Lexical function

Last updated

A lexical function (LF) is a tool developed within Meaning-Text Theory for the description and systematization of semantic relationships, specifically collocations and lexical derivation, between particular lexical units (LUs) of a language. [1] [note 1] LFs are also used in the construction of technical lexica (Explanatory Combinatorial Dictionaries) and as abstract nodes in certain types of syntactic representation. Basically, an LF is a function ƒ( ) representing a correspondence ƒ that associates a set ƒ(L) of lexical expressions with an LU L; in f(L), L is the keyword of ƒ, and ƒ(L) = {L´i} is ƒ’s value. Detailed discussions of Lexical Functions are found in Žolkovskij & Mel’čuk 1967, [2] Mel’čuk 1974, [3] 1996, [4] 1998, [5] 2003, [6] 2007, [7] and Wanner (ed.) 1996; [8] analysis of the most frequent type of lexical functions—verb-noun collocations—can be found in Gelbukh & Kolesnikova 2013. [9]

Contents

Standard Lexical Functions

Standard LFs form a proper subset of normal LFs. A normal LF ƒ is called Standard if and only if it satisfies both following conditions:

1. Broadness of the domain of ƒ: ƒ is defined for a relatively large number of keywords;
2. Diversity of the range of ƒ: ƒ has a relatively large number of expressions as elements of its possible values and these expressions are more or less equitably distributed between different keywords.

Normal LFs that do not satisfy both Conditions 1 and 2, on the one hand, and degenerate LFs, on the other, are called Non-Standard. An example of a Non-Standard LF is the meaning ‘without addition of dairy product’. It has two expressions in English, a phraseological one—BLACK (with COFFEE: black coffee), and a free one—WITHOUT MILK (tea without milk is not *black tea). This meaning fails Condition 1: it is too specific and applicable only to one beverage. It thus corresponds to a Non-Standard LF.

Simple Standard LFs

1. Syn [Lat. synonymum] = synonym.

Syn(helicopter) = copter, chopper
Syn(telephoneV) = phoneV

2. Anti [Lat. antonymum] = antonym.
3. Convijk [Lat. conversivum] = conversive.

This LF returns for L an LU L´ with the same meaning as L but with its Deep Syntactic Actants (roughly, syntactic arguments) i, j and k permuted —for example, the DSyntAs k, i and j of L are permuted in L´ such that [i→k, j→i, and k→j].
Conv21(include) = belong
Conv231⊃(opinion) = reputation
Conv21(behind) = in front of
Conv21(precede) = follow

4. Gener [Lat. genus] = the closest generic concept for L.

The value of this LF must appear in one of the following two constructions:
1) ‘Gener(L)−ATTR→DER(L)’ = ‘L’ [where DER is any DSynt-derivative, see 6–9 below]; or
2) L, X1, X2, ..., Xn and other (kinds of)Gener(L).
Gener(republic) = state [republican state = republic]
Gener(liquidN) = substance [liquidAsubstance = liquidN]
Gener(arrestN) = reprisals [arrests and other (kinds of) reprisals]

5. Figur [Lat. figuraliter ‘figuratively’] = standard received metaphor for L.

Figur(fog) = wall [wall of fog ≈ fog]
Figur(rain) = curtain [curtain of rain ≈ rain]
Figur(remorse) = pangs [pangs of remorse ≈ remorse]

6. S0 = Substantival, output N having a congruent meaning to L (which can be of any part of speech except N):

S0(analyze) = analysis

7. A0 = Adjectival, output A having a congruent meaning to L (which can be of any part of speech except A):

A0(city) = urban

8. V0 = Verbal, output V having a congruent meaning to L (which can be of any part of speech except V):

V0(analysis) = analyze

9. Adv0 = Adverbial, output Adv having a congruent meaning to L (which can be of any part of speech except Adv):

Adv0(followV [N]) = after [N]

10. Si = standard name of the i-th (Deep-Syntactic) actant of L.

For the verb TEACH: ‘Person X1 teaches subject Y2 to people Z3
S1(teach) = teacher
S2(teach) = subject/matter [in high school]
S3(teach) = pupil
For the noun LETTER: ‘Letter by person X to person Y about Z’
S1(letter) = author, sender [of the letter]
S2(letter) = addressee [of the letter]
S3(letter) = contents [of the letter]

11. Sinstr = standard name of the instrument used in the situation denoted by L.

Sinstr⊃(shoot) = firearmSinstr(murderV,N) = murder weapon

12. Smed = standard name of the means used to bring about the situation denoted by L.

Smed⊃(shoot) = ammunition

13. Smod = standard name of the mode through which the situation denoted by L is realized.

Smod(consider [an issue]) = approach [I consider this issue  ... ~ My approach to this issue ...]

