Attributive verb
In linguistics, an attributive verb is a verb or verb form that directly modifies a noun by expressing an attribute or property of it, functioning similarly to an attributive adjective rather than serving as the independent predicate of a clause.[1] This contrasts with predicative verbs, which link to their subject via a copula or stand alone in main clauses, and reflects a syntactic role where the verb integrates into a noun phrase to provide descriptive information about the head noun.[1] The concept is particularly emphasized in functional grammar frameworks, where attributive verb forms are classified as dependent structures that both predicate subordinate clauses and attribute qualities to nominal elements, avoiding reliance on finite/non-finite distinctions in favor of functional analysis.[1]
Attributive verbs manifest differently across languages, often involving specific morphological forms or participial constructions. In Japanese grammar, verbs conjugate into the rentaikei (attributive form) to modify nouns directly, as in hashiru kuruma ("running car"), where the verb hashiru ("to run") attributes the action to the noun kuruma ("car") without requiring a relative clause particle.[2] This form is essential for embedding verbal descriptions within noun phrases, a hallmark of Japanese syntax that parallels adjectival modification.[2] Similarly, in some Bantu languages like those of the Narrow Grassfields group, gradable verbs and adjectives can take attributive roles to express properties such as size or quality, integrating into noun phrases via verbal morphology that blurs the line between verbs and adjectival predicates.[3]
In English, attributive verbs are commonly expressed through participles, which function adjectivally to attribute qualities to nouns, such as boiling water (where boiling describes the state of water) or the elected official (where elected indicates the status of official).[4] These constructions, known as attributive participles, agree with the noun they modify and can appear prenominally without linking verbs, though they often derive from transitive or intransitive verbs to convey ongoing or completed actions.[4] This usage highlights the flexibility of English verbal forms in attributive positions, though constraints like agentivity or aspect can limit their distribution compared to full adjectival phrases.[5]
Conceptual Overview
Definition
An attributive verb is a verb that functions attributively by directly modifying a noun, expressing one of its attributes or properties without serving as the main predicate of a clause.[6] This construction allows the verb to describe or qualify the noun in a manner analogous to an attributive adjective, embedding the verbal information within a nominal phrase rather than asserting an independent action or state.[7]
Key characteristics of attributive verbs include their syntactic dependency on the modified noun and their typical restriction to non-clausal environments, distinguishing them from predicative verbs that form the core of a clause and link to a subject via agreement or tense marking.[6] In many languages, attributive verbs exhibit specialized morphology, such as infinitival or participial endings, to signal this modifier role, though the exact form varies cross-linguistically.[7]
The term "attributive verb" derives from the Latin attributus (past participle of attribuere, "to assign"), entering English via French attributif in the early 17th century to describe modifiers assigning qualities, initially applied to adjectives in traditional grammar.[8] Its extension to verbs emerged in 19th-century comparative linguistics, particularly in analyses of non-Indo-European languages where verbs routinely modify nouns without relativizers, as documented in early Western grammars of Japanese from the 1880s.[9]
A basic typology distinguishes finite attributive verbs, which retain tense and agreement features while modifying nouns, from non-finite ones, which lack independent tense and often appear as participles or infinitives; this variation highlights language-specific strategies for integrating verbal modification into nominal syntax, sometimes overlapping with relative clause formations.[7][6]
Relation to Other Grammatical Constructions
Attributive verbs frequently underpin relative clause constructions, particularly in reduced or non-finite forms where no relative pronoun appears, allowing the verb to directly modify the head noun either internally (within the noun phrase) or externally (adjacent to it).[10] This connection arises because attributive verbs encode restrictive or non-restrictive modifications akin to full relative clauses, but with aspectual or temporal constraints that limit their equivalence, as analyzed in syntactic frameworks like Role and Reference Grammar, where attributive phrases are represented as relative clause junctures.[11]
