Plural
Plural
Main page
P

Plural

logo
Community Hub0 subscribers
Read side by side
Plural
View on Wikipedia
from Wikipedia
Grammatical features
Related to nouns
Related to verbs
General features
Syntax relationships
Semantics
Phenomena

In many languages, a plural (sometimes abbreviated as pl., pl, PL., or PL), is one of the values of the grammatical category of number. The plural of a noun typically denotes a quantity greater than the default quantity represented by that noun. This default quantity is most commonly one (a form that represents this default quantity of one is said to be of singular number). Therefore, plurals most typically denote two or more of something, although they may also denote fractional, zero or negative amounts. An example of a plural is the English word boys, which corresponds to the singular boy.

Words of other types, such as verbs, adjectives and pronouns, also frequently have distinct plural forms, which are used in agreement with the number of their associated nouns.

Some languages also have a dual (denoting exactly two of something) or other systems of number categories. However, in English and many other languages, singular and plural are the only grammatical numbers, except for possible remnants of dual number in pronouns such as both and either, and in tendency for stock phrases to use "two" as an umbrella term for "many" (eg "double jeopardy" includes prosecuting a person three, four or a dozen times on the same charge).

Use in systems of grammatical number

[edit]

In many languages, there is also a dual number (used for indicating two objects). Some other grammatical numbers present in various languages include trial (for three objects) and paucal (for an imprecise but small number of objects). In languages with dual, trial, or paucal numbers, plural refers to numbers higher than those. However, numbers besides singular, plural, and (to a lesser extent) dual are extremely rare. Languages with numerical classifiers such as Chinese and Japanese lack any significant grammatical number at all, though they are likely to have plural personal pronouns.

Some languages (like Mele-Fila) distinguish between a plural and a greater plural. A greater plural refers to an abnormally large number for the object of discussion. The distinction between the paucal, the plural, and the greater plural is often relative to the type of object under discussion. For example, in discussing oranges, the paucal number might imply fewer than ten, whereas for the population of a country, it might be used for a few hundred thousand.

The Austronesian languages of Sursurunga and Lihir have extremely complex grammatical number systems, with singular, dual, paucal, greater paucal, and plural.

Traces of the dual and paucal can be found in some Slavic and Baltic languages (apart from those that preserve the dual number, such as Slovene). These are known as "pseudo-dual" and "pseudo-paucal" grammatical numbers. For example, Polish and Russian use different forms of nouns with the numerals 2, 3, or 4 (and higher numbers ending with these) than with the numerals 5, 6, etc. (genitive singular in Russian and nominative plural in Polish in the former case, genitive plural in the latter case). Also some nouns may follow different declension patterns when denoting objects which are typically referred to in pairs. For example, in Polish, the noun "oko", among other meanings, may refer to a human or animal eye or to a drop of oil on water. The plural of "oko" in the first meaning is "oczy" (even if actually referring to more than two eyes), while in the second it is "oka" (even if actually referring to exactly two drops).

Traces of dual can also be found in Modern Hebrew. Biblical Hebrew had grammatical dual via the suffix -ạyim as opposed to ־ים‎ -īm for masculine words. Contemporary use of a true dual number in Hebrew is chiefly used in words regarding time and numbers. However, in Biblical and Modern Hebrew, the pseudo-dual as plural of "eyes" עין / עינים‎ ʿạyin / ʿēnạyim "eye / eyes" as well as "hands", "legs" and several other words are retained. For further information, see Dual (grammatical number) § Hebrew.

Certain nouns in some languages have the unmarked form referring to multiple items, with an inflected form referring to a single item. These cases are described with the terms collective number and singulative number. Some languages may possess a massive plural and a numerative plural, the first implying a large mass and the second implying division (like the English modifer "respective[ly]"). For example, "the [combined] waters of the Atlantic Ocean" versus, "the waters of [each of] the Great Lakes [respectively]".

Ghil'ad Zuckermann uses the term superplural to refer to massive plural. He argues that the Australian Aboriginal Barngarla language has four grammatical numbers: singular, dual, plural and superplural.[1]: 227–228  For example:

  • wárraidya "emu" (singular)
  • wárraidyalbili "two emus" (dual)
  • wárraidyarri "emus" (plural)
  • wárraidyailyarranha "a lot of emus", "heaps of emus" (superplural)[1]: 228 

Formation of plurals

[edit]

A given language may make plural forms of nouns by various types of inflection, including the addition of affixes, like the English -(e)s and -ies suffixes, or ablaut, as in the derivation of the plural geese from goose, or a combination of the two. Some languages may also form plurals by reduplication, but not as productively. It may be that some nouns are not marked for plural at all, like sheep and series in English. In languages which also have a case system, such as Latin and Russian, nouns can have not just one plural form but several, corresponding to the various cases. The inflection might affect multiple words, not just the noun; the noun itself need not become plural as such, with other parts of the expression indicating the plurality.

In English, the most common formation of plural nouns is by adding an -s suffix to the singular noun. (For details and different cases, see English plurals.) Just like in English, noun plurals in French, Spanish, and Portuguese are also typically formed by adding an -s suffix to the lemma form, sometimes combining it with an additional vowel. (In French, however, this plural suffix is often not pronounced.) This construction is also found in German and Dutch, but only in some nouns. Suffixing is cross-linguistically the most common method of forming plurals.