14. Sloc = standard name of the location where the situation denoted by L is realized.

Sloc(fightV [two armies]) = battlefieldSloc(war) = theater (of war)

15. Sres = standard name of the result of the situation denoted by L.

Sres⊃(learn) = knowledge, skillsSres⊃(explosion) = shockwaveSres⊃(copyV) = copyN, reproduction

16. Ablei [Lat. habilis ‘able, manageable’] = determining property of the i-th potential DSyntA of L (‘such that it can L easily’/‘such that it can be L-ed easily’):

Able1(cryV) = tearfulAble1(vary) = variableAble2(prove) = provableAble2(trustV) = trustworthy

17. Quali [Lat. qualitas] = determining property of the i-th probable DSynt-actant of L (‘such that it is predisposed to L’/‘such that it is predisposed to be L-ed’):

Qual1(cryV/N) = sadQual1(laughV/N) = cheerfulQual2(doubtV/N) = implausibleQual2(laughV/N) = awkward, absurd

18. Ai = determining property of the i-th DSyntA of L from the viewpoint of its role in the situation ‘L’.

A1 is semantically roughly equivalent to an active participle (≈ ‘which is L-ing’), and A2 to a passive participle (≈ ‘which is being L-ed’):
A1(anger) = in [anger] //angryA1(speed) = with [a speed of ...]  A2(analyze) = //under analysisA2(conduct [an orchestra])= //under the baton [of N]

19. Advi = the determining property of the action by the i-th DSyntA of L from the viewpoint of the role of the DSyntAi of L in the situation denoted by L.

Adv1 is semantically roughly equivalent to an active verbal adverb (≈ ‘while L-ing’), and Adv2, to a passive verbal adverb (≈ ‘while being L-ed’):
Adv1(anger) = with [~] //angrilyAdv1(decreaseN,V) = //down [… a decrease of 2.7% = ... down 2.7%.]  Adv2(applause) = to [the ~]   Adv2(bombard) = //under bombardment [They came under heavy bombardment.]

20. Imper [Lat. imperāre ‘[to] command’] = imperative expression meaning ‘do L!’

Imper(shoot) = Fire!Imper(speak low) = Shhh!Imper(stop [to a horse]) = Whoa!

21. Result [Lat. resultāre ‘[to] result’] = ‘[to] be the expected result of L’:

Result(buyV) = ownVResult(lie down) = be lyingResult(have learnt) = know [how],  have the necessary skills

22. Centr [Lat. centrum ‘center’] =‘the center/culmination of L’

Syntactically, Centr(L) is a noun that takes the name of L as its DSyntA II: Centr−II→L, etc.
Centr(forest) = the thick [of the ~] Centr(crisis) =  the height [of the ~] Centr(glory) = summit [of ~] Centr(life) = prime [of ~]

23. Magn [Lat. magnus ‘big, great’] = ‘very’, ‘to a (very) high degree’, ‘intense(ly)’:

Magn(naked) = starkMagn(laughV) = heartily; one’s head offMagn(patience) = infiniteMagn(skinny [person]) = as a rake

24. Ver [Lat. verus ‘real, genuine’] = ‘as it should be’, ‘meeting intended requirements’:

Ver(surprise) = sincere, genuine, unfeignedVer(punishment) = well-deserved, justVer(instrument) = preciseVer(walkV) = steadily

25. Bon [Lat. bonus ‘good’] = ‘good’:

Bon(cutV) =  neatly, cleanlyBon(proposal) =  tempting Bon(service) = first-class Bon(assistance) =  invaluable

26. Locin [Lat. locus ‘place’] = preposition governing L and designating a containing spatial location (‘being in’):

Locin(height) = at [a height of ...]

27. Locad [Lat. locus ‘place’] = preposition governing L and designating an entering spatial relation (‘moving into’):

Locad(height) = to [a height of ...]

28. Locab [Lat. locus ‘place’] = preposition governing L and designating an exiting spatial relation (‘moving out of’):

Locab(height) = from [a height of ...]

