called a naming-predicative interpretation (Matushansky 2015). That is, that they convey the meaning that an entity exists and bears a specific name; for example, (22) shows that sentence (22a) would be interpreted as (22b); or, more precisely, that (22a) always entails (22b). Consequently, we can say that a name is true of an entity if that entity exists and bears the name.

(22) a. Johann Sebastian is a composer.

b. An individual named Johann Sebastian exists and is a composer.

This entailment can be explained by assuming that PNs establish a naming relation between their external argument and a phonological string (that of the name). To account for this, compositionally, Matushansky (2008) argues that two semantic arguments are needed. The first one is the individual argument slot, that had already been proposed by Geurts (1997), in other words, a slot for the name bearer, as seen in (23). However, Matushansky argues that (23) constitutes an incomplete meaning, since an entry like (23) provides no indication as to what kind of naming convention (Recanati 1997) holds between the name and its bearer. For instance, PNs may be nicknames, titles, names, surnames, or even unspecified relations; for this reason, she suggests formalizing the naming convention as R. R, therefore, is a free variable that relates the entity to the phonological string through an existing naming convention that specifies what kind of name is the name. The entry in (24) illustrates this process and constitutes the meaning of proper names that I follow in this paper.

(23) [[Pablo]] = λx (x is an entity named /pablo/)

(24) [[Pablo]] = λx [λx ∈ D_e] [λR_<e,<n,t>>] R(x) (/pablo/)

Before moving on, let’s break down (24). This entry should be read as: for the function of x such that it belongs to the set of entities, and for the function R where R is a naming convention that associates an entity with a phonological chain, x holds that relation with /pablo/. It should be noted that n is a type of entity that Matushansky includes, consisting of the phonological string of a PN.

Now, regarding the R slot, the logical step would be to explain how it is saturated. There are two answers provided by the author: for argumental PNs, R is a free variable, contextually saturated by the naming convention in force between the speaker and the hearer. However, as we’ll see, the proposed mechanism for predicate PNs makes incorrect predictions about NVs and NCs, so I will turn now to present the complications that arise in the interaction of PNs and NCs, both syntactically and semantically, and then to provide a solution that accounts correctly for the data reviewed so far.

3.2.Matushansky’s (2008)analysis of NCs

Given that NCs are the main focus of this paper, I will only discuss how could an entry like (24) be compatible with a predicative structure. First, Matushansky’s (2008) original idea must be reviewed in order for mine to be explained. Let’s use (25) as the NC that will exemplify how Matushansky’s theory works. Assuming the meaning of a PN as in (24), the SC The teacher Stinky would be composed as shown in (26); it should be noted that the subject of the SC, the teacher, saturates the individual slot required by the PN. This occurs because the meaning of the name demands that we deal with an individual and with the link between a phonological string and the individual (R); consequently, it is assumed that there is, indeed, an individual being referred to by the name. Since there is no referentiality in predicative uses of PNs, the individual argument slot must be filled differently, namely, with the subject of the SC.

(25) The class nicknamed the teacher Stinky.

(26) [[the teacher Stinky]] =

= [[λx_e ] [λR_<e,<n,t>>] R(x) (/stinki/)] (the teacher)

=[λR] R (the teacher) (/stinki/)

Once this happens, the second argument slot, the R variable, is still empty; however, we can’t recur to a contextual naming convention associating the predicate PN and the individual beforehand to saturate it. In turn, the author’s solution is that a NV is in charge of saturating this naming convention; that is, the verb determines what kind of link is established between the phonological string and the individual, rendering the following compositional meaning:

(27) [[nickname the teacher Stinky]] =

= [[ λw] [∃ R (NICKNAME (W) (R) & R (THE TEACHER) (/STINKI/))]]

According to Matushansky, (27) supposes that the NV evokes a specific naming function, in this case NICKNAME, that evaluates whether a relation (type <e,<n,t>>) between an individual and the phonological string of a PN is, in fact, a nicknaming relation. This function determines if this relation exists and holds between the subject of the SC and its predicate, i.e., the PN. Furthermore, the inclusion of possible worlds is essential in this perspective because predicate PNs are not rigid, so the naming convention only needs to exist in a possible world, not in every world. As such, (27) denotes a proposition that is true in the possible world w if, and only if, there is a relation between the DP the teacher and the phonological chain /stinki/, such that this relation is a nicknaming convention in w.