In Welsh, the reference form, or default quantity, of some nouns is plural, and the singular form is formed from it, e.g., llygod, mice -> llygoden, mouse; erfin, turnips -> erfinen, turnip.

Plural forms of other parts of speech

[edit]

In many languages, words other than nouns may take plural forms, these being used by way of grammatical agreement with plural nouns (or noun phrases). Such a word may in fact have a number of plural forms, to allow for simultaneous agreement within other categories such as case, person and gender, as well as marking of categories belonging to the word itself (such as tense of verbs, degree of comparison of adjectives, etc.)

Verbs often agree with their subject in number (as well as in person and sometimes gender). Examples of plural forms are the French mangeons, mangez, mangent – respectively the first-, second- and third-person plural of the present tense of the verb manger. In English a distinction is made in the third person between forms such as eats (singular) and eat (plural).

Adjectives may agree with the noun they modify; examples of plural forms are the French petits and petites (the masculine plural and feminine plural respectively of petit). The same applies to some determiners – examples are the French plural definite article les, and the English demonstratives these and those.

It is common for pronouns, particularly personal pronouns, to have distinct plural forms. Examples in English are we (us, etc.) and they (them etc.; see English personal pronouns), and again these and those (when used as demonstrative pronouns).

In Welsh, a number of common prepositions also inflect to agree with the number, person, and sometimes gender of the noun or pronoun they govern.

Nouns lacking plural or singular form

[edit]

Certain nouns do not form plurals. A large class of such nouns in many languages is that of uncountable nouns, representing mass or abstract concepts such as air, information, physics. However, many nouns of this type also have countable meanings or other contexts in which a plural can be used; for example water can take a plural when it means water from a particular source (different waters make for different beers) and in expressions like by the waters of Babylon.

Certain collective nouns do not have a singular form and exist only in the plural, such as "clothes".

There are also nouns found exclusively or almost exclusively in the plural, such as the English scissors. These are referred to with the term plurale tantum. Occasionally, a plural form can pull double duty as the singular form (or vice versa), as has happened with the word "data".

Usage of the plural

[edit]

The plural is used, as a rule, for quantities other than one (and other than those quantities represented by other grammatical numbers, such as dual, which a language may possess). Thus it is frequently used with numbers higher than one (two cats, 101 dogs, four and a half hours) and for unspecified amounts of countable things (some men, several cakes, how many lumps?, birds have feathers). The precise rules for the use of plurals, however, depends on the language – for example Russian uses the genitive singular rather than the plural after certain numbers (see above).

Treatments differ in expressions of zero quantity: English often uses the plural in such expressions as no injuries and zero points, although no (and zero in some contexts) may also take a singular. In French, the singular form is used after zéro.

English also tends to use the plural with decimal fractions, even if less than one, as in 0.3 metres, 0.9 children. Common fractions less than one tend to be used with singular expressions: half (of) a loaf, two-thirds of a mile. Negative numbers are usually treated the same as the corresponding positive ones: minus one degree, minus two degrees. Again, rules on such matters differ between languages.

In some languages, including English, expressions that appear to be singular in form may be treated as plural if they are used with a plural sense, as in the government are agreed. The reverse is also possible: the United States is a powerful country. See synesis, and also English plural § Singulars as plural and plurals as singular.

POS tagging

[edit]

In part-of-speech tagging notation, tags are used to distinguish different types of plurals based on their grammatical and semantic context.[2] Resolution varies, for example the Penn-Treebank tagset (~36 tags) has two tags: NNS - noun, plural, and NPS - Proper noun, plural,[3] while the CLAWS 7 tagset (~149 tags)[4] uses six: NN2 - plural common noun, NNL2 - plural locative noun, NNO2 - numeral noun, plural, NNT2 - temporal noun, plural, NNU2 - plural unit of measurement, NP2 - plural proper noun.

See also

[edit]

Notes

[edit]

Further reading

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Plural

View on Grokipedia
from Grokipedia
In linguistics, the plural is a value of the grammatical category of number, denoting a quantity greater than one referent, in contrast to the singular which specifies exactly one.[1] This category is typically expressed through inflectional morphology on nouns, pronouns, and sometimes verbs, allowing speakers to indicate multiplicity in reference.[2] While the plural form often applies to two or more entities, its semantic interpretation can vary, sometimes extending to collective or abstract senses even for singular objects in certain contexts.[2] Grammatical number systems, including the plural, differ widely across languages in complexity and marking strategies. Most languages distinguish only between singular and plural, with the plural commonly formed by adding suffixes such as -s in English (e.g., "cat" to "cats").[3] However, some languages feature more elaborate systems, incorporating a dual form for exactly two referents, as seen in Inuktitut where nouns inflect distinctly for singular (titiraut, "pen"), dual (titirautiik), and plural (titirautit).[3] In languages like Arabic or Yupik, number marking involves specific affixes that reflect these extended categories such as the dual, highlighting how number inflection adapts to cultural and cognitive needs for precise quantification.[2] The plural's role extends beyond nouns to influence agreement in sentences, ensuring consistency in verbs, adjectives, and pronouns. For instance, in English, plural subjects trigger plural verb forms (e.g., "They run"), while irregular plurals like "children" or "geese" preserve historical patterns of formation.[4] This agreement mechanism is crucial for syntactic coherence, and in agglutinative languages, number markers can accumulate on multiple elements within a phrase.[3] Overall, the plural underscores the universality of quantity encoding in human language while showcasing typological diversity.