Participles represent a specific subtype of non-finite attributive verbs, derived through morphological processes that adapt the verb stem for adjectival-like modification, such as present participles indicating ongoing action (e.g., via -ing affixation) or past participles denoting completion (e.g., via -ed or -en).[10] The derivation can occur in the lexicon for adjectival participles, which lack verbal complements, or in the syntax for verbal ones, which permit them, leading to a gradient analysis rather than a strict binary distinction.[10]
In contrast to adjectives, which typically express static qualities or inherent properties of nouns, attributive verbs introduce dynamic attributes reflecting actions, processes, or temporary states, thereby expanding the semantic range of nominal modification.[12] This distinction blurs in languages lacking a dedicated adjective class, where property concepts are often conveyed through stative or attributive verbs, effectively merging verbal and adjectival functions to describe qualities like size or color via verbal constructions.[12]
Syntactically, attributive verbs occupy pre-nominal positions in head-final languages, where they precede the noun they modify, or post-nominal positions in head-initial languages, following the noun; this variation aligns with broader typological patterns of phrase order.[10] For illustration, a basic constituent structure in a head-initial language might appear as:
[NP](/page/NP)
├── N (head [noun](/page/Noun))
└── Attributive Verb (post-nominal modifier)
[NP](/page/NP)
├── N (head [noun](/page/Noun))
└── Attributive Verb (post-nominal modifier)
Whereas in a head-final language:
NP
├── Attributive Verb (pre-nominal modifier)
└── [N](/page/N+) (head noun)
NP
├── Attributive Verb (pre-nominal modifier)
└── [N](/page/N+) (head noun)
These positions influence the verb's ability to host complements: pre-nominal attributive verbs are typically reduced and complement-free, while post-nominal ones allow fuller clausal embedding.[10]
Usage in English
In English, attributive verbs primarily take the form of non-finite participles that function adjectivally to modify nouns, including present participles ending in -ing and past participles typically ending in -ed, -en, -d, -t, -n, or irregular variants.[13] Present participles denote ongoing or characteristic actions, as in the running man, where "running" describes the man's habitual or current activity.[13] Past participles indicate completed actions or resultant states, such as the broken vase, implying the vase's state after being broken.[13] Gerunds, also ending in -ing but functioning nominally, appear in limited attributive roles, often as determiners in compound noun phrases like swimming pool, where "swimming" specifies the type of pool without verbal tense or aspect.[14]
Deverbal adjectives derive from verbs through suffixation, particularly -ing for active or process-oriented senses and -ed for passive or stative ones, behaving fully as adjectives without retaining verbal properties like agentivity.[15] For instance, exciting story uses the -ing form from "excite" to describe the story's effect on the reader, distinct from the verbal "The story excites the audience."[15] Similarly, painted door employs the -ed form from "paint" to denote a completed state, contrasting with the active verbal use in "She painted the door."[15] These forms integrate seamlessly into noun phrases, allowing modification by adverbs or intensifiers, as in the highly exciting story.[15]
Attributive verb constructions modify nouns within larger sentences, often contrasting with predicative uses to highlight scope. In The cat sitting on the mat fled, "sitting on the mat" is a present participial phrase attributively modifying "cat," specifying its position at the time of fleeing, unlike the finite predicative The cat sat on the mat, which asserts the action independently.[13] The modification scope is restricted to the head noun, as in the broken vase on the shelf, where "broken" applies only to "vase" and not the entire phrase.[13]
Historically, participial forms in Old English were inflected for case, number, and gender when used attributively, as in lifigendne ("living," accusative masculine singular) modifying a noun like Apollonium in Apollonium lifigendne ("Apollonius alive").[16] Past participles, such as geborenum ("born," dative plural), similarly agreed with nouns in phrases like geborenum in descriptions of origins.[16] By Middle English, adjectival inflections were lost, and past participle suffixes like -n or -d simplified; present participles merged with -ing forms around 1500, leading to modern uninflected, adjectival uses with restrictions on complex verbal arguments.[16] Fossilized forms, such as boiling water, preserve participial adjectival status from earlier periods, denoting a characteristic state without implying ongoing action in contemporary contexts.[17]
Syntactic Constraints
In English, attributive verbs primarily manifest as non-finite participial forms, which are constrained to pre-nominal position when modifying a noun directly, as in "the running water" or "a broken vase." While primarily pre-nominal, attributive participles can appear post-nominally in reduced relative clauses without a relative pronoun or finite verb, as in "the water running from the tap," though standalone "*the water running" may require context or additional modifiers for clarity.[18][19]