3.3. Problematization of the compositionality of NCs

There are two major issues with Matushansky’s compositional account to be solved. First, the conformation of the naming SC as outlined in (26) leads to a theoretical incompatibility with the meaning of the PN and the syntactic structure itself. Let’s take (26) and see where the problem lies. Assuming that the meaning of a PN is the entry in (24), the saturation of its two arguments cannot happen the way Matushansky claims it does in (27). As I mentioned, a SC should predicate a state on its own (Harley 2011), therefore, it should have a complete meaning. However, in Matushansky’s terms, a naming SC is not interpretable, because one of its elements, the PN, is semantically incomplete by the time the SC is assembled. Moreover, the fact that a PN would need the presence of a NV in order to saturate its second argument (R) would cause that a naming SC would have an incomplete meaning, dragging a PN with no full use inside the SC. Vice versa, having a NV that demands the presence of a PN makes the wrong predictions about the type of complement NVs can display.

Precisely, the second issue requires going back to the examples of (12b) and (14), repeated below in (28), that exhibit NCs without SCs, only the DO of the verb. It can be observed that an entry like (27) is clearly not supporting NVs without a SC complement. As explained, since (27) would require the verb to quantify over naming conventions, and there are no proper names in the complements of these examples to satisfy this condition, these sentences should be ungrammatical. Given that they are not, an alternative compositional account is in order.

(28) a. Eva nombró a su gato.

Eva gave her cat a name’

b. The boy named his new pet.

Consequently, a model that can deal with both event semantics and compositional semantics must be used to address the differences between Matushansky’s proposal and mine, as well as to link this new proposal to the syntactic structure for a resultative construction presented in (20) and successfully account for a state interpretation of PNs. To achieve this, I’ll recur to a neo-Davidsonian model of event semantics (Parsons 1990). From this perspective, the event described by the verb is its only argument, and the relation with its syntactic arguments are expressed through thematic roles, which syntactic correlate are theta-roles (Chomsky 1995). Additionally, the particular approach that I will adopt leads to posing quantification over events, so that it can be said that a verb entails the existence of the event, which is its argument.

While from some perspectives event and compositional semantics are not compatible (Heim & Kratzer 1998), there are other models to combine them, like that of Pylkkänen (2008) or Champollion (2015). The latter allows accounting for both semantic aspects without the need to commit with argument order, distinction between arguments and adjuncts, among others. The fundamental idea for my purposes is that both consider that verbs and their projections denote existential quantification over events, in contrast to other compositional perspectives that treat events as sentence-level variables.

Following along, I suggest departing from the conformation of a SC in Matushansky’s terms (26), repeated below as (29). As I pointed out, I consider naming SCs to have the expected characteristics of any type of clause, i.e., to consist of a subject and a predicate, and to express a state or event. A key to solving the issue of a SC that depends on a NV to be fully propositional is precisely its ability to predicate. As mentioned back in §3.3, Matushansky’s SC has no predicative value, since she proposes two v heads to account for the bi-eventive NC. On the contrary, my account regards that the SC does predicate, so we should embrace the intrinsic predication of this structure instead of the verb’s, which is ensembled afterwards.

(29) [[the teacher Stinky]] =

= [[λx_e ] [λR_<e,<n,t>>] R(x) (/stinki/)] (the teacher)

=[λR] R (the teacher) (/stinki/)

Another point that should be noted in (29) is that, according to this approach, the SC would not only be propositionally nullified but structurally. (29) aims to saturate the individual argument of the PN, which is clearly desirable, however, not in this way. Matushansky ensembles what should be the subject of a SC inside the name’s argument structure, in other words, the subject of the SC is contained inside its own predicate. As a result, a PN would be equivalent to a SC. Moreover, this perspective also leaves no room for an event or state described by the clause. In contrast, I suggest that the PN, indeed, saturates its arguments with the contents of the SC, but not until it is fully ensembled, as illustrated in (30):

(30) [[the teacher Stinky]]

= λx.∃s[Stinky(s) ^ theme (s, x)] (the teacher)

