ADJECTIVES OF COMPLETENESS AS MAXIMIZERS OF EVENT NOMINALIZATIONS 1

Aspectual modifiers that have adverbial and adjectival counterparts are an important source of information for researching lexical aspect in the nominal domain. One class of such modifiers are adverbs of completeness (completamente 'completely', totalmente 'totally'), which are maximizers when modify adjectives, and their correspondent adjectives (completo 'complete', total 'total'). This paper addresses the inheritance of aspectual features and its relation to degree in event nominalizations of incremental theme verbs (traduccción 'translation', destrucción 'destruction') through the analysis of the modification performed by adjectives of completeness in Spanish. The proposal combines a syntactic account of deverbal nominalizations with a scalar approach to aspect. Adjectives of completeness are argued to be aspectual modifiers, with a contribution equivalent to that of their adverbial counterparts both in the verbal and in the adjectival domain, establishing thus a parallelism between degree modification of adjectives, verbs and nominals.


Introduction
Aspectual modifiers that have adverbial and adjectival versions are an important source of information for researching lexical aspect in the nominal domain.The distribution and interpretation of adjectives of completeness [henceforth, ACs] in Spanish (completo 'complete', total 'total') strongly correlates with those of their corresponding adverbs (completamente 'completely', totalmente 'totally'), which have been analyzed in the literature as degree modifiers sensitive to scale maximums (Rotstein and Winter 2004;Kennedy and McNally 2005;a.o.).In both examples in (1), the modifier is acceptable only when the direct object denotes an individual with precise limits, and its contribution is to assert that the translating event has reached its endpoint.
a. Paula tradujo completamente {la novela / ??novelas}.Paula translated completely the novel novels 'Paula completely translated {the novel / ??novels}.'b. la completa traducción de {la novela / ??novelas} (por parte de the complete translation of the novel novels by Paula) Paula 'the complete translation of {the novel / ??novels} (by Paula).' The data in (1) is interesting for two reasons.On the one hand, from the perspective of the verbal predicate, these data raise the question of whether aspectual features are transferred in nominalization and, if so, how aspect is manifested in the nominal domain.Incremental theme verbs constitute an interesting case study because their telicity is not dependent on the verbal predicate itself, but on the nominal reference of its internal argument, and the same seems to be the case in nominals derived from those verbs.
On the other hand, from the point of view of the modifiers, the examples in (1) suggest that some adjectives can act as aspectual modifiers if they combine with the right kind of nominal.In addition, since these modifiers are related to degree in the adjectival domain (as in, e.g., completely full), exploring this issue brings up a possible connection between aspectual and degree modification across categories.
This paper argues that ACs are aspectual modifiers in cases like (1b) and analyzes them as maximizers, unifying thus the semantics of ACs modifying event nominalizations of incremental theme verbs with the semantics of adverbs of completeness in their modification of adjectives (completely full) and VPs (completely translated the novel).In order to do so, I adopt a scalar approach to aspectual composition (Krifka 1998;Hay, Kennedy & Levin 1999;Piñón 2005Piñón , 2008;;Caudal & Nicolas 2005;Beavers 2008; Kennedy & Levin 2008;Kennedy 2012;a.o.).In particular, I combine Ramchand's (2008) and Kennedy's (2012) proposals for incremental theme verbs in order to account for the syntax and semantics of their nominalizations.The main idea is that both the verbal predicate and its nominalization include a degree argument provided by their internal argument.This degree argument provides the measure of the parts of the object affected during the event.As a consequence, the referential properties of the internal argument, and ultimately the value of the degree argument, are responsible for the telicity of the event description.ACs are defended to be degree modifiers able to saturate that degree argument and, therefore, interact with the aspect of the nominalization.
This paper is organized as follows.Section 2 lays out the properties of maximizers in the adjectival domain and section 3 shows that these modifiers are present in the verbal domain as well, which provides the basis for the analysis of ACs later on.Section 4 is devoted to show that ACs do have the properties of maximality modifiers when modifying event nominalizations of incremental theme verbs.The syntax and semantics of these nominalizations are discussed in section 5. Section 6 provides the analysis of ACs as maximality modifiers and discusses its consequences.Finally, section 7 concludes.

Maximality modification in the adjectival domain
Gradable adjectives lexicalize scales that may include a maximal and/or a minimum value (Rotstein & Winter 2004;Kennedy & McNally 2005).Scale endpoints are relevant for setting the standard for the predicate.In particular, the standard corresponds to the upper or lower bound whenever there is one.Otherwise, the standard is calculated contextually (Kennedy 2007).The typology of adjectives based of their scale structure is provided in (2).These modifiers indicate that the referent has a maximal degree of the gradable property denoted by the adjective they combine with.More formally, maximizers set the value of the degree argument of the adjective G to the maximum in its scale S G (4) (Kennedy and McNally 2005).Since the function max only yields a value if the scale has a maximum defined, the restriction to upper and totally closed adjectives is accounted for.
( Kennedy and McNally 2005) Maximality semantics is associated with a number of properties.First, maximizers entail that the end of the scale has been reached.Consequently, it is contradictory to assert that the referent can have a higher degree of the property (5a) (Kennedy & McNally 2005).Second, the construction maximizer G is a total construct, in the sense that it has the distribution of an upper-closed scale adjective (Rotstein & Winter 2004).This is shown by the fact that it is compatible with casi 'almost' (5b).And third, because of the universal quantification in the semantics of the max function, maximizer G accepts exceptive phrases (5c).To sum up, maximizers are degree modifiers restricted to adjectives that use a scale closed (at least) in its upper end.They set the degree of the property denoted by the adjective to that maximum.I turn now to showing that adverbs of completeness are maximality modifiers in the verbal domain.

Maximality modification in the verbal domain
Adverbs of completeness modify verbs as well.As shown in ( 6), the presence of completamente 'completely' implies that the event of translating the novel was developed to its culmination.