These participial forms exhibit no subject-verb agreement, as they lack finiteness and thus do not inflect for person, number, or tense in relation to the modified noun; for instance, "the escaping prisoners" uses the same form regardless of plurality. Tense is frozen in the participial morphology, typically conveying a resultative or ongoing state without explicit temporal marking, which can lead to ambiguities such as "the escaping prisoner" potentially implying either a current flight or a recently completed one.[18][20][21]
Attributive participles in English are typically interpreted as stative or resultative, derived from dynamic (often telic) verbs but denoting a state or result rather than an ongoing event, as in "the solved problem." Eventive interpretations are generally incompatible with attributive positions. This usage often involves a shift from verbal to adjectival function, emphasizing state over event. Participles from stative verbs are possible, though less common without modifiers, as in "a well-known fact." Modals and auxiliaries are prohibited in attributive position due to their inability to inflect as participles, making constructions such as "*the will go man" or "*the being seen event" ungrammatical.[20][18]
Exceptions occur in informal or innovative usages, where non-standard forms like "a no-go area" employ infinitive-like structures for attributive modification, bypassing traditional participial constraints. Dialectal variations may occasionally permit flexible placement of participles in certain regional varieties, such as extended post-nominal uses in some American English dialects, though these remain marginal and context-dependent.[22][23]
Usage in Japanese
Syntactic Integration
In Japanese syntax, attributive verbs serve as the predicates of noun-modifying clauses that directly precede the head noun, forming attributive phrases without relative pronouns or complementizers. This core mechanism allows a finite verb to embed a full clause as a modifier, as in aruita hito ("the person who walked"), where aruita is the past tense form of the verb aruku ("to walk"). The construction relies on juxtaposition and contextual cues for interpretation, reflecting Japanese's head-final structure.
These noun-modifying clauses embed tense, aspect, and polarity markings on the attributive verb, mirroring their use in independent predicative clauses, while omitting politeness forms such as -masu.[24] The entire clause functions as a unified modifier, equivalent to an adjectival phrase, with no explicit markers of subordination.[24] Unlike non-finite relative clauses in languages such as English, Japanese attributive constructions consistently use finite verb forms.[24]
Japanese's strict subject-object-verb (SOV) word order enforces a head-final arrangement, positioning the attributive verb immediately before the noun it modifies, with inflections identical to those in predicative contexts in modern Japanese. For instance, watashi ga kinō atta hito translates to "the person I met yesterday," where atta carries past tense and the subject marker ga integrates into the clause.[24]
This syntactic flexibility supports nested modifications, enabling complex embeddings within a single phrase. An example is tabeta neko ga nomu miruku ("the milk that the cat that ate drinks"), where the inner clause tabeta neko ("the cat that ate") modifies neko, and the outer verb nomu ("drinks") governs the entire structure with the subject particle ga linking the embedded subject.[24] Such nesting highlights the clause-like nature of attributive phrases, allowing recursive modification without additional morphological cues.
Morphophonological Aspects
In modern Japanese, verbs in attributive positions, such as relative clauses modifying nouns, inflect using the plain forms to indicate tense and aspect without dedicated attributive morphemes. For instance, the past tense is formed with the -ta suffix, as in tabeta neko ("the cat that ate"), where tabeta derives from the verb taberu ("to eat") via stem tabe- plus -ta. This plain past form parallels the finite past but omits polite markers, ensuring the verb integrates directly with the following noun without additional phonological adjustments beyond standard suffixation.[25]
Phonological adaptations in attributive constructions are minimal in contemporary Japanese, lacking vowel harmony and relying instead on occasional elision or liaison in rapid speech or compounds. For example, in verb-noun sequences like aruita michi ("the road that was walked"), no elision occurs, but in some lexicalized compounds involving verbal elements, vowel shortening or devoicing may apply for euphony, though these are not systematic rules specific to attributives. Unlike languages with robust harmony systems, Japanese attributive verbs follow the language's general moraic phonology, where high vowels may devoice in unstressed positions but without altering the morphological integrity of the form.
