Universals Archive
A database of proposed crosslinguistic laws and tendencies in grammar across the world’s languages. 2,029 published entries.
1,237 entries · page 25 of 25
| # | Standardized / Original | Domain | Type | Keywords |
|---|---|---|---|---|
| 1979 | A grammatical element is less likely to be omitted the greater the ambiguity which would result from its omission. | syntax, semantics | implication | ellipsis, ambiguity |
| 1981 | When a language with a full copula begins to lose it, or when a language with a zero copula begins to develop one, the use in which the change is initiated is that of identity or equivalence with a noun complement. | syntax | implication | copula, zero |
| 1982 | When a language with zero copula inserts a pro-copula it does so between the subject and complement regardless of the normal position of verbs in the language. | syntax | unconditional | copula, zero, order, verb |
| 1996 | IF there is a gender system, THEN the domain of animate nouns is further differentiated; IF there is a system of noun classification, THEN the domain of inanimate nouns is further differentiated. |
inflection, syntax, semantics | implication | gender, noun class, animacy |
| 1998 | IF there are gender-number-person contrasts on the predicate agreeing in these categories with the subject, THEN there are the same contrasts on subject pronouns. | inflection, syntax | implication | agreement, verb, gender, number, person, pronoun |
| 1999 | IF a bound form is morphologically replaced through analogical levelling or otherwise, THEN the corresponding free form is also replaced, or has already been replaced. | inflection, syntax | implication | free form, bound form, analogy, levelling |
| 2000 | All grammars leak. | phonology, morphology, syntax, semantics, etc. | unconditional | analogy, iconicity, economy, plumbing |
| 2001 | Some semantic criteria never define a gender completely. For example, there is never a gender exclusively for the names of languages. | inflection | unconditional | gender |
| 2002 | Gender cannot be assigned on the basis of syntactic properties of nouns (such as, e.g., that of taking prepositional complements). | inflection, syntax | unconditional | gender, gender assignment |
| 2003 | When gender is assigned according to morphological properties of nouns (such as declension class, or other inflectional or derivational categories) or to phonological properties of nouns, this is always secondary, being limited to residues of semantic assignment or being overruled by semantic assignment rules. | phonology, morphology | unconditional | gender, gender assignment, declension |
| 2004 | IF there is a gender distinction in any pronoun other than 3rd person, THEN there will be a gender distinction (the same?) in 3rd person. | inflection | implication | pronoun, person, gender |
| 2007 | IF infinitive phrases are non-prototypical controllers of agreement in gender and number, requiring neutral agreement forms, THEN so are clauses. | inflection, syntax | implication | agreement, gender, number, infinitive, clause |
| 2008 | The Agreement Hierarchy:attributive < predicate < relative pronoun < personal pronoun(i) Possible agreement pattern are constrained as follows: As we move rightwards along the hierarchy, the likelihood of semantic agreement [in gender or number] [with hybrid nouns] will increase monotonically (that is, with no intervening decrease).(ii) Whenever, in a given position on the Agreement Hierarchy there is a difference between the agreements found in the nominative and in the oblique cases, the likelihood of semantic agreement in the nominative will be as high as or higher than the likelihood of semantic agreement in the oblique cases.(iii) When stacked targets of a given controller stand in different agreement forms, the further target will show semantic agreement.(iv) If parallel targets show different agreement forms, then the further target will show semantic agreement.(v) For any particular target type, the further it is removed from its controller, the greater the likelihood of semantic agreement. | inflection, syntax, semantics | implicational hierarchy | agreement, gender, number, case, nominative, oblique |
| 2009 | When gender agreement is in decline, IF target words other than personal pronouns continue to agree in gender, THEN personal pronouns also continue to agree in gender. | inflection, syntax | implication | agreement, gender, personal pronoun |
| 2010 | In agreement with conjoined NPs and person resolution is required, when the conjuncts include a 1st person, 1st person agreement forms will be used; otherwise, when the conjuncts include a 2nd person, 2nd person agreement forms will be used; otherwise, 3rd person agreement forms (the default) are used. | inflection, syntax | unconditional | agreement, coordination, resolution rules, person, person hierarchy, 1, 2, 3 |
| 2011 | The type of gender resolution system found is constrained by the [gender] assignment system of the language. Languages with strict semantic assignment systems [...] have semantic resolution systems, as do those with predominantly semantic systems [...]. Languages with formal assignment systems may have semantic resolution [...], syntactic (=formal) resolution [...], or mixed resolution [...]. Informally speaking, gender resolution may be "more semantic" but not "less semantic" than gender assignment. | inflection, syntax, semantics | implication | agreement, coordination, resolution rules, gender |
| 2012 | When vowel length is contrastive, pitch is the most important cue, intensity the second-most important cue, and duration the least important cue for word stress.When vowel length is not contrastive, duration is the most important cue for word stress.When there is contrastive tone, duration is the most important cue for word stress and pitch the least important. | prosodic phonology | unconditional | stress, pitch, vowel length, intensity, duration, tone |
| 2013 | If there is contrastive pitch accent, then there is a vocalic length contrast.If there is a vocalic length contrast, then there cannot also be contrastive dynamic (=stress) accent.If there is contrastive dynamic (=stress) accent, then there cannot also be contrastive vowel length.[i.e., and vice versa] | prosodic phonology | implication | pitch, stress, vowel length |