Paula
tradujo completamente la novela.Paula translated completely the novel 'Paula completely translated the novel.'Intuitively, the culmination of an event corresponds with the end of a scale and, consequently, its maximum degree.If adverbs of completeness modifying VPs were acting as maximizers, and the analysis in (4) could thus be extended, two predictions would follow.First, adverbs of completeness would be restricted to events that are durative (that develop in time and thus can be mapped onto a scale) and that are telic (that have an upper bound).Second, they would show the properties of maximizers presented in section 2. Both predictions are borne out.
First, adverbs of completeness only occur with accomplishments (durative and telic) (7a), and are not acceptable with states (punctual and atelic) (7b), activities (durative and atelic) (7c), or achievements (punctual and telic) (7d).Therefore, they require that the event be extended in time (durative) and that it have an endpoint (telic) (Caudal & Nicolas 2005;Piñón 2005 Event descriptions display certain characteristics that are close to those found in degree semantics.For instance, both event descriptions and degree expressions can be modified by certain expressions, such as proportional adverbs (half, completely) (Caudal and Nicolas 2005;Piñón 2005).Scalar accounts of aspectual composition draw on this fact to provide a common framework for different types of verbs with variable telicity (Krifka 1998;Hay, Kennedy & Levin 1999;Piñón 2005Piñón , 2008;;Caudal & Nicolas 2005;Beavers 2008;Kennedy & Levin 2008;Kennedy 2012;Demirdache & Martin 2015;a.o.).In these accounts, progress of an event corresponds to movement along a scale that measures the change undergone by a participant in the event (the incremental theme, the patient, or a figure participant).In every case, the affectedness of the participant (or the position in a path) provides an appropriate bound for the scale, which in turn corresponds to a bound (or telos) for the event.I adopt here a scalar approach to variable telicity in order to give a unified treatment to adverbs of completeness in the adjectival and the verbal domain, and to the corresponding adjectives in the domain of eventive nominalizations.
According to degree accounts of aspect, verbs with variable telicity share in their meaning a function that measures the degree to which an object changes relative to some scalar dimension over the course of the event.In the case of incremental theme verbs, their telicity depends on the referential properties of the theme argument (Verkuyl 1972;Dowty 1979;Krifka 1989;Filip 1999;Kennedy 2012;Bochnak 2013; 2 Some speakers find this type of examples slightly degraded, especially when compared to the versions without the modifier (e.g., Paula tradujo la novela, excepto dos páginas 'Paula translated the novel, except for two pages.')(see also (18c)).I would like to argue that this is related to the fact that, although maximality modifiers have a semantic effect (cf.Sassoon and Zevakhina 2013), their use has the side effect of decreasing the amount of imprecision allowed in the context.Consequently, the number of possible exceptions is reduced, making exceptives slightly less felicitous than when no maximizer is present.a.o.).Specifically, when the theme argument has quantized reference, the event is telic (9a).This is indicated by their compatibility with temporal in-adverbials (Vendler 1957;Dowty 1979).On the other hand, if the theme has cumulative reference, the event is telic (9b).According to Krifka (1989), a predicate has quantized reference if, whenever it holds of something, it does not hold of any of its proper parts.In (9a), la novela 'the novel' is quantized because a proper part of a novel is not a novel.By contrast, a predicate is cumulative if whenever it holds of two things, it also holds of their collection.For instance, novelas 'novels' in (9b) has cumulative reference because the collection of two sets of novels can also be referred to as novels.As a summary, adverbs of completeness in the VP display the same restriction to scales with a maximal degree and the same properties as when they modify adjectives (see section 2).This supports an extention of the analysis of adverbs of completeness as maximizers when modifying adjectives to cover their modification of VPs.The fact that only telic interpretations of verbs of incremental theme verbs are compatible with adverbs of completeness (and, looking ahead, the same is true of their nominalizations) makes them interesting to delve into the interaction between aspect, degree and maximality in the verbal and nominal domain.For this reason, incremental theme verbs and their nominalizations constitute the focus of this paper.
Before proceeding to the nominalizations, it is worth mentioning that there are two other classes of verbs showing variable telicity, namely degree achievements and directed motion verbs.As I will mention in the conclusion, the analysis put forward in this article for (nominalizations of) incremental theme verbs can be extended to these verbs.In the case of degree achievements, they describe the event of increasing or decreasing the values of a property of the internal argument, such as closeness or As for motion verbs, their telicity is related to the boundedness of their path (Dowty 1991;Tenny 1992;Jackendoff 1996;Krifka 1998;Zwarts 2005;Beavers 2008).If the path one of the participants of the event moves along is bounded, the event is telic (12a).Otherwise, the event is atelic (12b).Next section introduces the different types of deverbal nominalizations and shows that ACs are sensitive to the aspect of event nominalizations (which they inherit from their base) and behave as maximizers when modifying these nominals.

Adjectives of completeness and deverbal nominalizations
Deverbal nominalizations do not form a homogeneous class.Rather, they differ with respect to whether they obligatorily take arguments and denote eventualities (Grimshaw 1990;Borer 2003;a.o.).According to these criteria, deverbal nouns have been divided into complex event nominals, result nominals and simple event nominals.The former, complex event nominals (14a), support arguments and denote events; result nominals (14b) neither take arguments obligatorily nor denote eventualities.Simple event nominals (14c) are underived, denote events but do not realize event participants as arguments.However, this classification is obscured by the fact that many of these nouns can be ambiguous between an event or process and a result reading.For example, a nominal like construcción 'building' can refer both to the event of building something or to an entity, the result of that building.From the point of view of the obligatoriness of arguments, the difference has been recast in terms of argument structure (AS) nominals (Grimshaw's complex event nominals) and referential (R) nominals (Grimshaw's result and simple event nominals) (Borer 2003).AS-nominals (or event nominalizations) are associated with a set of properties that sets them apart from R-nominals (Grimshaw 1990;Borer 2003;cf., e.g., Newmeyer 2009;Grimm & McNally 2013).For instance AS-nominals accept aspectual modifiers such as in/for-adverbials (15a) (cf.( 15b)).a. La completa traducción está encima de la mesa.the complete translation is on.top of the table 'The complete translation is on the table.' (='the comprehensive translation') b.La traducción completa está encima de la mesa.the translation complete is on.top of the table 'The complete translation is on the table.' (='the whole translation') When ACs combine with event nominalizations, they display the properties of maximizers.First, ACs entail that the event has come to its endpoint.There is thus a contradiction in adding, for instance, that two chapters are left to be translated after a complete translation (18a) (cf.( 5a) and ( 8a)).Second, they are compatible with casi 'almost' (18b) (cf.( 5b) and ( 8b)).And third, they accept exceptive phrases (18c) (cf.( 5c) and ( 8c)).5 a. #La completa traducción de la novela tuvo lugar en abril; the complete translation of the novel had place in April quedan dos capítulos.remain.3PLtwo chapters 'The complete translation of the novel took place in April; there are two chapters left.' b.La casi completa traducción de la novela tuvo lugar en abril.
the almost complete translation of the novel had place in April 'The almost complete translation of the novel took place in April.' c. La completa traducción de la novela, excepto dos páginas, the complete translation of the novel except two pages tuvo lugar en abril.had place in April 'The complete translation of the novel, except for two pages took place in April.' As a summary, ACs appear with nominalizations of incremental theme verbs with different effects.With result nominalizations, as expected from their lack of eventive meaning, ACs have their literal reading.However, when ACs occur with event nominalizations, they are sensitive to aspect (they only combine with nominalizations denoting telic events) and show the properties of maximizers.This constitutes evidence that the degree analysis of adverbs of completeness can be extended to ACs' modification of event nominalizations of incremental theme verbs.The next section is dedicated to spell out such an analysis.