Historically, classical and Old Japanese (8th century) featured distinct attributive (rentaikei) endings separate from conclusive forms, often involving suffixes like -ru added to the verb stem for adnominal use, as in sinuru ("dying," attributive) versus sinu (conclusive) for the verb "to die." This distinction arose from Proto-Japonic morphemes, with the attributive -ru marking modification of nouns in relative-like constructions. Over time, phonological reduction and reanalysis during the transition to Middle Japanese (12th–16th centuries) led to the merger of these forms, where the attributive shape absorbed the conclusive, resulting in the unified plain forms of modern Japanese; for example, the classical quadrigrade verb kaku ("to write") had an attributive kakuru, which simplified without the -ru in contemporary usage.[26][27]
Attributive verbs in Japanese consistently employ plain forms, omitting polite suffixes like -masu even in formal contexts, as politeness is conveyed by the matrix clause's ending. For example, tabemasu ("eat" polite) becomes tabeta in tabeta hito ("the person who ate"), regardless of overall sentence politeness, to maintain morphological simplicity and syntactic embedding. This omission reflects a broader constraint on embedded clauses, where polite morphology is incompatible with non-final positions.[25]
Usage in Bantu Languages
Morphological Marking
In Bantu languages, attributive verbs, which function as modifiers in relative constructions, employ a pre-prefix system to indicate their attributive role. This involves the addition of a class concord prefix, often realized as an augment or pre-prefix, that agrees with the noun class of the head noun it modifies; for instance, prefixes such as mu- for singular human classes or ba- for their plural counterparts ensure morphological harmony within the noun phrase.[28][29]
The agreement mechanics require the attributive verb to index the noun class of the head noun through a relative marker, typically a pronominal prefix that occupies a pre-initial or initial position in the verb complex, signaling the verb's status as a modifier rather than an independent predicate. This prefix functions dually as an argument licenser and an agreement element, integrating the verb syntactically with the modified noun while maintaining tense, aspect, and mood specifications.[28][29]
Unlike non-finite participles in other language families, attributive verbs in Bantu retain their finite nature but acquire dedicated relative morphology via these prefixes, allowing them to inflect for subject agreement and other verbal categories without losing sentential properties. This pattern holds generally across the Bantu family, though variations occur in prefix realization, such as zero marking in certain noun classes or positional shifts relative to tense-aspect markers depending on the language's diachronic stage in the relative agreement cycle.[28][29]
In subgroups like the Narrow Grassfields Bantu languages (e.g., Tswefap), gradable verbs and adjectives can take direct attributive roles to express properties such as size or quality. These integrate into noun phrases via verbal morphology, often without full relative clause structures, blurring distinctions between verbs and adjectival predicates. For example, in Tswefap, a verb like "be big" may directly modify a noun to indicate relative size, functioning attributively similar to an adjective.[3]
Examples from Key Languages
In Luganda, a Bantu language spoken in Uganda, attributive verbs in relative clauses are formed by prefixing the noun class agreement marker to the verb, often without an overt subject prefix inside the relative, while maintaining agreement with the head noun. For instance, the phrase abasajja abatambula translates to "men who walk," where aba- serves as the augment and class 2 prefix for plural humans, and tambula is the verb root inflected for the relative construction.[30] This structure integrates into full sentences, as in Abasajja abatambula baagala ennyo, meaning "The men who walk love it very much," demonstrating how the relative clause modifies the subject without additional relativizers.[30]
In Swahili, another Bantu language widely used in East Africa, attributive verbs incorporate a relative marker infix, typically -o-, between the tense prefix and the verb stem, alongside the subject prefix agreeing with the head noun's class. A common example is watoto wanaosoma, or more precisely watoto wanaosoma, meaning "children who are reading," where wa- is the class 2 plural subject prefix, na- indicates the present progressive tense, and -o- marks the relative function before the verb root soma "read."[31] Full sentence integration appears in constructions like Alitaja watoto wanaosoma, "He mentioned the children who are reading," highlighting the clause's role in modifying the object.[31]
Variations in attributive verb forms occur across Bantu dialects, including differences in prefix vowels due to phonological harmony or regional patterns; for example, some Swahili dialects like Kiamu may alter vowel quality in subject prefixes (e.g., wa- becoming we- in certain contexts) or infixes to match surrounding sounds.[32] In learner language, agreement errors are frequent, such as mismatched class prefixes in relatives (e.g., using class 1 a- instead of class 2 wa- for watoto), leading to ungrammatical forms like atoto wanaosoma among non-native speakers.[33]