29. Instr [Lat. instrumentum ‘instrument’] = preposition meaning ‘by means of L’:

Instr(typewriter) = on [ART ~] Instr(satellite) = via [~] Instr(mail) = by [~] Instr(argument) = with [ART ~]

30. Propt [Lat. propter ‘because of’] = preposition meaning ‘because of’/‘as a result of L’:

Propt(fear) = from, out of [~] Propt(love) = out of [one’s ~ of ...]

31. Operi [Lat. operāri ‘[to] do, carry out’] = a light verb used as part of the expression ‘realize/carry out L':

The DSyntA I of this verb (and its Subject) is the expression that is described in the Government Pattern of L as the i-th DSyntA of L, and Operi’s DSyntA II (= its Primary Object) is L itself.
Oper1(blowN) = [to] deal [ART ~ to N] Oper1(supportN) = [to]  lend [~ to N] Oper2(blowN) = [to] receive [ART ~ from N] Oper2(supportN) = [to] receive [~ from N] 

32. Funci [Lat. functionāre ‘[to] function’]:

The DSyntA I of this verb (and its Subject) is L itself, and its DSyntA II (its Primary Object) is the i-th DSyntA of L.
Func1(blowN) = comes [from N] Func2(blowN) = falls [upon N]
If Funci has no object, the subscript 0 is used:
Func0(snowN) = fallsFunc0(option) = is openFunc0(preparations) = are under wayFunc0(rumors) = circulate

33. Laborij [Lat. labōrāre ‘[to] work, toil’]:

The DSynt-actant I of this verb (and its Subject) is the i-th DSyntA of L, its DSyntA II (its Primary Object) is the j-th DSyntA of L, its DSyntA III (= its Secondary Object) is the j+1-st DSyntA of L, and its further DSyntA (its TertiaryObject) is L itself.
Labor12(interrogation) =  [to] subject [N to an ~] Labor32(leaseN) =  [to] grant [N to N on ~]

34. Incep [Lat. incipere] = ‘begin’:

Incep(sleep) = fall asleep

35. Cont [Lat. continuāre] = ‘continue’:

Cont(sleep) = stay asleep

36. Fin [Lat. fīnīre] = ‘cease’:

Fin(sleep) = wake up

37. Caus [Lat. causāre] ‘cause’ [≈ ‘do something so that a situation occurs’]
38. Perm [Lat. permittere] =‘permit/allow’ [≈ ‘do nothing which would cause that a situation does not occur]’
39. Liqu [Lat. *liquidāre] = ‘liquidate’ [≈ ‘do something so that a situation does not occur’]
40. Reali [Lat. realis ‘real’]

Real1(accusation) = [to] prove [ART ~]  Real1(car) = [to] drive [ART ~ ] Real1(illness) = [to] succumb [to  ART ~] Real2(law) = [to] abide [by ART ~] Real2(hintN) = [to] take [ART ~] Real2(demandN) = [to] meet [ART ~]

41. Fact0/i [Lat. factum ‘fact’]

42. Labrealijk [a hybrid of Labor and Real]

43. Involv [Lat. involvere ‘[to] drag along’] = verb meaning ‘[to] involve Y’, ‘[to] affect Y’

Involv links L and the name of a non-participant Y which is affected or acted upon by the situation ‘L’; Y is DSyntA II of Involv, and L is its DSyntA I:
Involv(lightN) = floods [N = Y, e.g. the room] Involv(snowstorm) = catches [Nhum=Y Locin N], hits [Narea= Y]

44. Manif [Lat. manifestāre ‘[to] manifest’] = verb meaning ‘L manifests itself [≈ becomes apparent] in Y’

The keyword L, a noun, is DSyntA I of Manif, and Y (= in which L manifests itself) is its DSyntA II:
Manif(doubt) = nagsManif(joy) = lights up [his eyes]

45. Degrad [Lat. degradāre ‘[to] degrade’] = verb meaning ‘[to] degrade’ ≈ ‘[to] become permanently worse or bad’.

Degrad takes its keyword L, which can be any noun, as its DSyntA I.
Degrad(milk) = goes sourDegrad(meat) = goes offDegrad(discipline) = weakensDegrad(house) = becomes dilapidated

46. Son [Lat. sonāre ‘[to] sound’] = verb meaning ‘[to] emit characteristic sound’.

Son also takes its keyword, which most often, but not necessarily, is a concrete noun, as its DSyntA I.
Son(dog) = barksSon(battle) = rumblesSon(banknotes) = rustleSon(wind) = howls