Event nominalizations of incremental theme verbs
Event nominalizations preserve the lexical aspect of their morphological base (Gross and Kiefer 1995;Haas, Huyghe & Marín 2008), unless the nominalizing morpheme contributes an aspectual effect.In the case of of incremental theme verbs, many of their event nominalizations are formed by the suffixes -ción, -ura, the past participle, or no nominalizer (that is, they are zero nominalizations) (19).6 a. traducir -traducción; destruir -destrucción; revisar -revisión translate translation destroy destruction revise revision b. leer -lectura; escribir -escritura; read reading write writing c. recoger -recogida; pelar -pelado; bordar -bordado collect collection peel peeling embroid embroidering d. dibujar -dibujo; estudiar -estudio draw drawing study study None of these ways of building an event nominalization induces an aspectual effect (see, e.g., Meinschäfer 2005;Fábregas, Marín & McNally 2012).Therefore, the lexical aspect of the incremental theme verb carries over to their nominalizations, as shown by their compatibility with in-and for adverbials.The nominalizations in (20a), where the incremental theme argument has quantized reference, denote telic events.As such, they are acceptable with in-but not for-aspectual phrases.By contrast, when the theme argument has cumulative reference, the event denoted by the nominalization is atelic, and the event description is only compatible with foradverbials (20b) (cf.( 9)).
a. La {traducción / escritura} de la novela {en / ??durante} dos the translation writing of the novel in for two semanas tuvo lugar en abril.weeks had place in April 'The {translation / writing} of the novel {in / ??for} two weeks took place in April.' TELIC b.La {traducción / escritura} de novelas {??en / durante} dos semanas the translation writing of novels in for two weeks tuvo lugar en abril.had place in April 'The {translation / writing} of novels {??in / for} two weeks took place in April.' ATELIC In short, nominalizations of incremental theme verbs preserve the aspect of their verbal base.From a degree-based account of aspectual composition, this means that these nominalizations include a degree argument that measures out the event.This measurement is based on an extent scale that tracks the part structure of the theme argument involved in the event.This section is devoted to provide a scalar account of nominalizations of incremental theme verbs based on Kennedy's (2012) for incremental theme verbs, which will be used in section 6 to argue for an analysis of ACs as aspectual modifiers.In what follows, I first introduce in section 5.1 the scalar approach to incremental theme verbs and discuss the source of the degree argument, which will be determined to be the internal argument.After, sections 5.2 and 5.3 provide the syntax and semantics, respectively, for nominalizations of incremental theme verbs.

Scalar approach to incremental theme verbs
Incremental theme verbs describe events in which the internal argument of the verb undergoes an incremental change over the course of the event (Verkuyl 1972;Krifka 1989Krifka , 1992;;Tenny 1994;Dowty 1991; a.o.).As mentioned above, the referential properties of the incremental theme argument determine the telicity of the event described by the incremental theme verb (Verkuyl 1972;Dowty 1979;Krifka 1989Krifka , 1992;;Filip 1999;a.o.) or its nominalization.As shown in (20) above, if the internal argument has quantized reference, the event is telic (20a).If it has cumulative reference, the event is atelic (20b).
The relation between the event structure of the verb and the part structure of an incremental theme argument can be thought of as an homomorphism whereby each subevent of the main event is mapped to a subpart of the object, and vice versa.This is known as the object-event homomorphism (Krifka 1989).The homomorphism ensures the correspondence between the part structure of the incremental theme argument and the development of the event.As a consequence, the referential properties of the object then carry over to the event: A quantized DP such as the novel denotes an object with precise limits, and its limit acts as a bound for the event of translating it.On the other hand, a cumulative DP like novels denotes something without clear limits, and its lack of limits leaves the event without an endpoint.Telicity is thus dependent on the boundedness of the object (cf.e.g., Rothstein 2004;Arsenijević 2006).
From a scalar perspective, the part structure of the referent of the internal argument constitutes a scale that is homomorphic to the progress of the event.The part structure of an object with quantized reference can be mapped onto bounded scales, whereas the part structure of an object with cumulative reference is mapped onto unbounded scales.Due to the homomorphism between the part structure of the object and the development of the event, the boundary of the part-structure scale corresponds with a telos for the event (Caudal and Nicolas 2005;Piñón 2005;Kennedy 2012;Bochnak 2013).
Therefore, a degree analysis of incremental theme verbs and their nominalizations assumes that there is a scale that tracks the parts of the internal argument affected during the course of the event.In order to provide a specific analysis, we first need to identify the source of the measure function and the role played by the nominal reference of the incremental theme.This is done in turn in the next two subsections.

Incremental theme verbs do not lexicalize a degree argument
There are two possibilities for the source of the degree argument in incremental theme verbs.It can either be lexically encoded in the verb or be provided by the theme argument.
According to the first option, verbs like translate or read include a measure function in their semantics that measures how much of their theme argument is affected by the event.That is, how much of the novel gets translated or read.This approach can be implemented either by including a degree argument directly in the lexical semantics of the incremental theme verb (Caudal and Nicolas 2005;Piñón 2008) or via a type shift that introduces the degree function (Piñón 2005).In addition, this sort of approach needs a way to ensure the connection between the degree function and the incremental theme involved in the event.
However, there is evidence against incremental theme verbs lexicalizing a degree argument, at least in English (Rappaport Hovav 2008;Levin and Rappaport Hovav 2010;Kennedy 2012).Rather, the scalar meaning seems to be associated with the quantity of the referent of the incremental theme argument involved in the event.A first piece of evidence comes from resultatives.As Rappaport Hovav (2008) points out, verbs that lexicalize scales as part of their meaning are very restrictive with the result phrases they can take.This is due to the constraint that prevents multiple scales to be specified within a single VP and the fact that resultatives either introduce their own scale or specify the scale lexicalized in the verb (Levin and Rappaport Hovav 1995;Wechsler 2005).In verbs that lexicalize scales, resultatives can only be used to specify the verbal scale and thus the range of resultatives is quite restricted (21a).Incremental theme verbs are felicitous with a wider range of resultatives (21b), which indicates that the result phrases are introducing a new scale, and thus the scale is not a lexical property of the verb.Since Romance languages lack resultative constructions like the ones in (21) (Green 1973;Talmy 1985Talmy , 2000;;Mateu 2002;Acedo Matellán and Mateu 2013;a.o.), the test cannot be replicated in Spanish.Nevertheless, as Rappaport Hovav (2008) mentions, the very fact that there are languages in which the association of a scale with a predicate is much more restricted is an argument in itself for not assuming that all verbs that show scalar behavior have scales as part of their lexical meaning.a.We froze the ice-cream {solid / ??blue / ??sweet}.b.We steamed the clothes {dry / clean / stiff}.
(Rappaport Hovav, 2008) A second argument in favor of incremental theme verbs not lexicalizing scales is related to argument realization.In particular, verbs that have a scale as part of their lexical meaning require their theme objects to be realized syntactically, provided that the theme denotes the entity that undergoes the change specified in the scale (22a).By contrast, incremental theme verbs are less dependent on their direct objects and can be used intransitively (22b).a. ??{Oscurecimos / ensanchamos / cerramos / enfriamos} toda la noche.
translated2PL read2PL wrote2PL ate2PL all the night 'We {translated / read / wrote / ate} all night long.' Third, VPs headed by degree achievements allow a full range of degree morphology, whereas incremental theme verbs are much more restricted (Gawron 2007).Degree achievements like acortar 'shorten' are compatible with comparatives (23a) and degree modifiers like demasiado 'too much' (23b) or tanto que 'so much that' with a comparison of degree reading (23c).That is, the sentences with acortar compare the degree to which the paper gets shortened to some other degree.That is not the case with incremental theme verbs such as traducir 'translate'.The sentences in (23) with traducir only have a marginal comparison of events reading (by which, for instance, Elena translated the paper more times than Vicent), but are not an acceptable way to compare the degree to which the paper gets translated.Instead, in order to get that reading, the degree morphology needs to combine directly with the theme argument (24).To sum up, in incremental theme verbs and, consequently, their nominalizations, the scale that tracks the extension of the theme argument that is involved during the course of the event is not lexically encoded in the verb itself but rather is associated with the incremental theme argument.Next, I provide an analysis of composition of the incremental theme verb and its internal argument.