Complex attributive constructions in these languages can involve multi-verb relatives with coordination, as in Swahili kitabu nilichonunua na kusoma, "the book that I bought and read," where the relative marker -chi- agrees with the class 7 noun kitabu, and the infinitival kusoma coordinates with the past tense nilichonunua to express sequential actions within the clause. Similarly, in Luganda, extended relatives like ekitabo ekikolaganira n'okisoma adapt the structure for multiple verbs, maintaining class agreement (class 7 e-) while linking actions.[30]
Typological Extensions
In Other Language Families
In Turkic languages, such as Turkish, attributive verbs typically occur in finite form within relative clauses, directly preceding the head noun without resumptive pronouns. For instance, the construction geldiği ev ("the house he came to") employs the past tense form geldiği from the verb gelmek ("to come"), marked with genitive case to indicate the relation to the head noun ev ("house"). This finite strategy contrasts with non-finite participles in other languages and reflects the head-final nature of Turkic syntax, where tense and agreement are preserved on the verb.[34]
In Sino-Tibetan languages, exemplified by Mandarin Chinese, verbs serve attributively in relative clauses, often linked to the head noun by the particle de (的). The phrase chī de rén ("the person who eats") uses the verb chī ("eat") as a modifier for rén ("person"), with de nominalizing the verbal element. Stative verbs, treated as a subtype of verbs in Chinese grammar, can function attributively without de in simple cases, such as gāo rén ("tall person") from the stative verb gāo ("tall"), allowing direct adjectival-like modification. This system highlights Mandarin's analytic structure, relying on particles for complex attributions rather than inflection.[35]
Uralic languages, like Finnish, utilize non-finite forms such as past participles for attributive verbs, positioned prenominally to modify nouns. An example is kirjoittama kirja ("the book written"), where kirjoittama is the past participle of kirjoittaa ("to write"), agreeing in case with the head kirja ("book"). This participial strategy enables compact relative clause formation, integrating verbal modification seamlessly into noun phrases.[36]
In Austronesian languages, such as Malay, attributive verbs appear in postnominal relative clauses introduced by the relativizer yang. The construction orang yang makan ("the person who eats") features the verb makan ("eat") following yang to specify orang ("person"), typical of head-initial word order in the family. This approach allows verbs to function descriptively without heavy morphological alteration, emphasizing the clause's restrictive role.[37]
Cross-linguistically, attributive strategies vary with word order parameters: languages with prenominal modification, such as Turkish (OV, head-final) and Mandarin (VO, head-initial but prenominal relatives), often use direct verbal forms or particles for concise noun phrase integration, while languages like Malay (head-initial, postnominal relatives) employ relativizers for clarity. Finnish, despite being VO (head-initial), favors prenominal participles akin to OV languages.
Theoretical Implications
Attributive verbs serve as a key indicator of the adjective-verb continuum in linguistic typology, particularly in languages lacking a dedicated adjective class. In such systems, stative or property-expressing verbs function attributively to modify nouns, distributing semantic roles typically associated with adjectives across verbal and nominal categories. Dixon's seminal survey of over 300 languages reveals that roughly 40% feature a large open class of adjectives, 30% have a small open or closed class, and 30% lack a distinct adjective class altogether, relying on attributive verbs for descriptive modification.[38] This variation underscores the non-universal status of adjectives and highlights how attributive verbs enable languages to encode attributes without rigid categorial boundaries.
Research on attributive verbs reveals notable gaps, especially in creoles and pidgins, where simplified grammars often blur verb-adjective distinctions but receive limited comparative attention. Analyses of Suriname creoles demonstrate a gradient continuum from verbal to adjectival uses, yet systematic studies across pidgin continua remain scarce, hindering typological generalizations.[39] Moreover, established linguistic resources lag in updating coverage beyond the 2010s.
Diachronically, attributive verbs exemplify shifts from verbal predicates to adjective-like elements in isolating languages, driven by grammaticalization. In Mandarin Chinese, result-state adjectives frequently derive from change-of-state verbs, evolving through reduced verbal inflection to serve direct attributive roles without copulas, as seen in pairs like pò 'break' (verb) yielding pò 'broken' (adjective). This pathway illustrates how stative verbs can reanalyze into modifiers, contributing to category expansion in analytic systems.[40]