Grammatical Foundations

Definition and Role in Number Systems

In linguistics, the plural is defined as an inflected form or syntactic construction that denotes multiplicity, specifically referring to two or more entities, as opposed to a single one.[5] This marking typically applies to nouns, pronouns, and sometimes adjectives or verbs through agreement, serving to indicate quantity in grammatical structures.[6] Grammatical number constitutes a fundamental category in many languages, encompassing distinctions such as singular (denoting exactly one entity), plural (more than one), and less commonly, dual (exactly two) or trial (exactly three).[7] These categories encode quantification over referents, influencing agreement patterns across syntactic elements like verbs and determiners, and are realized morphologically in inflectional languages or analytically in others.[7] The plural plays a central role in this system by contrasting with singular to express non-unitary sets, while dual and trial provide finer-grained numerosity in select languages such as Classical Arabic (dual) or some Austronesian languages (trial).[7] The origins of the plural trace back to Proto-Indo-European (PIE), the reconstructed ancestor of the Indo-European language family, where plural forms often evolved from earlier collective suffixes denoting groups or aggregates.[8] In PIE, the neuter plural, for instance, utilized a collective suffix *-h₂, which manifested as -a in many daughter languages and shifted from indicating undivided wholes to distributive plurals over time.[9] This evolution persisted in major Indo-European branches, such as Germanic and Romance, where plurals retained traces of these collective origins while adapting to mark countable multiplicity.[9] In English, an Indo-European language, the plural is morphologically realized by adding -s or -es to nouns, as in the singular "cat" becoming the plural "cats" to denote multiple animals.[10] By contrast, isolating languages like Mandarin Chinese lack obligatory morphological plurals on nouns, instead relying on quantifiers (e.g., "many" or "several") or context to convey multiplicity, with limited inflectional marking via suffixes like -men primarily on pronouns referring to humans.[11] This typological variation highlights the plural's role as a flexible component of number systems across language families.[12]

Distinction from Singular and Other Numbers

The singular grammatical number marks a single entity or referent, as exemplified by the English noun phrase "the dog," which refers to one individual dog. In contrast, the plural number indicates more than one, such as "the dogs," encompassing two or more entities. This distinction forms the core binary opposition in most languages' number systems, where the plural serves as the default for multiplicity beyond unity. Number agreement varies across languages: in English, it is obligatory, requiring verbs, adjectives, and determiners to match the noun's number, as in "the dog runs" versus "the dogs run." Conversely, in Japanese, number marking on nouns is optional and lacks obligatory agreement with other elements, allowing forms like inu ("dog") to contextually imply singular or plural without inflectional changes.[13] Beyond the singular-plural binary, some languages distinguish additional numbers, such as the dual for exactly two referents. In Arabic, the dual is obligatorily marked on nouns, verbs, adjectives, and pronouns; for instance, kitāb ("book," singular) becomes kitābāni ("two books," dual).[14] Other systems include the trial for three referents or the paucal for a small but unspecified number greater than two. In Austronesian languages like Samoan, pronouns exhibit trial forms alongside singular, dual, and plural, such as matou (trial inclusive "we three"). These extended distinctions follow Greenberg's Universal 34, which posits that no language has a trial without a dual, and no dual without a plural.[15] Such non-binary numbers are declining in modern usage across many languages. In Russian, the dual has been lost diachronically, with its former forms reanalyzed into countability distinctions between singular and plural. Similarly, in the Austronesian language Kala, the dual persists in pronouns but is eroding in nouns due to contact influences, reflecting a broader trend toward simplification in globalizing contexts.[16] Functionally, the plural often conveys indefiniteness or generality, contrasting with the singular's potential for definiteness. In Bantu languages, noun classes pair singular and plural forms (e.g., class 1 singular with class 2 plural), where bare plural noun phrases typically imply indefinite or generic reference, as in Zulu abantu ("people," indefinite plural) versus definite singular umuntu ("the person").[17] Bantu systems feature up to 20 classes, with singular-plural pairings enabling nuanced distinctions beyond mere count, such as augmentative or diminutive senses.[18] Linguistically, number distinctions are marked through fusional or agglutinative typology. In fusional languages like Latin, a single ending fuses number with other categories, such as -i in puerī ("boys," plural nominative).[19] Agglutinative languages like Turkish add discrete suffixes for number, as in köpek-ler ("dogs," where -ler solely marks plural), allowing clearer separation from case or other features.[20] This typological contrast affects how number integrates with broader inflectional paradigms.[19]