Complex LFs and Configurations of LFs

Simple LFs can be combined to form complex LFs:

AntiMagn(applause) = scatteredIncepOper1(loveN) = [to]  fall [in ~] Adv1Real1(whim) = on [a ~]

Certain expressions (collocations or derivations) can simultaneously realize two LFs, resulting in a configuration of LFs:

[Magn + Oper1](doubt) = [to] be plagued [by ~] (≈ ‘[to] have [= Oper1] strong [= Magn] doubts’) [Ver + Oper1](health) = [to] have a clean bill [of ~] (≈ ‘[to] have [= Oper1] good [= Ver] health’)

LFs in the lexicon

Lexical Functions play an important part in the lexicon, which of necessity must include information about the collocational and derivational properties of LUs. In MTT, the LFs for L are included in the entry for L in the Explanatory Combinatorial Dictionary:

REVULSION
Definition
X’s revulsion for Y ≡ ‘X’s (strong) negative feeling about Y caused by X’s perception of Y, similar to what people normally experience when they perceive something that makes them sick and such that it causes that X wants to stop perceiving Y’.

Government Pattern

X = IY = I
1. N’s
2. Aposs
3. A
1. against N

2. at N
3. for N
4. toward N

1) CII.2 : N denotes something that can be seen or felt
2) CII.4 : N denotes people
[‘C’ stands for ‘column;’ the Roman numeral identifies the column, and the Arabic numeral, the cell in it.]
John’s 〈his〉 revulsion against racism 〈against dismal results of his endeavors〉. John’s 〈his〉 revulsion at the sight of sea food; John’s 〈his〉 revulsion for work 〈for all those killings〉; John’s 〈his〉 revulsion for 〈= toward〉 these scoundrels 〈toward the government〉
Impossible:
John’s 〈his〉 revulsion *at these words [correct expression: ... for these words] [by Constraint 1]
John’s 〈his〉 revulsion *towards these words [by Constraint 2]

Lexical Functions

Syn: distaste; repugnance; repulsion;
    disgust; loathing
  

Anti∩: attraction

Conv21Anti∩: appeal  A1: revulsed
Able2: repulsive  Magn : deep, extreme
AntiMagn : mild  Adv1: in [~]
Propt: from [~]  Oper1: experience, feel [~]
Magn + Oper1: be filled [with ~]  Magn + Labor12: fill [N with ~]
Adv1Manif: with [~]    

Examples He did it from deep revulsion for the bitterness of the sectarian strife. Any revulsion they might feel from fat-ass bastards they ran up against professionally was ad hominem and not ad genus [A. Lurie]. Mary turned her head away in revulsion. I felt no revulsion for her maternal fantasies, only a practical concern. She met his advances with revulsion. Pam was driven to revulsion (by the sight of the dead animal) 〈*The sight of the dead animal drove Pam to revulsion〉. Revulsion at slaughter cut war short [newspaper heading].

Notes

  1. The material in this article is based to a large extent on Mel’čuk, Igor A. (2007). Lexical Functions. In H. Burger, D. Dobrovol’skij, P. Kühn & N. Norrick (eds.), Phraseology. An International Handbook of Contemporary Research, 119–13. Berlin/New York: W. de Gruyter. The material in the list of LFs and the sample ECD entry below is reproduced from that source with kind permission of the author.

Related Research Articles

A syntactic category is a type of syntactic unit that theories of syntax assume. Word classes, largely corresponding to traditional parts of speech are syntactic categories. In phrase structure grammars, the phrasal categories are also syntactic categories. Dependency grammars, however, do not acknowledge phrasal categories.

Idiom Combination of words that has a a literal meaning

An idiom is a phrase or expression that typically presents a figurative, non-literal meaning attached to the phrase; but some phrases become figurative idioms while retaining the literal meaning of the phrase. Categorized as formulaic language, an idiom's figurative meaning is different from the literal meaning. Idioms occur frequently in all languages; in English alone there are an estimated twenty-five thousand idiomatic expressions.

In mathematics and computer science, a higher-order function is a function that does at least one of the following:

In grammar, the instrumental case is a grammatical case used to indicate that a noun is the instrument or means by or with which the subject achieves or accomplishes an action. The noun may be either a physical object or an abstract concept.