The incremental theme and its composition with the verb
Internal arguments of incremental theme verbs belong to a type of arguments that contribute to describe the event, as argued by Verkuyl (1972); Krifka (1989); Tenny (1994); Ramchand (2008); a.o. 7More concretely, these internal arguments contribute the measuring scale that is homomorphic with the process (as in Krifka 1989).In the case of incremental themes, as mentioned above, the event is correlated with a physical property (extension, volume) of the object they denote (that can be conceptualized as a set of points).The quantization of the referent of the incremental theme corresponds to a boundary for the scale and, consequently, a termination point 7 Not all internal arguments have an influence in the telicity of the predicate.Internal arguments that refer to entities that experience the process or event but do not play a role in delimiting it are, following Ramchand (2008), undergoers.An example of a verb that selects an undergoer is, for instance, mover 'to move'.As shown in (i), the aspectual properties of the event of moving are not affected by the referential properties of its internal argument.Regardless of whether the internal argument is has cumulative (ia) or quantized reference (ib), the verbal predicate has an atelic reading.
Claudia moved chairs for ten minutes b.Claudia movió una silla durante diez minutos.
Claudia moved a chair for ten minutes for the event.Other such arguments that provide a scalar structure that can be mapped to the process denoted by the verb are PPs of directed motion verbs (cross), which represent a locational path, and APs of deadjectival verbs (close), which provide the scale of a gradable property (see, e.g., discussion in Hay, Kennedy, and Levin 1999).Syntactically, I follow Ramchand (2008) in assuming that this kind of elements that provide structure to construct the specific event property (semantic paths) are merged in the complement of Proc, the designated head for process meaning, and the head associated with the event argument (25) (see Ramchand 2008 for details).
The internal argument itself, as a DP, does not denote a scale, but gives rise to one.That is, there needs to be a function at the level of Rheme that maps the mereological structure of the object with a scale along some dimension that needs to be monotonic (Schwarzschild 2002;Ramchand 2008), such as extension.Following scalar analyses of incremental theme verbs (Kennedy 2012;Bochnak 2013), I propose that there is a partitive head that introduces a degree argument that tracks the quantity of the incremental theme affected by the event.For this head I adopt Kennedy's ( 2012 This function takes thus the DP and introduces a degree argument that measures the amount of the object affected by the event.I represent this as the PATH role, that is, path(d, e, x).This differs from the PATH role in Ramchand (2008) in having a degree argument.According to her definition of path (Ramchand 2008: 49-51), this role denotes a relation between an individual (the internal argument) and the event, and ensures that there is a correspondence between the degrees tracking the mereological structure of the object and the event, in the spirit of (Krifka 1989(Krifka , 1992)'s object-event homomorphism.Given that incremental themes can be modified by degree terms (el fuego destruyó 100 ha de la ciudad 'the fire destroyed 100 ha of the city') and degree modifiers (24), and, looking ahead, that they have a contribution to the aspect of the event description, it seems reasonable that the degree argument is accessible for semantic composition.In addition, the presence of a degree argument would tease apart incremental theme verbs from other transitive verbs whose internal argument does not interact with their telicity (see fn. 7), which in Ramchand's (2008) system have the same denotation as the former.
The Proc head is associated with the event argument (Ramchand 2008), so I assume that the verb at Proc denotes a property of events.In order to combine with its internal argument, and since the latter includes a degree argument I use as mode of composition a version of Solt (2015)'s Degree Argument Introduction generalized to events (27).This is a variant of Kratzer (1996) The scale used by the PATH tracks the mereological structure of the internal argument affected by the event.Since at least some part of the object is involved in the event, the scale has necessarily a minimum, which corresponds with the beginning of the event.Depending on the reference of the object, the scale may include a maximum or not.Specifically, objects that denote limited amounts of stuff (quantized) provide a bound for the scale, while objects denoting unlimited amounts of stuff (cumulative) would provide no maximum.In the absence of degree morphology, the degree argument is set (via typeshift or through a null POS morpheme, see Kennedy & McNally (2005); Kennedy ( 2012)) to the contextually most relevant standard.This standard is a minimum or maximum by the principle of Interpretative Economy (Kennedy 2007).The maximum and minimum standards correspond, respectively, with telic and atelic interpretations.Given that the maximum entails the minimum, in the case of verbs with incremental themes that have quantized reference, the maximum-standard interpretation (telic) will be preferred (Kennedy 2012).The minimum standard one (atelic) is still available in some specific contexts, as expected ( 29 In short, the internal argument of an incremental theme verb is generated in the complement of the ProcP and provides the scale that determines the telicity of the verb.To do so, a degree argument that tracks the part structure of the object affected by the event is introduced by a partitive head that combines with the DP.The boundedness of the scale depends on the reference properties of the object.The next two sections elaborate on this analysis of incremental theme verbs to provide an account of syntax and semantics of the nominalizations of these verbs.