Plural Formation

Mechanisms in Nouns

Plural formation in nouns primarily occurs through morphological processes that alter the word's form to indicate multiplicity. These mechanisms vary across languages and include affixation, where a suffix is added to the singular stem; ablaut, involving internal vowel changes; suppletion, where the plural form is entirely unrelated to the singular; and reduplication, which repeats part or all of the stem.[21][22] Affixation is the most common method, as seen in English where regular nouns typically add the suffix -s (e.g., cat/cats) and in German where many nouns, particularly weak masculines and feminines, add -en (e.g., der Student/die Studenten; die Frau/die Frauen), though umlaut occurs in some cases with other endings like -e (e.g., Apfel/Äpfel).[21][23] Ablaut, or vowel gradation, marks plurality through stem-internal changes without affixation, exemplified in English by foot/feet and goose/geese, where a back rounded vowel shifts to a front unrounded one.[21][24] Suppletion involves replacing the singular form with a phonologically distinct plural, as in English child/children, where the plural derives from an Old English weak ending -rum that evolved irregularly.[21][24] Reduplication creates plural forms by copying a portion of the stem, often at the onset, and is prominent in Salishan languages such as St'át'imcets (Lillooet), where initial consonant reduplication marks plurality (e.g., sqʷəqʷyíc 'rabbits' from sqʷəyíc 'rabbit').[25] Irregular plurals frequently arise from historical sound changes that disrupted regular patterns, leading to exceptions preserved through analogy or lexical retention.[24] In Latin, sound shifts like vowel lengthening affected fourth-declension nouns, resulting in forms such as domus (singular) / domūs (plural nominative), which upon borrowing into English retain the Latin plural in specialized contexts (e.g., academic or architectural usage: domus / domūs).[26][27] Zero plurals, where the singular and plural forms are identical without morphological marking, occur in English for certain animal names like sheep and deer, relying on context for interpretation.[21] This pattern is widespread in classifier languages such as Mandarin Chinese, which lacks obligatory plural inflection on nouns and instead uses numeral classifiers or quantifiers (e.g., yī gè rén 'one person' vs. duō gè rén 'many people') to convey number, treating bare nouns as number-neutral.[28] Cross-linguistically, gender influences plural formation in Romance languages, where masculine and feminine nouns often share the same plural suffix despite differing singular endings. In French, for instance, both genders typically add -s (e.g., masculine chat/chats 'cat/cats'; feminine chatte/chattes 'female cat/female cats'), though pronunciation varies by phonological context, with the suffix silent in isolation but realized before vowels.[29][30]

Forms in Verbs, Adjectives, and Pronouns

In many languages, verbs exhibit number agreement with their subjects through inflectional morphology, ensuring concord in person and number. For instance, in English, singular subjects pair with singular verb forms, as in "he runs," while plural subjects require plural forms, such as "they run."[31] This agreement highlights the unmarked nature of singular count nouns compared to marked plural forms, influencing processing and error patterns in production.[32] In pro-drop languages like Spanish, where subject pronouns can be omitted due to rich verbal inflection, number agreement remains explicit on the verb. For example, the third-person plural form "corren" (they run) distinguishes plurality without a overt pronoun, as the affixal morphology encodes both person and number sufficiently for recovery.[33] This contrasts with non-pro-drop languages like English, where poorer agreement morphology necessitates overt subjects to convey number clearly.[33] Adjectives in inflectional languages often inflect for number to agree with the nouns they modify, a process absent in analytic languages like English where adjectives remain invariant. In Italian, a Romance language with synthetic features, adjectives adjust endings for both gender and number; for example, the masculine singular "grande" (big) becomes "grandi" in the masculine plural to agree with nouns like "case grandi" (big houses).[34] This concord ensures syntactic harmony within noun phrases, with adjectives typically following the noun but still marking plurality explicitly.[34] Pronouns frequently encode plurality through distinct forms that may carry additional semantic nuances beyond mere count. In many Austronesian languages, first-person plural pronouns distinguish inclusive and exclusive variants: the inclusive form, such as Chamorro "ta" (you and I, plus others), includes the addressee, while the exclusive "in" (I and others, excluding you) does not.[35] This clusivity distinction is a hallmark of the family, appearing in nearly all Austronesian languages surveyed.[35] Certain plural pronoun uses serve honorific or stylistic purposes, as seen in the majestic plural or royal "we," where a singular referent employs a plural form to convey authority or grandeur. Historically employed by monarchs and deities, this pluralis majestatis appears in languages like Latin and English literary traditions, though it lacks clear attestation in Biblical Hebrew pronouns or verbs.[36] In modern English dialects, the second-person plural "you" has developed regional variants to address the absence of a dedicated form; Southern American English uses "y'all" (a contraction of "you all") specifically for plural reference, as in "y'all come back now," distinguishing it from singular "you."[37] In polysynthetic languages such as Inuktitut, an Eskimo-Aleut language, number marking extends beyond nouns and verbs to other elements, including certain particles and adverbial affixes incorporated into complex words. Inuktitut features a three-way number system (singular, dual, plural) that applies to verbal agreement and can influence incorporated modifiers functioning adverbially, allowing holistic expression of plurality within single words.[38] For example, verbal complexes may include affixes denoting plural actions or locations that behave like adverbial particles, reflecting the language's agglutinative nature where plurality permeates the entire predicate.[39]