Collocation Frequent occurrence of words next to each other

In corpus linguistics, a collocation is a series of words or terms that co-occur more often than would be expected by chance. In phraseology, collocation is a sub-type of phraseme. An example of a phraseological collocation, as propounded by Michael Halliday, is the expression strong tea. While the same meaning could be conveyed by the roughly equivalent powerful tea, this expression is considered excessive and awkward by English speakers. Conversely, a corresponding expression in technology, powerful computer, is preferred over strong computer. Phraseological collocations should not be confused with idioms, where an idiom's meaning is derived from its convention as a stand-in for something else while collocation is a mere popular composition.

Greenlandic language Eskimo-Aleut language spoken in Greenland

Greenlandic is an Eskimo–Aleut language with about 57,000 speakers, mostly Greenlandic Inuit people in Greenland. It is closely related to the Inuit languages in Canada such as Inuktitut. It is the most widely spoken Eskimo-Aleut language.

Crow language

Crow is a Missouri Valley Siouan language spoken primarily by the Crow Nation in present-day southeastern Montana. The word, Apsáalooke, translates to "children of the large beaked bird." It is one of the larger populations of American Indian languages with 2,480 speakers according to the 1990 US Census.

c-command is a relationship between the nodes of grammatical syntax trees. A working definition of c-command is that node X c-commands node Y if a mother of X dominates Y. C-command is closely associated with the generative phrase structure grammars of the Chomskyan tradition, and is not applicable to the tree structures of other theories of syntax, such as dependency grammars. C-command relations have served as the basis for many explorations and explanations of phenomena within the field of syntax. It has been taken to be the basic configurational relation underlying binding, and has played a central role in the analysis of diverse syntactic mechanisms, such as parasitic gaps and the scope of quantifiers.

In linguistics, an adverbial phrase ("AdvP") is a multi-word expression operating adverbially: its syntactic function is to modify other expressions, including verbs, adjectives, adverbs, adverbials, and sentences. Adverbial phrases can be divided into two types: complement adverbs and modifier adverbs. For example, in the sentence She sang very well, the expression very well is an adverbial phrase, as it modifies the verb to sing. More specifically, the adverbial phrases very well contains two adverbs, very and well: while well modifies the verb to convey information about the manner of singing, very is a degree modifier that conveys information about the degree to which the action of singing well was accomplished.

In linguistics, phraseology is the study of set or fixed expressions, such as idioms, phrasal verbs, and other types of multi-word lexical units, in which the component parts of the expression take on a meaning more specific than or otherwise not predictable from the sum of their meanings when used independently. For example, ‘Dutch auction’ is composed of the words Dutch ‘of or pertaining to the Netherlands’ and auction ‘a public sale in which goods are sold to the highest bidder’, but its meaning is not ‘a sale in the Netherlands where goods are sold to the highest bidder’. Instead, the phrase has a conventionalized meaning referring to any auction where, instead of rising, the prices fall.

Igor Melčuk

Igor Aleksandrovič Mel'čuk, sometimes Melchuk, is a Russian and Canadian linguist, a retired professor at the Department of Linguistics and Translation, Université de Montréal.

In linguistic morphology, the bracketing paradox concerns morphologically complex words which have more than one analysis, or bracketing, e.g., one for phonology and one for semantics, and the two are not compatible, or brackets do not align.

A phraseme, also called a set of thoughts, set phrase, idiomatic phrase, multi-word expression, or idiom, is a multi-word or multi-morphemic utterance where at least one of whose components is selectionally constrained or restricted by linguistic convention such that it is not freely chosen. In the most extreme cases, there are expressions such as X kicks the bucket ≈ ‘person X dies of natural causes, the speaker being flippant about X’s demise’ where the unit is selected as a whole to express a meaning that bears little or no relation to the meanings of its parts. All of the words in this expression are chosen restrictedly, as part of a chunk. At the other extreme, there are collocations such as stark naked, hearty laugh, or infinite patience where one of the words is chosen freely based on the meaning the speaker wishes to express while the choice of the other (intensifying) word is constrained by the conventions of the English language. Both kinds of expression are phrasemes, and can be contrasted with ’’free phrases’’, expressions where all of the members are chosen freely, based exclusively on their meaning and the message that the speaker wishes to communicate.

Meaning–text theory (MTT) is a theoretical linguistic framework, first put forward in Moscow by Aleksandr Žolkovskij and Igor Mel’čuk, for the construction of models of natural language. The theory provides a large and elaborate basis for linguistic description and, due to its formal character, lends itself particularly well to computer applications, including machine translation, phraseology, and lexicography.