The syntax of nominalizations of incremental theme verbs
As discussed in section 4, deverbal nominalizations can have eventive meanings and support arguments.I assume here that the verbal properties of event nominalizations are a consequence of the presence of verbal structure, and therefore that the differences between event and result nominalizations is structural (Hazout 1991;Van Hout and Roeper 1998;Fu, Roeper, and Borer 2001;Alexiadou 2001b;Harley 2009;a.o.)As a consequence, the different types of nominalizations (event, state, quality, result) can be derived from the layers present under the nominalizer and the position this categorizer occupies in the configuration (Harley and Noyer 1998;Alexiadou 2001b;Alexiadou, Iordăchioaia, and Schäfer 2011;Borer 2003Borer , 2005;;Ramchand 2008;Fábregas 2016;a.o.).
The structure I propose for nominalizations of incremental theme verbs in -ción is the following (30).
Event nominalizations of incremental theme verbs denote events.Since their telicity is built up from the referential properties of the theme argument (see ( 20)), just as in the verbal predicate, the nominalization includes a ProcP that introduces the event argument, and a rheme path as its complement, which introduces the scale that measures out the event (see above, section 5.1.2).
The presence of an AspectP depends on the participial morpheme being part of the nominalization, as in nominalizadions in -ado/-ada (Fábregas 2010, 2016: IV.8.5) and arguably -(d)ura (see Santiago Lacuesta and Bustos Gisbert 1999; Fábregas 2016: IV.5.1).By contrast, -ción does not interact with the verbal structure and its role is limited to nominalize it. 8The nominalizer -ción is thus a nominal head that gives nominal category to the array of verbal projections and has no semantic effect.
Event nominalizations also display nominal features, such as gender (-ción and -(d)ura nominalizations are feminine; nominalizations with the past participle can have either masculine or feminine gender) and adjectival modification (31).This is associated with the presence of nominal layers, such as ClassP, in charge of gender (Picallo 2006) To sum up, the syntactic structure of event nominalizations of incremental theme verbs consists of verbal projections, including at least a ProcP and its rheme complement, and nominal projections.Next section puts forward a scalar account of the nominalizations under discussion.

The semantics of nominalizations of incremental theme verbs
Event nominalizations of incremental theme verbs denote events.I assume that the semantic composition follows the order of syntactic composition.Consequently, the semantic tree in (32) parallels the structure in (30), with some layers omitted for clarity.
First, the internal argument combines with the partitive head that introduces the scale mapping the part structure of the object to the development of the event.That is, PART inc takes an individual-denoting DP and turns it into a gradable properties of events ( , , ,  ).In this case, the rheme denotes de property of the novel being affected by an event e to a degree d (33a).At the level of PART inc P and before proceeding, the part y of the object x involved in the event is existentially closed (33b) (cf.( 26b)).In order to be consistent with Ramchand's (2008) notation, the denotation of the rheme (PART inc P) is represented using the path function, as in (33c), which Then the internal argument combines with the verbal predicate (which, I assume, starts as a root that is verbalized (Chomsky 1970;Picallo 1991;Alexiadou 2001b;Borer 2003;a.o.)), via DAI (see ( 27)), so the two event arguments are identified.The result is a gradable property of events, the property of the novel being translated to degree d (34).

traduc-la novela = 𝜆𝑑𝜆𝑒[path(d, the-novel, e) & translate(e)]
Up to this point, the structure is verbal.An InitP that introduces the initiator of the event would be projected on top, but this is omitted because it has no consequence for the analysis.The categorizer -ción then attaches above the whole verbal structure and nominalizes it.This has no semantic effect, as the string still denotes a property of events ( 35), but has syntactic consequences, such as the assignment of case 9 or the licensing of adjectives.

traducción (de) la novela = 𝜆𝑑𝜆𝑒[path(d, the-novel, e) & translate(e)]
For nominalizations, I suggest that the degree argument gets saturated at this stage, either by degree terms and quantifiers, or by a type shift in lack thereof.If that is the case, the value of the degree argument is set to a contextually relevant standard, a minimum or a maximum (Kennedy 2007(Kennedy , 2012)).The scale used by the predicate comes from the mereological structure of the internal object.As a consequence, if the theme argument has quantized reference, the scale has a maximum and a minimum; by contrast, if the internal argument is cumulative, the scale has a minimum but no maximum.This is consistent with the aspectual properties of event nominalizations of incremental theme verbs, which preserve the aspect of their verbal base (20) (see discussion around ( 29)).
The mapping between the referential properties of the theme argument onto a scale that is homomorphic with the event predicts that nominalizations of incremental theme verbs with quantized objects as internal arguments can receive atelic readings in some specific contexts.This is actually the case.In both examples in (36) the nominalization denotes an atelic event, even though the internal argument has quantized reference.b.Si la destrucción de la ciudad continúa, no quedará nada if the destruction of the city continues NEG remain.FUT nothing en pie. in foot 'If the destruction of the city continues, nothing will remain.' After the degree argument is saturated, the determiner turns the property of events into a unique event.The final semantics for the whole DP is thus as in (37), where the degree argument is set to the standard, which, in this case, is its maximal value, 1.

la traducción (de) la novela = 𝜄𝑒[path(1, the-novel, e) & translate(e)]
To sum up, event nominalizations of incremental theme verbs have been analyzed from a degree-based perspective.In particular, following Kennedy (2012), a partitive head introduces de scale that maps the parts of the internal argument to the progress of the event.The internal theme then composes with the verb and contributes to its aspectual configuration.If the object has limits (if it is quantized) the scale that measures out the event has an upper endpoint, which constitutes the standard for the predicate, and thus the event is telic.Otherwise, if no maximal bound for the scale is present, the value of the degree argument is set to the minimum of the scale, and the event is atelic.The degree argument is saturated at the NP level for nominalization, as will be discussed in the next section.
This semantic composition has a syntactic correspondence, based on Ramchand's (2008) analysis of incremental theme verbs.The internal argument is syntactically a rheme path, and it is merged in the complement of the Proc phrase.Above the verbal layers, which provide the event semantics, a nominal head (-ción in this case) nominalizes the structure, but has no semantic effect.Now that the syntax and semantics of the nominalizations are set, we are ready to address the analysis of ACs.

Adjectives of completeness are maximality modifiers
Section 4 showed that ACs behave as maximality modifiers when combined with nominalizations of incremental theme verbs and that they are sensitive to the aspect of the predicate.In this section, I put forward an analysis of ACs as maximality modifiers of nominalizations of incremental theme verbs that covers those two facts.In particular, I argue that they are degree morphology able to saturate the degree argument associated with those nominalizations that tracks the parts of the theme affected by the event.