Special Cases and Exceptions

Non-Countable and Mass Nouns

Mass nouns, also known as uncountable or non-count nouns, denote substances, materials, collectives, or abstract concepts that are conceptualized as undifferentiated wholes without discrete units, thereby lacking inherent plurality. In English, examples include "water," referring to a liquid substance, and "information," an abstract entity that cannot be enumerated individually. These nouns exhibit cumulative reference, where the combination of two portions yields a larger portion of the same kind, and they are syntactically singular, incompatible with cardinal numbers or indefinite articles like "a" or "an."[40] When plurality must be expressed for mass nouns in English, speakers rely on indirect strategies rather than direct morphological pluralization. Partitive constructions quantify portions via containers or measures, such as "bottles of water" or "items of information," allowing reference to multiple discrete amounts. Alternatively, zero-derived plurals can indicate varieties or types, as in "wines" for different kinds of wine, while lexical shifts enable plurals for servings, like "beers" denoting individual drinks. These methods preserve the mass semantics while accommodating contexts requiring multiplicity.[41] Cross-linguistic variation reveals greater flexibility in pluralizing mass nouns in some languages compared to English's stricter uncountability. In Russian, for example, the mass noun "voda" (water) can form the plural "vody" to refer to types or portions, such as different mineral waters, a usage unavailable for equivalents like "pivo" (beer). This parametric difference underscores how number morphology on mass nouns, though uncommon, occurs in various languages to encode subtypes or distributed quantities.[42][43] Historical semantic shifts illustrate how former countable plurals can become mass nouns in English. The term "data," derived from the Latin plural "data" (neuter plural of "datum," meaning "things given"), was initially treated as plural in scientific English but has largely evolved into a singular mass noun in contemporary usage, reflecting a reconceptualization as an undifferentiated aggregate.[44]

Invariant or Defective Forms

In linguistics, invariant nouns are those that do not inflect for number, maintaining the same form regardless of whether they refer to a single entity or multiple ones.[45] These include collectives like English "cattle," which functions as a mass noun without singular or plural distinction, derived from Old Northern French "catel" meaning movable property and entering English as an uncountable term for livestock.[46] Similarly, "sheep" and "deer" remain unchanged in both singular and plural contexts, reflecting historical patterns where certain animal nouns avoided morphological marking for plurality. Defective nouns exhibit partial paradigms, inflecting for number only in specific contexts. Proper names, for instance, typically lack plural forms but can form them when referring to families or groups, as in "the Smiths" for multiple members of the Smith family, following standard English pluralization rules by adding "-s" to names not ending in sibilants.[47] Collectives like "people" (plural of "person") or "police" often resist further pluralization, treating the group as a unitary concept despite denoting multiplicity.[45] A subset of invariant forms includes pluralia tantum, nouns that occur exclusively in the plural and lack a singular counterpart, such as "scissors," "trousers," and "glasses." These often arise etymologically from compound structures or paired objects; for example, "scissors" derives from Latin caesorium (a cutting instrument), treated as plural due to its two blades functioning as a set, while "trousers" evolved from Old French trebus (a garment with two legs), entering English in the 16th century as a plural form emphasizing duality.[45] Conversely, singularia tantum are nouns restricted to singular form, including abstracts like "news," "furniture," and "information," which denote uncountable concepts or masses and do not pluralize; "news," for instance, stems from Middle English neves (plural of Latin nova, new things) but standardized as a singular mass noun by the 17th century.[48] Cross-linguistically, invariant forms vary. Japanese nouns generally lack obligatory number marking, with forms like neko (cat/cats) remaining unchanged and plurality conveyed contextually via classifiers or quantifiers such as hon (counter for long objects) in san-hon no enpitsu (three pencils).[49] In Slavic languages, singularia tantum predominate among abstract nouns, as in Russian udivlenie (surprise) or ispug (fright), which occur only in singular due to their non-discrete, qualitative semantics, contrasting with count nouns that fully inflect.[50]

Semantic and Syntactic Usage

Indicating Multiplicity and Collectivity

Plural forms in language primarily serve to indicate multiplicity, denoting more than one entity, and collectivity, referring to a group treated as a unified whole. Semantically, plural marking distinguishes between distributive readings, where a predicate applies to each member of the set individually, and collective readings, where it applies to the entire group. For instance, in the sentence "The children are asleep," a distributive interpretation requires that each child is asleep separately, whereas "The children gathered in the yard" conveys a collective action of the group as a unit. This distinction arises from the semantics of plurality, where plural noun phrases can sum individuals into a group but predicates may distribute over atoms or apply holistically.[51] Syntactically, plural forms trigger agreement across elements in a sentence, ensuring concord in number features. In English, a plural subject like "the cats" requires a plural verb form, as in "The cats run," where the verb conjugates to match the plural number of the subject noun phrase. This agreement extends to other categories, such as adjectives and pronouns, maintaining syntactic harmony. Additionally, plurals facilitate coordination, allowing conjoined noun phrases like "cats and dogs" to form plural entities that agree collectively with predicates, e.g., "Cats and dogs play together." Quantification also interacts with plurals, where expressions like "many cats" or "three dogs" license plural marking to denote sets of multiple items, influencing the interpretation of numerical or indefinite quantifiers. Generic plurals represent a distinct semantic use, where bare plural noun phrases refer to kinds, species, or classes rather than specific individuals, often conveying general properties or habits. For example, "Cats are mammals" uses the plural "cats" to generalize about the feline species as a whole, without implying a particular group of cats. This contrasts with specific plurals, such as "The cats in the yard are mammals," which refer to a definite set of individuals. Generic interpretations arise in subject position and involve a kind-referring semantics, distinct from episodic or referential uses of plurals. Edge cases illustrate further nuances in plural usage. Honorific contexts may employ plural forms to denote respect for a single entity, as in English historical usage "His Majesties" for a king, triggering plural agreement despite singular reference. Similarly, indefinite singular constructions like "one of the boys" can treat the embedded plural "the boys" as controlling agreement in relative clauses, e.g., "One of the boys who are late will apologize," where "who are" agrees with the plural antecedent "boys." These cases highlight how plurality can override strict numerical singularity for pragmatic or syntactic reasons.