Norbert Schmitt is an American linguist and a Professor of Applied Linguistics at the University of Nottingham in the United Kingdom. He is known for his work on second language vocabulary acquisition and second language vocabulary teaching. He has published numerous books and papers on vocabulary acquisition.

An explanatory combinatorial dictionary (ECD) is a type of monolingual dictionary designed to be part of a meaning-text linguistic model of a natural language. It is intended to be a complete record of the lexicon of a given language. As such, it identifies and describes, in separate entries, each of the language's lexemes and phrasemes. Among other things, each entry contains (1) a definition that incorporates a lexeme's semantic actants (2) complete information on lexical co-occurrence ; (3) an extensive set of examples. The ECD is a production dictionary — that is, it aims to provide all the information needed for a foreign learner or automaton to produce perfectly formed utterances of the language. Since the lexemes and phrasemes of a natural language number in the hundreds of thousands, a complete ECD, in paper form, would occupy the space of a large encyclopaedia. Such a work has yet to be achieved; while ECDs of Russian and French have been published, each describes less than one percent of the vocabulary of the respective languages.

Aramba (Arammba), also known as Serki or Serkisetavi, is a Papuan language of Papua New Guinea. It is spoken to the south of Western Province in the Trans Fly region. Aramba belongs to the Tonda Sub-Family, which is next to the Nambu Sub-Family region and the Suki language. Alternative names for the language include Upper Morehead, Rouku, Kamindjo and Tjokwasi.

Upper Necaxa Totonac is a native American language of central Mexico spoken by 3,400 people in and around four villages— Chicontla, Patla, Cacahuatlán, and San Pedro Tlaloantongo —in the Necaxa River Valley in Northern Puebla State. Although speakers represent the majority of the adult population in Patla and Cacahuatlán, there are very few monolinguals and few if any children are currently learning the language as a mother tongue, and, as a consequence, the language must be considered severely endangered.

Wandala, also known as Mandara or Mura', is a language in the Chadic branch of the Afro-Asiatic language family, spoken in Cameroon and Nigeria.

Irarutu language

Irarutu, Irahutu, or Kasira, is an Austronesian language of most of the interior of the Bomberai Peninsula of north-western New Guinea in Teluk Bintuni Regency. The name “Irarutu” comes from the language itself, where “ira” conjoins with “ru” to create “their voice”. When put together with “tu”, which on its own means “true”, the meaning of the name becomes “Their true voice” or “The people’s true language”.

References

  1. Fontenelle, Thierry. (2008) Using a bilingual dictionary to create semantic networks. In Thierry Fontenelle (ed.), Practical Lexicography: A reader, 175–185. Oxford: Oxford University Press.
  2. Žolkovskij, Aleksandr & Igor A. Mel’čuk, I. (1967). O semantičeskom sinteze. Problemy kibernetiki 19, 177–238.
  3. Mel’čuk, Igor A. (1974). Opyt teorii lingvističeskix modelej «Smysl ⇔ Tekst». Moscow: Nauka.
  4. Mel’čuk, Igor A. (1996). Lexical Functions: A tool for the description of lexical relations in the lexicon. In Leo Wanner (ed.), Lexical Functions in Lexicography and Natural Language Processing, 37–102. Amsterdam: John Benjamins.
  5. Mel’čuk, Igor A. (1998). Collocations and Lexical Functions. In Anthony P. Cowie (ed.) Phraseology. Theory, analysis, and applications, 23–53. Oxford: Clarendon.
  6. Mel’čuk, Igor A. (2003). Collocations dans le dictionnaire. In Th. Szende (ed.), Les écarts culturels dans les dictionnaires bilingues, 19–64. Paris: Honoré Champion.
  7. Mel’čuk, Igor A. (2007). Lexical Functions. In H. Burger, D. Dobrovol’skij, P. Kühn & N. Norrick (eds.), Phraseology. An International Handbook of Contemporary Research, 119–131. Berlin/New York: W. de Gruyter.
  8. Wanner, Leo. (ed.) (1996). Lexical Functions in Lexicography and Natural Language Processing. Amsterdam/Philadelphia: John Benjamins.
  9. Gelbukh, A.; Kolesnikova, O. (2013). "Semantic Analysis of Verbal Collocations with Lexical Functions". Studies in Computational Intelligence. 414. doi:10.1007/978-3-642-28771-8. ISBN   978-3-642-28770-1.Cite journal requires |journal= (help)