= ∃s[Stinky(s) ^ theme (s, the teacher)]

Therefore, the compositional process of the SC and the argument saturation of the PN follow (30). Inside the SC, there is a subject (the teacher), that saturates the variable x corresponding to the subject of a state, and it is specified that it is the Theme of the state Stinky. Thanks to the predicative nature of SCs this could be interpreted as BEING Stinky, and so the R argument of the PN may be saturated by this naming predicate, as it would with a contextual naming convention for argument PNs. This, in my view, is a more desirable theoretical outcome because it unlinks the semantics of PNs from that of NVs, therefore, we address and solve the first issue that I pointed out from Matushansky’s theory. Besides, we now have a PN that can be interpreted as a result state inside the SC.

Once the SC is set, the PN can saturate its individual argument with the variable x that corresponds to the teacher. Summing up, the SC, as a whole, describes a proposition where there is a state s (the name) and a participant that experiences such state. It is worth noticing that the PN has already saturated its two arguments, in contrast with what we observed in (29), where a NV was still required to saturate R.

3.4. A neo-Davisonian resultative approach to NCs

Having explained what happens with the SC, it is now the turn to explain my proposal for the complete NC. The semantic account I suggest for a NC with a SC complement appears in (31). Basically, it affirms that there is an event e such that it is of nicknaming and the agent of this event is the class; and that there is a state s such that Stinky is the state and its theme is the teacher; and that the event e causes the state s. Thus, we can account for the presence of two eventualities (an event and a state) conforming a causative meaning for the NV. Next, I’ll examine what (31) implicates.

(31) [[The class nicknamed the teacher Stinky]]

= ∃e[nickname(e) ^ agent (e, the class) ^ ∃s[Stinky(s) ^ theme

(s, the teacher)] ^ CAUSE(S, E)

A syntactic tree associating the structure proposed in §2.2 and the meaning of a NC in (31) features in (32). Focusing on the SC first, it displays a semantic type <s,t>, where s refers to an event or state and is obtained interpreting the predicate of the SC as a change of state event (Pylkkänen 2008). Then, as we saw, the predication coming from the SC itself saturates the naming convention required for R. It is worth noting, however, that the naming convention R associated with the state of BEING Stinky can be further specified pragmatically by whatever naming verb is merged to the SC, but this specification cannot be part of the compositional requirements of the PN in order to preserve the SC’s propositional and compositional autonomy.

Following the composition of a NC as described in (31), the CAUSE component in (32) is indicating that, by the means on nicknaming, the agent causes the teacher to bear the name Stinky. I should also point out that the root that is merged with the SC has no semantic type, in accordance with the DM notion that roots have no category. Following Acquaviva (2008: 4), if we mean to take seriously the category-less nature of roots, then we should suppose that they have no coherent meaning by themselves, nor a categorization in semantic types; rather, they are morphosyntactic pieces that give rise to words subject to categorization.³ The structure in (32) also commits to the idea that it is possible for roots to take complements (Acedo-Matellán & Mateu 2015); as illustrated above, the SC is the complement of the root, yielding a √P. Continuing, once the verb is formed in the v head, it acquires the type <e,<s,t>>, so it must select a state with type <s,t>. I argue that the v head selects, via Event Identification (Kratzer 1996), the state encoded by the SC, which has the required type. Last, the verbalizer head also requires the presence of an agent type e, that is satisfied by merging the DP the class as its specifier.

Finally, the perspective that I’ve outlined to account for the semantics of NCs can successfully predict and deal with NCs without SC complements if we consider that a NV always takes a propositional complement, even if it is not explicit. To exemplify, take again (33). In this sentence, there would be a naming event and because of it an entity experiences the state of being named; even without specifying a PN, the proposition entails its existence. Therefore, we can account for the alternation of the complement that a NV can take, thus addressing the second issue with Matushansky’s (2008) proposal. The event decomposition in (34) seeks to capture the bieventive nature of naming verbs. This is accomplished considering an abstract state -being x (where x is a PN)-⁴ that predicates that an entity enters such state as a result of the naming event; in other words, the naming event still causes an entity to bear a name, even if it is not spelled out.

(33) Eva nombró a su gato.

‘Eva named her cat.’