Variations Across Languages

Plural marking exhibits significant typological variation across language families, with some requiring obligatory expression of plurality for count nouns while others treat it as optional or context-dependent. In Indo-European languages such as English, plural marking is generally obligatory for countable nouns in non-generic contexts, ensuring that multiplicity is explicitly indicated through suffixes or other morphological changes.[52] In contrast, many Sino-Tibetan languages, including Mandarin Chinese, feature optional plural marking, where suffixes like -men (们) appear primarily with human or animate referents and can be omitted when plurality is inferable from context.[53] This optionality aligns with the isolating morphological profile common in the family, prioritizing pragmatic inference over obligatory inflection. Certain language families introduce semantic distinctions in pluralization based on animacy hierarchies, further diversifying how plurality is encoded. Algonquian languages, for instance, classify nouns into animate and inanimate categories, with plural suffixes differing accordingly: animate plurals often end in -ag or -ak (e.g., Ojibwe waabizheshiwaag 'martens'), while inanimate plurals use -an or -ag (e.g., Ojibwe inaakanaawaan 'dishes').[54] This animate/inanimate split influences not only plural forms but also verb agreement and possession, reflecting a worldview where animacy extends beyond biological life to culturally significant entities.[55] In agglutinative languages, plural markers integrate seamlessly into stacked suffix sequences, allowing complex expressions of number alongside other grammatical categories. Turkish exemplifies this through its vowel-harmonic plural suffix -ler or -lar, which attaches to noun roots before additional case or locative endings; for example, ev (house) becomes evler (houses) and evlerde (in the houses), demonstrating how plurality is layered without altering the root.[56] This agglutinative strategy enables concise yet information-rich forms, characteristic of Turkic and Uralic languages where suffixes accumulate in a fixed order to convey multiple relations.[57] Socio-cultural factors also shape plural usage, particularly in honorific systems where plurality signals respect or social distance. In Korean, plural pronouns for the first person, such as jeohui-deul (we, humble), incorporate honorific elements akin to singular forms, extending deference to groups and aligning with the language's hierarchical speech levels.[58] Conversely, some languages employ plural avoidance in taboo contexts to mitigate offense or supernatural risks; for example, in certain African languages like Kambaata, speakers select alternate lexical forms over standard plurals when referring to in-laws or prohibited topics, preserving social harmony through circumlocution.[59] Contemporary linguistic evolution, driven by contact and globalization, often results in simplified plural systems, especially in creoles and pidgins. These contact varieties typically reduce obligatory marking to optional or invariant forms, as seen in Nigerian Pidgin English where the English-derived -s plural appears inconsistently, relying instead on quantifiers or context for multiplicity (e.g., di boy dem 'the boys').[60] Globalization amplifies this trend by promoting invariant or simplified plurals in global Englishes and hybrid varieties, where exposure to dominant languages erodes traditional inflections in favor of pragmatic efficiency, particularly in urban multilingual settings.[61]

Formal and Computational Treatment

Part-of-Speech Tagging

In part-of-speech (POS) tagging, grammatical number for nouns is typically distinguished through specific tags or morphological features to indicate singular versus plural forms. The Penn Treebank tagset, a widely used scheme for English, employs NN for singular or mass nouns (e.g., "car" or "data" in singular contexts) and NNS for plural nouns (e.g., "cars").[62] Similarly, for verbs, tags like VB (base form) contrast with VBP (present tense, non-3rd person singular), reflecting number agreement.[62] This scheme enables taggers to capture plurality as an inherent property of the word's inflection.[63] The Universal Dependencies (UD) framework extends this to cross-lingual settings by using a universal POS tagset where nouns are uniformly tagged as NOUN, with an additional Number feature to specify Sing (singular) or Plur (plural), such as Number=Plur for "cars."[64] This feature-based approach facilitates consistent annotation across over 100 languages, supporting multilingual models that propagate plural information via dependencies.[65] UD datasets, like UD English-EWT, include explicit plural markings, aiding in training taggers for number-sensitive tasks.[66] Challenges in POS tagging for plural forms arise primarily from ambiguity, where words like "data" can function as singular (NN, e.g., "the data is reliable") or plural (NNS, e.g., "the data are analyzed") based on syntactic agreement and context.[63] Resolving such cases requires contextual disambiguation, as rule-based systems struggle with exceptions while machine learning models depend on sufficient training data to learn patterns.[67] In low-resource languages, error rates for plural tagging in UD can exceed 10-15% higher than in high-resource ones like English (where accuracies reach 95-97%), due to limited annotated data and morphological complexity.[68] Approaches to POS tagging have evolved from rule-based methods, which apply hand-crafted heuristics for plural inflections, to probabilistic models like Hidden Markov Models (HMMs) that estimate tag sequences based on emission and transition probabilities.[67] Modern systems leverage transformer-based architectures, such as BERT fine-tuned for POS, achieving state-of-the-art accuracies by incorporating bidirectional context for number resolution.[69] These neural methods outperform earlier HMMs on plural disambiguation, particularly in UD benchmarks.[70] Historically, POS tagging began with rule-based systems in the 1970s, such as early parsers using affix rules for plural detection in English.[67] By the 1990s, stochastic approaches like HMMs dominated, as seen in the Brill tagger's transformation rules for error correction.[67] Post-2010, the shift to neural models, including recurrent and transformer networks, addressed limitations in handling rare plural forms and cross-lingual transfer, with UD enabling scalable annotation since 2014.[65]