Analysis of ACs as maximizers
As argued in the last section, event nominalizations of incremental theme include a degree that tracks the development of the event based on the quantity of the incremental theme affected.My proposal is that ACs are degree morphology that can saturate that degree.As shown in (38a), completo 'complete' takes a gradable property of events and sets the degree of affectedness of the object during the course of the event to its maximal value.If the rheme denotes an object with quantized reference (e.g., la novela 'the novel'), the scale has a maximum and ACs are acceptable.By contrast, if the reference of the object is cumulative (novels 'novelas'), the scale has no upper bound, so no maximum is defined.As a consequence, the max function in the semantics of ACs is undefined and the modifiers are not acceptable.ACs combine with the nominalizations above NP, and I assume that they are placed in one of the Note that the semantics for la completa traducción de la novela 'the complete translation of the novel' in (38b) is exactly the same as the semantics given for the same DP without the modifier in (37).The semantics of ACs differs from the type shift that closes the degree argument in the absence of degree morphology in that ACs set the value of the degree necessarily to the maximum in the scale, while the type shift sets the value to the standard of the scale, which can be a maximum or a minimum.In other words, by means of the type shift, the event description can be either telic or atelic, whereas the presence of the AC forces, and emphasizes, the telic reading.Therefore, the reason why the examples in (39) are not grammatical is that the presence of the ACs excludes the atelic interpretation (cf.( 36 The semantics for ACs are identical to the semantics for adverbs of completeness.The previous sections showed that ACs modify the event denoted by the nominal with the same effect adverbs of completeness have when modifying the VP.It is thus desirable that both modifiers have the same semantics, that is, that are the same modifier, their morphological expression being a consequence of their merge site.In other words, if the modifier merges in the VP, it adopts adverbial morphology, and if it merges in the NP, it takes adjectival morphology (see Alexiadou 2001bAlexiadou , 2009;;Fu, Roeper, and Borer 2001;Borer 2013;López 2015).
The analysis can be extended to other modifiers affecting the extent to which an event is realized, such as parcial(mente) 'partially'.This modifier sets the degree of development of the event to a value lower than the maximum of the scale.As a consequence, it resembles adjectives and adverbs of completeness in requiring that the event be telic, that is, in the case of incremental theme verbs and their nominalizations, that the internal argument have quantized reference (40) (cf.Piñón 2005). 10a. la parcial traducción de la novela the partial translation of the novel b. ??la parcial traducción de novelas the partial translation of novels In conclusion, ACs modifying event nominalizations of incremental theme verbs are aspectual modifiers.In a scalar accout of aspect as the one adopted here, they are maximality modifiers that set the degree of development of the event to its maximum.Next section examines some consequences of the analysis.

Consequences of the analysis
ACs, as degree modifiers of event nominalizations, form a especial subclass of adjectives.They are often included in the class of adverbial adjectives (see, e.g., Demonte 1999; Picallo 2002; Real Academia Española and Asociación de Academias de la Lengua Española 2009), because their modification is closer to the type of modification adverbs perform and usually can be paraphrased with an adverb (as in (1) above).This sets them apart from regular adjectives.In this section I discuss some facts regarding the position and distribution of ACs that confirm that they are not to be analyzed as qualitative, regular adjectives.
First, if ACs are degree modifiers of eventive nominalizations of incremental theme verbs, they should not be subject to the correlation between position with respect to the noun and interpretation that applies to regular adjectives in Spanish and other Romance languages.This is borne out.Compare ( 41) to (17) (repeated here as( 42)).While (42b) refers to the translation having all its parts (restrictive, R) and (42a) to the translation having all the parts the speaker deems necessary (being thus comprehensive) (non-restrictive, NR), the difference is not present in (41a) and (41b).Both examples are equivalent and refer to the completeness of the translating event.
a. La completa traducción de la novela tuvo lugar en abril.
the complete translation of the novel had place in April 'The complete translation of the novel took place in April.' b.La traducción completa de la novela tuvo lugar en abril.
the translation complete of the novel had place in April 'The complete translation of the novel took place in April.' a. La completa traducción está encima de la mesa.
NR the complete translation is on of the table 'The complete translation is on the table.' (=the comprehensive translation) b.La traducción completa está encima de la mesa.
R the translation complete is on of the table 'The complete translation is on the table.' (=the whole translation) In addition, degree uses of ACs are predicted to be restricted to attributive position to be able to access the degree argument, since they need the degree modifier to be adjacent, but ACs are at times found in predicative position with nominalizations of incremental theme verbs.However, these uses of ACs are suspicious as degree uses for two reasons.First, there is a wide range of acceptability from example to example (43).And second, when ACs refer to the result state of a process, it only appears with the copula estar,11 but that is never the cases in the examples below (44).a. La traducción de la novela fue ??{completa / total} the translation of the novel was ser complete total 'The translation of the novel was {complete / total}.' b.La destrucción urbana (en Siria) fue {completa / ?total}. the destruction urban in Siria was ser complete total 'The urban destruction (in Siria) was {complete / total}.' a. La traducción de la novela {fue / *estuvo} completa.
the translation of the novel was ser was estar complete 'The destruction of the city was complete.'b.La destrucción de la ciudad {fue / *estuvo} completa.
the destruction of the city was ser was estar complete 'The destruction of the city was complete.'Therefore, ACs in ( 43) and ( 44) appear to be operating on something different from the degree of development of the event.They may be modifying a result nominalization, as the possibility of having a relational adjective such as urbana 'urban' (43b) (which favors the result reading, see Bosque and Picallo 1996) points to.
Also, the configuration proposed has two advantages over analyses that saturate the open degree argument at the level of the incremental theme argument (Kennedy 2012;Bochnak 2013).First, it accounts for the fact that adverbs of completeness are modifiers internal to the VP.Two facts constitute evidence that these adverbs are VP modifiers, namely, they are restricted to a postverbal position (45) (Jackendoff 1972;Piñón 2005) and they take narrow scope with respect to the two readings of again (46) (Piñón 2005;see Dowty 1979;von Stechow 1996).Especifically, in both readings in (46) the adverb is in the scope of de nuevo 'again'.That is, the example cannot mean either that Paula had translated the novel before, but this time she translated it completely (repetitive reading with wide scope of completely) or that the novel was translated before but this time it was translated completely (restitutive reading with wide scope of completely).Second, saturating the degree argument at the level of VP (or NP in the case of the nominalizations) is consistent with the fact that the adverb and the incremental theme argument do not form a constituent.In pronominalizing the direct object, the modifier stays outside.That is, [completamente la novela] 'completely the novel' cannot be replaced by the accusative pronoun la (47).The same is true of ACs and the nominalizations of incremental theme verbs.As shown in (48) [completa de la novela] 'complete of the novel' cannot be substituted by a possessive pronoun.Finally, ACs, as degree modifiers, are not expected to combine with event nominals that do not have aspect, because aspect is what provides the relevant degree argument to be saturated by ACs.That is the case of simple event nominals (Grimshaw 1990;Borer 2003).As expected, ACs only receive their literal reading with them.In (49a), the AC receives a non-restrictive interpretation whereby the class is said to have had all the elements a class must have, while in (49b), it is the whole class, with all its elements, what took place in April.Note also that only completo is acceptable, pointing to the fact that total is only an aspectual modifier.However, there are other adverbial adjectives that interact with the event itself, rather than with its aspectual properties.For example, frequency adjectives (Stump 1981;Schäfer 2007;Gehrke and McNally 2015), which modify an event argument, occur both with event nominalizations of incremental theme verbs (50a) and with simple event nominals like clase 'class' (50b).To sum up, this section has discussed the consequences of an analysis of ACs as maximizers.In particular, as adverbial adjectives, degree uses of ACs are expected to have a different behavior from regular adjectives.To begin with, they do not change their meaning depending on their position with respect to the noun and do not seem to appear in predicative position.The position at which ACs and adverbs of completeness have been proposed to be situated is also consistent with their syntactic relation to the internal argument of the event predicate.Finally, ACs have been compared to frequency adjectives to show that they do not modify an event directly, but a degree argument.