(34) [[Eva named her cat]]

=∃e[name(e)^agent (e, Eva) ^ ∃s [BEING(S) ^ theme (s, her cat)]

^ CAUSE (s, e)

Summing up, I consider that the neo-Davidsonian and compositional account of the semantics of NCs that was presented in this section has accurately described this construction and solved the empirical and theoretical problems regarding both its syntactic possibilities and the interaction with the meaning of PNs that were pointed out from Matushansky’s (2008) account. Furthermore, this analysis allows to treat NCs with SC complements as resultatives, where a predicate PN is the result state of the main naming event.

4. Conclusion

In this paper, I advocated for and supported a resultative analysis of naming constructions, within a DM framework and a minimalist approach to argument structure. Prominently, I argued in favor of analyzing the complement of naming verbs as a small clause consisting of a subject DP and a predicate proper name. However, I also presented data showing that NCs can also appear with a direct object complement alone. Likewise, I thoroughly described the semantics of these constructions. Like any change of state construction, NCs are bi-eventive in nature, for that reason, I analyzed both their compositional and event semantics. On one hand, I provided an account for the composition of the SC provided the meaning of proper names proposed by Matushansky (2008). On the other hand, the complete compositional an eventive semantics of the NC were addressed and two major issues with Matushansky’s proposal were highlighted. First, an empirical problem: this author’s account failed to predict and explain the uses of NVs without SC. Second, a theoretical objection: Matushansky´s meaning of proper names was heavily liked to NVs, causing for the SC to be propositionally dependent on a verb, which ultimately was the root of the first problem. Therefore, I presented a solution using a neo-Davidsonian model, under which I claimed that a NV selects a result state, interpretable as BEING x where x is a PN, which can be either expressed through a SC or entailed by the NV. Consequently, my proposal grants that NCs without a SC are accounted for, broadening the scope of Matushansky’s (2008) proposal and resulting in a more robust theory of NVs and PNs in predicate positions, able to account for variation across languages in the syntax of NCs.

Referencias

Acedo-Matellán, Víctor & Mateu, Jaume. 2015. From syntax to roots: A syntactic approach to root interpretation. In Alexiadou, Artemis & Borer, Hagit & Schäfer, Florian (eds.), The syntax of roots and the roots of syntax, 14-32. Oxford: Oxford University Press.

Acquaviva, Paolo. 2008. Roots and lexicality in distributed morphology. (MS., University College Dublin/ Universität Konstanz)

Arad, Maya. 2003. Locality constraints on the interpretation of roots: The case of Hebrew denominal verbs. Natural Language & Linguistic Theory 21. 737-778.

Bach, Kent. 2015. The predicative view of proper names. Philosophy Compass 10(11). 772-784.

Bauer, Winifred. 2003. The Reed reference grammar of Māori. Auckland, New Zealand: Reed.

Boër, Steven. 1975. Proper names as predicates. Philosophical Studies 27(6). 389-400.

Borer, Hagit. 2005. Structuring sense 1: In name only. Oxford: Oxford University Press.

Bowers, John. 1993. The syntax of predication. Linguistic Inquiry 24(4) 591-656.

Champollion, Lucas. 2015. The interaction of compositional semantics and event semantics. Linguistics and Philosophy 38. 31-66.

Chomsky, Noam. 1995. The minimalist program. Cambridge: MIT Press.

Chomsky, Noam. 2001. Derivation by phase. In Kenstowicz, Michael (ed.), Ken Hale: A life in language, 1-52. Cambridge: MIT Press .

Elbourne, Paul. 2005. Situations and individuals. Cambridge: MIT Press.

Fara, Delia Graff. 2015. Names are predicates. Philosophy Review 124(1). 59-117.

Fong, Vivienne. 2003. Resultatives and depictives in Finnish. In Nelson, Diane & Manninen, Satu (eds.), Generative approaches to Finnic and Saami linguistics, 201-233. Stanford: CSLI.

Frege, Gottlob. 1970 [1892]. On sense and reference. In Geach, Peter & Black, Max (eds.), Translations from the philosophical writings of Gottlob Frege, 58-78. Oxford: Blackwell.

Gallego, Ángel J. 2015. Roots and phases. In Alexiadou, Artemis; Borer, Hagit & Schäfer, Florian (eds.), The syntax of roots and the roots of syntax, 192-207. Oxford: Oxford University Press.