| 2016 | Overall Dependency Hierarchy between Grammatical Systems: 1. Polarity (I) 2. Tense; Aspect; Evidentiality (II) 3. { Person; Reference classification (gender etc.) (III) 4. { Number (III) } 5. Case (IV) } This basically indicates that a grammatical system may be dependent on any system above it in the hierarchy. The lower system may have a different set of choices available according to which term is chosen from the higher system. The braces indicate mutual dependencies (i) between person and reference classification, and number; and (ii) between number and case. Further expectations: 1. Mood should not be dependent on aspect, number, or case; mood should not be dependent on polarity, nor vice versa. 2. Definiteness is probably to be added quite low in the hierarchy, just above Case and below Number, or at the same level as Case or Number (Definiteness may depend on Number, and vice versa; Definiteness may depend on number, and vice versa). |
inflection | implicational hierarchy | polarity, tense, aspect, evidentiality, person, gender, number, case, neutralization, syncretism |
| 2017 | When a system does show functional markedness it is generally the case that (i) when the number of choices in the system is reduced, in a particular context, the unmarked term will always be retained, and will have its semantic scope extended; and (ii) if the choices in system X depend on that made in system Y, we will always expect more choices in X to be associated with the unmarked than with the marked terms from Y.Markedness distributions:Polarity: positive is functionally and almost always formally unmarked, and negative marked;Number: singular is almost always functionally and often also formally unmarked;Person: if any person, then 3rd;Tense, Aspect: often no unmarked term, or crosslinguistically variable;Case: generally case that covers intransitive subject function is unmarked, but sometimes also accusative;Reference classification: often masculine unmarked, sometimes feminine. | inflection | no genuine implication; rather: provided that | markedness, polarity, tense, aspect, evidentiality, person, gender, number, case |
| 2018 | IF there is switch-reference marking, THEN alignment for (all) syntactic purposes is accusative. | inflection, syntax | implication | switch-reference, alignment, accusative |
| 2019 | (i) IF there are classifiers for other contrasts, THEN there are ones for the contrast human/nonhuman or animate/inanimate;(ii) IF there are genders or noun classes for other contrasts, THEN there are ones for the contrast human/nonhuman or animate/inanimate.(iii) There is a contrast human/nonhuman or animate/inanimate recognized by some grammatical forms (such as interrogative pronouns) or rules. | inflection, syntax | implication | classifier, gender, animacy |
| 2020 | There are no (clear cases of) long-distance reciprocals. | syntax | unconditional | reciprocal, binding |
| 2021 | There is a word for FEELING (verb, adjective, or noun), undifferentiated between "bodily feelings" (sensations) and "cognitively based" feelings ("emotions"), and distinct from THINKING. | lexicon | unconditional | sensation, emotion, cognition, word class |
| 2022 | Some feelings can be described as "good" and some as "bad" (while some may be viewed as neither "good" nor "bad"). | lexicon | unconditional | sensation, emotion |
| 2023 | There are words comparable, though not necessarily identical in meaning, with cry/weep and smile/laugh; that is words referring to bodily expression of good and bad feelings. | lexicon | unconditional | emotion, sensation, body |
| 2024 | There are "emotive" interjections (i.e. interjections expressing cognitively based feelings). | syntax, lexicon | unconditional | emotion, sensation, cognition, interjection |
| 2025 | There are some "emotion terms" (i.e. terms designating some cognitively based feelings). | lexicon | unconditional | emotion, sensation, cognition |
| 2026 | There are words linking feelings with (i) the thought that "something bad can happen to me", (ii) the thought that "I want to do something", and (iii) the thought that "people think something bad about me", that is words overlapping (though not identical) in meaning with the English words afraid, angry, and ashamed. | lexicon | unconditional | emotion, sensation |
| 2027 | (i) Cognitively based feelings can be described via observable bodily "symptoms" (that is, via some bodily events regarded as characteristic of these feelings). (ii) In all languages, cognitively based feelings can be described with reference to bodily sensations, general or localized. (iii) In all languages, cognitively based feelings can be described via figurative "bodily images" (referring to imaginary events and processes taking place inside the body). |
semantics | unconditional | emotion, sensation, cognition, body, metaphor |
| 2028 | There are alternative grammatical constructions for describing (and interpreting) cognitively based feelings, conceptualizing "emotions" as "involuntary" or as "uncontrollable", or as "overwhelming" and "irresistible", or as "active" and in some sense "voluntary", and so on. | syntax, semantics | unconditional | cognition, sensation, emotion, synonymy |
| 2029 | The main universal mode for describing cognitively based feelings is in terms of a comparison, that is, via LIKE, and in this, the main human strategy for talking about feelings is analogous to the main human strategy for talking about colours. | lexicon | unconditional | emotion, sensation, cognition, colour, metaphor |
| 2030 | IF there is Scrambling, then basic order is SOV. | syntax | implication | order, SOV, scrambling |
| 2031 | IF there are word-initial geminates, THEN there are also medial geminates. | phonology | implication | consonant, geminate, word |
| 2032 | IF there are voiced word-initial geminate stops, THEN there are also voiceless word-initial geminate stops. | phonology | implication | consonant, geminate, voice |
| 2033 | IF there are contour tones in place X on any of the hierarchies in (a)-(d), THEN there are contour tones also lower on each hierarchy:(a) syllable type: CVV >> CVR >> (CVO) >> CV(b) number of syllables: monosyllabic >> bisyllabic >> polysyllabic(c) prominence: stressed >> unstressed, root >> affix syllable(d) tone: R >> F >> H, L | phonology | implicational hierarchy | tone, contour tone |
| 2034 | Generally agreed upon "facts" about contour tones (marked = more complex, rarer): (a) contour tones are more marked than level tones; (b) rising tones are more marked than falling tones; (c) rising-falling and falling-rising tones are more marked than rising and falling tones. Markedness generalizations:The more marked a tone is, (i) the more likely it is absent in a tone system;(ii) the more likely it is to be restricted to a hospitable tone bearing unit (e.g. long, prominent, sonorous). |
phonology | implication | tone, contour tone, level tone, markedness |