Conclusion
This paper has argued that ACs modifying event nominalizations of incremental theme verbs are maximality modifiers.In order to do so, I have first showed that ACs are sensitive to the aspect of the event nominalization and have the properties of maximizers.Then, incremental theme verbs and their nominalizations have been analyzed as denoting gradable properties of events.The scale, however, is not present in the verb, but is provided by the internal argument through the presence of a functional head that maps the extension of the object to the development of the event.
By adopting a scalar approach to aspectual composition, a unified account of the degree uses of adverbs and adjectives of completeness has been provided.This highlights the correspondence between aspect and degree modification across categories and contributes to the understanding of it in the nominal domain.Without leaving the subclass of adjectives related to aspect, the examinations of their interaction with some other nominalizations of verbs of variable telicity would contribute further to pin down the ways in which aspect is composed and transferred under nominalizations and in which nominal modifiers interact with it.For instance, two candidates for this are event nominalization of degree achievements (el completo {cierre / ??ensanchamiento} de la grieta.'the complete {closing / ??widening} of the crack), whose telicity depends partly on the scale structure used by the source adjective (for the verbs, see Hay, Kennedy, and Levin 1999; for an analysis, see Masià 2017) and of directed motion verbs (el completo {aterrizaje / ??ascenso} del avión 'the complete {landing / ascending} of the plane'), where the path the object moves along determines telicity (for the verbs, see, e.g., Zwarts 2005)