Geurts, Bart. 1997. Good news about the description theory of names. Journal of Semantics 14. 319-348.

Hale, Kenneth & Keyser, Samuel Jay. 2002. Prolegomena to a theory of argument structure. Cambridge: MIT Press.

Halle, Morris & Marantz, Alec. 2005. Distributed morphology and the pieces of inflection. In Hale, Kenneth & Keyser, Samuel, Jay (eds.), The view from building20, 111-176. Cambridge: MIT Press.

Harley, Heidi. 2005. How do verbs get their names? Denominal verbs, manner incorporation, and the ontology of verb roots in English. In Erteschik-Shir, Nomi & Rapoport, Tova (eds.), The syntax of aspect. Deriving thematic and aspectual interpretation, 42-64. Oxford: Oxford University Press.

Harley, Heidi. 2011. A minimalist approach to argument structure. In Boeckx, Cedric (ed.), The Oxford handbook of linguistic minimalism. Oxford: Oxford University Press.

Harley, Heidi. 2012. Lexical decomposition in modern syntactic theory. In Hinzen, Wofram; Werning, Markus & Machery, Edouard (eds.), The Oxford handbook of compositionality, 328-350. Oxford: Oxford University Press.

Heim, Irene & Kratzer, Angelika. 1998. Semantics in generative grammar. Oxford: Blackwell.

Julien, Marit. 2005. Nominal phrases from a Scandinavian perspective. Amsterdam: John Benjamins.

Kratzer, Angelika. 1996. Severing the external argument from its verb. In Rooryck, Johan & Zaring, Laurie (eds.), Phrase structure and the lexicon, 109-137. Dordrecht: Kluwer.

Kripke, Saul. 1980. Naming and necessity. Oxford: Blackwell.

Levin, Beth. 2019. Resultatives and causation. (MS., Stanford University)

Levinson, Lisa. 2007. The roots of verbs. New York University. (Ph.D. thesis.)

Marantz, Alec. 1997. No escape from syntax: Don’t try morphological analysis in the privacy of your own lexicon. In Dimitriadis, Alexis; Siegel, Laura; Surek-Clark, Clarissa & Williams, Alexander (eds.), Proceedings of the 21st Annual Penn Linguistics Colloquium, vol. 4, 201-225. Philadelphia: University of Pennsylvania, Penn Linguistics Club.

Marantz, Alec. 2005. Objects out of the lexicon! Argument structure in the syntax. (MS., MIT)

Mateu, Jaume. 2015. La estructura argumental. In Gallego, Ángel J. (ed.), Perspectivas de sintaxis formal, 111-148. Madrid: Akal.

Matushansky, Ora. 2006. Why Rose is the rose: On the use of definite articles in proper names. Empirical Issues in Syntax and Semantics 6. 285-307.

Matushansky, Ora. 2008. On the linguistic complexity of proper names. Linguistics and Philosophy 31(5). 573-627.

Matushansky, Ora. 2015. The other Francis Bacon: On non-bare proper names. Erkenntnis 80. 335-362.

Munakata, Tetsuo. 2006. Japanese topic-constructions in the minimalist view of the syntax-semantics interface. In Boeckx, Cedric (ed.), The syntax of roots and the roots of syntax, 115-159. Amsterdam: John Benjamins .

Parsons, Terence. 1990. Events in the semantics of English. Cambridge: MIT Press.

Pylkkänen, Liina. 2008. Introducing arguments. Cambridge: MIT Press.

Recanati, François. 1997. Direct reference: From language to thought. Oxford: Blackwell.

Sánchez Sánchez, Aarón Pablo. 2022. Complejidad sintáctica y semántica de los nombres propios. UNAM. (Ph.D. thesis.)

Stowell, Tim. 1981. Origins of phrase structure. MIT. (Ph.D. thesis.)

Von Stechow, Arnim. 1995. Lexical decomposition in syntax. In Egli, Urs; Pause, Peter E.; Schwarze, Christoph & von Stechow, Arnim & Wienold, Götz (eds.), Lexical knowledge in the organization of language, 81-118. Amsterdam: John Benjamins .

Notes