Applications in Linguistics and NLP

In linguistics, finite-state transducers (FSTs) are widely employed for morphological analysis, enabling the generation and parsing of plural forms by modeling inflectional rules as bidirectional mappings between surface forms and underlying morphemes. For instance, FSTs can systematically produce plural variants such as English "cat" to "cats" or more complex cases in agglutinative languages like Turkish, where suffixes indicate number alongside other features. This approach, foundational in computational morphology, facilitates efficient processing of large lexicons and supports applications in dictionary building and language documentation.[71][72] Sociolinguistic research utilizes plural forms to examine dialectal variations, revealing how social factors influence morphological marking across communities. Studies on regional dialects, such as those in American English, highlight differences in plural realization, like the use of zero plurals for mass nouns in certain Southern varieties, which reflect identity and socioeconomic patterns. These investigations employ plural markers as indicators of language shift and variation, aiding in the preservation of endangered dialects.[73][74] In natural language processing (NLP), handling plural mismatches is critical for machine translation systems, particularly between languages with differing number agreement rules, such as English and French. Neural machine translation models often struggle with morphological mismatches, leading to errors in gender-number concord, as seen in translations where English singular "the child" becomes French plural "les enfants" inappropriately due to context ambiguity. Techniques like subword regularization and attention mechanisms mitigate these issues, improving accuracy by up to 5-10% on benchmarks involving number-sensitive pairs.[75][76] Text generation in chatbots and language models requires maintaining plural consistency to ensure grammatical coherence across responses. Large language models, such as those underlying GPT variants, are evaluated for factual and syntactic consistency, including number agreement, where inconsistencies in plural usage can degrade user trust and output quality. Prompt engineering and fine-tuning strategies enforce plural harmony, as demonstrated in consistency analyses showing reduced error rates in multi-turn dialogues.[77][78] Advanced NLP tasks like semantic role labeling (SRL) incorporate plural distinctions to differentiate collective versus distributive interpretations in predicate-argument structures, as annotated in resources like PropBank. In PropBank, plural noun phrases may receive roles such as ARG1 (agent-like) where collectivity implies group action, versus distributive readings for individual instances, enhancing inference in event understanding. This granularity supports downstream applications in question answering and summarization by resolving ambiguities in multi-entity scenarios.[79][80] Multilingual models like multilingual BERT (mBERT) demonstrate robust handling of plurals across languages, achieving above-chance performance in number agreement tasks through pretraining on diverse corpora. Evaluations show mBERT distinguishing singular and plural forms in zero-shot settings for over 100 languages, though performance varies by morphological richness, with agglutinative languages benefiting from shared subword representations. This capability enables cross-lingual transfer for plural-sensitive tasks without language-specific annotations.[81][82] In emerging low-resource NLP, transfer learning addresses plural morphology challenges by leveraging high-resource models to infer inflections in under-resourced languages. Cross-lingual morphological taggers, trained on related languages, achieve up to 80% accuracy in plural tagging for languages like those in the Universal Dependencies treebanks, using techniques such as multilingual embeddings to bridge data gaps. This approach has been pivotal for revitalizing minority languages with sparse plural paradigms.[83][84] Ethical considerations in plural handling arise from biases in NLP systems, particularly where plural assumptions reinforce gender stereotypes in non-binary or gender-neutral contexts. For example, models trained on gendered corpora may default to binary plural forms, marginalizing neutral language and perpetuating exclusion in translation or generation tasks. Mitigation strategies, including debiasing datasets and inclusive prompting, are essential to promote equitable representations across diverse identities.[85][86]