(
Totally) open scale ○----○ (tall, wide, beautiful, wise) Lower closed scale •----○ (dirty, impure, dangerous) Upper closed scale ○----• (dry, clear, free, dark) (Totally) closed scale •----• (open, closed, visible, full) Maximizers like completamente 'completely' or totalmente 'totally' are modifiers sensitive to the scale structure of the adjective.In particular, they only occur with upper-and totally-closed scale adjectives (3) (Rotstein & Winter 2004; Kennedy & McNally 2005).a. completamente {seco / oscuro / libre / cerrado / visible / lleno} completely dry dark free closed visible full b.??completamente {alto / ancho / bello / impuro / sucio} completely tall wide beautiful impure dirty fila. in the first row 'The plane is completely full; there is an empty seat in the first row.' b.El avión está casi completamente lleno.the plane is almost completely full c.El avión está completamente lleno, excepto un asiento en la the plane is completely full except a seat in the primera fila.first row 'The plane is completely full, except for a seat in the first row.' ; Lenepveu 2013).a. Roberto dibujó completamente un círculo.Roberto drew completely a circle 'Roberto completely drew a circle.'ACCOMPLISHMENT b. ?? Roberto sabe francés totalmente.Roberto knows French totally 'Roberto totally knows French.' STATE c. ?? Roberto caminó completamente.Roberto walked completely 'Roberto completely ran.' ACTIVITY d. ?? Roberto advirtió totalmente el nuevo cuadro.Roberto noticed totally the new painting 'Roberto totally noticed the new painting.'ACHIEVEMENT Second, adverbs of completeness behave as maximizers when modifying VPs.First, they entail that the event has come to an endpoint.As a consequence, asserting that the novel can be further translated constitutes a contradiction (8a) (cf.(5a)).Second, they are compatible with casi 'almost' (8b) (cf.(5b)).And third, they accept exceptive phrases (8c) (cf.(5c)). 2 a. #Paula tradujo completamente la novela, pero quedan dos Paula translated completely the novel but remain two capítulos por traducir.chapters to translate 'Paula completely translated the novel, but there are two chapters left to be translated.' b.Paula tradujo casi completamente la novela.Paula translated almost completely the novel 'Paula completely translated the novel.' c.Paula tradujo completamente la novela, excepto dos páginas.Paula translated completely the novel except two pages 'Paula completely translated the novel, except for two pages.' tradujo la novela {en / ??durante} una hora.Paula translated the novel in for a hour 'Paula translated the novel {in / ??for} an hour.' TELIC b.Paula tradujo novelas {??en / durante} una hora.Paula translated novels in for a hour 'Paula translated novels {??in / for} an hour.' ATELIC As expected, maximizers only combine with the telic versions of incremental theme verbs (10).a. Paula tradujo totalmente la novela.Paula translated totally the novel 'Paula totally translated the novel.b. ??Paula tradujo totalmente novelas.Paula translated totally novels 'Paula totally translated novels'.
a. El avión cruzó el Atlántico {en / ??durante} siete horas.the plane crossed the Atlantic in for seven hours 'The plane crossed the Atlantic {in / ??for} seven hours.' TELIC b.El avión ascendió {??en / durante} cinco minutos.the plane ascended in for five minutes 'The plane ascended {??in / for} five minutes.'ATELIC In these cases as well, adverbs of completeness are only acceptable with the telic versions of degree achievements (13a) and directed motion verbs (13b).a. La grieta se {cerró / ??ensanchó} completamente. the crack se closed widened completely 'The crack {closed / ??widened} completely.' b.El avión {cruzó completamente el Atlántico / ??ascendió the plane crossed completely the Atlantic ascended completamente}.completely 'The plane completely {crossed the Atlantic / ascended}.' a. La construcción ??(del puente) llevó dos años.the building of.thebridge took two years 'The building of the bridge took two years' b.La construcción (*del puente) es de piedra.the building of.thebridge is of stone 'The building (*of the bridge) is made of stone.' c.La clase (de matemáticas) (a los alumnos de tercero) llevó mucho the class of maths to the students of third took a.lot tiempo.time 'The (math) class (to the third-year students) took a very long time.' 4 a.La construcción del puente en tres horas tuvo lugar en abril.the building of.thebridge in three hours had place in April 'The building of the bridge in three hours took place in April.' b.La construcción *(en tres horas) es de piedra.the building in three hours is of stone 'The building *(in three hours) is made of stone.' postnominal position (17b), as expected from qualitative (non-aspectual) adjectives (e.g., Demonte 1999; Alexiadou 2001a).
a. Elena {acortó / ??tradujo} el artículo más que Vicent.Elena shortened translated the article more than Vicent 'Elena {shortened / translated} the article more than Vicent did.' b.Elena {acortó / ??tradujo} mucho el artículo Elena shortened translated a.lot the article 'Elena {shortened / translated} the article a lot.' c.Elena {acortó / ??tradujo} tanto el artículo que no quedó Elena shortened translated so.much the article that NEG remained nada que hacer para Vicent.' nothing to do for Vicent 'Elena {shortened / translated} the article so much that Vicent didn't have anything left to do.' a. Elena tradujo más del artículo que Vicent.Elena translated more of.thearticle than Vicent 'Elena translated more of the article than Vicent did.' b.Elena tradujo mucho del artículo Elena translated a.lot of.thearticle 'Elena translated a lot of the paper.' c.Elena tradujo tanto del artículo que no quedó nada Elena translated so.much of.thearticle that NEG remained nothing que hacer para Vicent.to do for Vicent 'Elena translated so much of the paper that Vicent didn't have anything left to do.' ) incremental partitive head Part inc .This head, by means of the measure of change function partof Δ , gives a measure of how much (what part y) of the internal argument x changes as a result of its participation in the event e (26a).For instance, if the internal argument is la novela 'the novel', [Part inc la novela] has the denotation in (26b), where d is the amount of the novel involved in the event.a. PART ./0= .partof  ()()() =  (Kennedy 2012:119) b.PART ./0la novela = .partof  (the-novel)()() = 's Event Identification that is used to combine the functional head that introduces degrees in the semantics of the NP with the NP itself(Abney 1987;Zamparelli 1996;Schwarzschild 2006; a.o.See Solt 2015 for details).Degree Argument Introduction (DAI) generalized to events: If α is a branching node, {β,γ} are the set of α's daughters, and β = λe.P(e), γ = λdλe.Q(e)(d), then α = λdλe.[P(e)∧ Q(e)(d)] (based on Solt 2015: 237)By this rule, the event argument of the verb and that of the internal argument are identified.At the level of Proc', then, the denotation of the verb and its direct object is as in (28).That is [traducir la novela] 'translate the novel' denotes a gradable property of events.In particular, it denotes the property of translating the novel to a degree d. traducir la novela = [path(d, the-novel, e) & translate(e)] ) (see also Arche 2014; Martin 2017).a.I ate Mr Unagi for a few minutes, then decided to switch to tofu.(Kennedy, 2012) b.Paula tradujo la novela durante dos horas antes de hacer yoga.Paula translated the novel for two hours before of do yoga 'Paula translated the novel for two hours before doing yoga.' , and NumberP.a. La {cuidadosa / repentina} traducción de la novela tuvo lugar en the.F careful.F sudden.F translation.F of the novel had place in abril.April 'The {careful / sudden} translation of the novel took place in April.' b.El {cuidadoso / constante} bordado de la colcha empezó the.M careful.M constant.M embroidering.M of the duvet started en abril. in April 'The {careful / constant} embroidering of the duvet started in April.' returns the degree d to which the object la novela is affected over the course of the event e. a. PART ./0la novela = .[partof  (the-novel)(y)(e) = d] b.PART ./0P = .∃y[partof  (the-novel)(y)(e) = d] c.PART ./0P = .[path(d, the-novel, e)] a. La traducción de la novela durante media hora la convenció de the translation of the novel for half hour ACC convinced of que era una gran obra.that was a great work 'The translation of the novel for half an hour convinced her of the quality of the work.' 9 For discussion on the position in the NP at which the arguments of the verb move after nominalization, see Adger (2013: §4); Fábregas (2016: §V.1.4).
functional projections where attributive adjectives are located in the extended NP (see, e.g., Demonte 2008; Cinque 2010).The final semantics for the nominalization and the modifier is as in (38b).a. AC = λg <d,<v,t> λe.∃d[d = max(S g ) & g(d)(e)] b. la completa traducción (de) la novela = [path(1, the-novel, e) & & translate(e)] )). a. *La completa traducción de la novela durante media hora la the complete translation of the novel for half hour ACC convenció de que era una gran obra.convinced of that was a great work 'The complete translation of the novel for half an hour convinced her of the quality of the work.' b. *Si la total destrucción de la ciudad continúa, no quedará if the total destruction of the city continues NEG remain.FUT nada en pie.nothing in foot 'If the total destruction of the city continues, nothing will remain.' a. (*Completamente) Paula tradujo (completamente) la novela completely Paula translated completely the novel (?completamente).completely 'Paula translated (completely) the novel.' b. (*Parcialmente) el incendio destruyó (parcialmente) la ciudad partly the fire destroyed partly the city (?parcialmente).partly 'The fire destroyed partly the city.' Paula tradujo completamente la novela de nuevo.Paula translated completely the novel again a. REPETITIVE: once more, Paula translated completely the novel.b.RESTITUTIVE: Paula translated the novel completely so it got completely translated again.
a. #Paula la tradujo.Paula ACC translated 'Paula translated it.'b.Paula la tradujo completamente.Paula ACC translated completely 'Paula translated it completely.'a. #Su traducción its translation b.Su traducción completa its translation complete 'its complete translation' a. La {completa / *total} clase sobre Alphaville tuvo lugar en abril.the complete total class on Alphaville had place in April 'The {complete / total } class on Alphaville took place in April.' (=comprehensive, NR) b.La clase {completa / *total} sobre Alphaville tuvo lugar en abril.the class complete total on Alphaville had place in April 'The {complete / total } class on Alphaville took place in April.' (=the whole class, R) a. La traducción {semanal / periódica} de noticias sobre Siria es esencial.the translation weekly periodic of news on Syria is essential 'The {weekly / periodic} translation of news about Syria is essential.'b.La clase {semanal / periódica} sobre Alphaville nos gustó.the class weekly periodic on Alphaville DAT.1PL liked 'We liked the {weekly / periodic} class on Alphaville.' Dowty 1979;Hay, Kennedy & Levin 1999;Winter 2006;Kearns 2007;Kennedy & Levin 2008)provided by the source adjective (seeDowty 1979;Hay, Kennedy & Levin 1999;Winter 2006;Kearns 2007;Kennedy & Levin 2008).If the source adjective uses a closed scale (such as in the case of cerrado 'closed'), the verb (cerrar 'close') has a telic interpretation (11a).If the scale lexicalized by the source adjective is open (such as in the case of ancho 'wide'), its derived verb (ensanchar 'widen') has a default atelic reading (11b).