References

  1. https://www.merriam-webster.com/dictionary/plural
  2. Major Morphological Categories
  3. 5.7 Inflectional morphology – ENG 200: Introduction to Linguistics
  4. Plural Nouns: Rules and Examples - Grammarly
  5. What is a Number | Glossary of Linguistic Terms - SIL Global
  6. Number - Grammatical Features
  7. Number in Language
  8. The Proto-Indo-European Collective and the Hittite Neuter Plural
  9. What Is a Plural Noun? | Examples, Rules & Exceptions - Scribbr
  10. Chinese Morphology | Oxford Research Encyclopedia of Linguistics
  11. Syntactic Typology: Studies in the Phenomenology of Language
  12. https://escholarship.org/content/qt5917s4b8/qt5917s4b8_noSplash_ee3bb93c85fc6ce64ec16f9e0981d99d.pdf
  13. Usage and Decline of the Dual Number in Kala - DOI
  14. [PDF] 202 bantu linguistic terminology
  15. A typology of northwestern Bantu gender systems - De Gruyter Brill
  16. 3.3 Morphology of Different Languages - BC Open Textbooks
  17. 10.3. Packaging words and morphemes
  18. [PDF] • Morphological analysis: What we can learn from plural nouns ... - 13
  19. 5.3 Morphology beyond affixes – Essentials of Linguistics, 2nd edition
  20. An ERP-study of German 'irregular' morphology - ScienceDirect.com
  21. 14.4 Morphological change – Essentials of Linguistics, 2nd edition
  22. [PDF] Going Radical in Salish - Cascadilla Proceedings Project
  23. Latin Nouns verb 'domus' conjugated - Verbix verb conjugator
  24. Latin : domus, dom-us f. - louis ha
  25. Acquisition of plurality in a language without plurality * | Request PDF
  26. The Romance noun: a comparative-historical study of plural formation
  27. https://oxfordre.com/linguistics/display/10.1093/acrefore/9780199384655.001.0001/acrefore-9780199384655-e-716
  28. Subject/Verb Agreement - Purdue OWL
  29. The Marked Effect of Number on Subject–Verb Agreement
  30. Reversing the Approach to Null Subjects: A Perspective ... - Frontiers
  31. Adjectives in Italian: Form and Agreement - ThoughtCo
  32. Chapter Inclusive/Exclusive Distinction in Independent Pronouns
  33. None
  34. Y'all, we need to talk about 'y'all' - NPR
  35. Inflectional Morphology – ENGL6360 Descriptive Linguistics for ...
  36. [PDF] Polysynthetic word-internal adjectives and verb-like adjectives in Inuit
  37. The Logic of Mass Expressions - Stanford Encyclopedia of Philosophy
  38. What is a Mass Noun? (With Examples) | Grammarly
  39. [PDF] Mass-to-Count Shifts and Number Morphology
  40. [PDF] 1 Reference to kinds and to other generic expressions in Spanish
  41. Data are or data is? | Language - The Guardian
  42. Pluralia tantum nouns and the theory of features: a typology of ...
  43. What's the singular form of 'cattle'? - Motivated Grammar
  44. Plural and Possessive Names: A Guide - Merriam-Webster
  45. Grammatical Gender
  46. Numeral classifiers in Japanese - Oxford Academic
  47. [PDF] CORPUS-BASED PROFILES OF RUSSIAN NOUNS
  48. [PDF] plurality - Semantics Archive
  49. [PDF] The Syntax of Semantic Agreement in British English - Peter W. Smith
  50. [PDF] A new approach to plurals-of-politeness and their number agreement
  51. [PDF] Nominal plurality in languages of the Greater Hindukush - DiVA portal
  52. https://oxfordre.com/linguistics/display/10.1093/acrefore/9780199384655.001.0001/acrefore-9780199384655-e-530
  53. [PDF] Introduction to Nouns: Gender
  54. [PDF] Inanimate nouns as subjects in Mi'gmaq
  55. The Production of Noun Suffixes by Turkish-Speaking Children with ...
  56. [PDF] MORPHOLOGY - İTÜ
  57. Korean Language Course Plurals (Speak Like Natives)
  58. 2.2 Language and offence – Essentials of Linguistics, 2nd edition
  59. [PDF] PLURAL MARKING PAITERNS IN NIGERIAN PIDGIN ENGLISH Sali ...
  60. https://www.jbe-platform.com/content/journals/10.1075/jpcl.19.1.06sie
  61. [PDF] Penn Treebank POS Tag Set
  62. [PDF] Part-of-Speech Tagging Guidelines for the Penn Treebank Project ...
  63. Number : number - Universal Dependencies
  64. [PDF] Universal Dependencies - Stanford NLP Group
  65. Universal Dependencies
  66. [PDF] Part-of-Speech Tagging - Stanford University
  67. [PDF] A Benchmark Study and Error Analysis of POS Tagging for ...
  68. Part-of-Speech Tagging with Rule-Based Data Preprocessing and ...
  69. Part of speech tagging: a systematic review of deep learning and ...
  70. 2.3 Morphological Analysis with Finite State Transducers
  71. Finite State Transducer (FSTs) in NLP - GeeksforGeeks
  72. [PDF] Lexical Plurals - Georgetown University
  73. Regional and Sociolinguistic Variation of Personal Pronouns in ...
  74. Morphological mismatches in machine translation - ResearchGate
  75. (PDF) Syntactic mismatches in machine translation - ResearchGate
  76. [PDF] Consistency Analysis of ChatGPT - ACL Anthology
  77. (PDF) Language Model Evaluation in Open-ended Text Generation
  78. [PDF] The Proposition Bank: An Annotated Corpus of Semantic Roles
  79. [PDF] Dependency-based Semantic Role Labeling of PropBank
  80. [PDF] This is a BERT. Now there are several of them. Can they generalize ...
  81. [PDF] Cross-Linguistic Syntactic Difference in Multilingual BERT
  82. [PDF] Cross-Lingual Morphological Tagging for Low-Resource Languages
  83. [PDF] Neural Sequence-to-Sequence Models for Low-Resource Morphology
  84. Gender Bias in Machine Translation - MIT Press Direct
  85. [PDF] A Survey on Gender Bias in Natural Language Processing
User Avatar
No comments yet.