A taxonomy of antonymy in Arabic: Egyptian and Saudi proverbs in comparison

This study has set out to identify, quantify, typify, and exemplify the discourse functions of canonical antonymy in Arabic paremiography by comparing two manually collected datasets from Egyptian and Saudi (Najdi) dialects. Building upon Jones’s (2002)most extensive and often-cited classification of the discourse functions of antonyms as they co-occur within syntactic frames in news discourse, the study has substantially revised this classification and developed a provisional and dynamic typology thereof. Two major textual functions are found to be quantitatively significant and qualitatively preponderant: ancillarity (wherein an A-pair of canonical antonyms project their antonymicity onto a more important B-pair) and coordination (wherein one antonym holds an inclusive or exhaustive relation to another antonym). Three new functions have been developed and added to the retrieved classification: subordination (wherein one antonym occurs in a subordinate clause while the other occurs in a main clause), case-marking (wherein two opposite cases are served by two antonyms), and replacement (wherein one antonym is substituted with another). Semicanonical and noncanonical guises of antonymy are left and recommended for future research.


Introduction
Proverbs compact everyday experiences and common observations in a succinct and formulaic language that renders them easy to remember and retrieve in social interaction (Mieder 2004). Arabic proverbs are no exception and their central and prevalent linguistic tool for doing so is canonical antonymy, a lexicalsemantic relation held by pairs of words considered to be conventional opposites in neutral contexts and hence perceived to be typical cases of canonical antonyms in and out of contextantonyms which elicit one another in free word association tasks and when no context is available (Deese 1964;Murphy 2003;Paradis et al. 2009;Jones et al. 2012).
Under proverb collection (i.e., paremiography), Mieder (2004: 12-14) includes proverbial expressions ("to bite the dust"), comparisons ("as busy as a bee"), interrogatives ("Does a chicken have lips?"), twin formulas ("give and take"), and wellerisms ("'Each to his own,' as the farmer said when he kissed his cow"). Mieder (2004: 6) has logged structural patterns of proverb chunks in terms of their syntagmatic co-occurrences, but confined them only to English, including "Better X than Y," "Like X, like Y," "No X without Y," "One X doesn't make a Y," and "If X, then Y," as in "Better poor with honour than rich with shame," "Like father, like son," "No work, no pay," "One robin doesn't make a spring," and "If at first you don't succeed, then try, try again." Coinnigh (2014) confirmed these structural aspects of proverbs by analyzing their commonest formulae (e.g., X is Y; No X, no Y, etc.) and their syntactic markers (e.g., syntactic parallelism, emphatic word order, parataxis, etc.) with few examples from (non-)European languages to provide a linguistic overview of proverbial style and structure in world languages.
"The numerous proverb collections make it possible to study proverbs on a comparative basis" (Mieder 2004: xii). Comparative paremiology has attracted scholarly attention to proverbial differences across two distinct languages (cf. Dabbagh 2016). Arabic proverbs were examined from several different linguistic perspectives, e.g., cognitive metaphor (Tohamy n.d.), semantic translation (Assaqaf 2019;Fahmi 2016;Naoum 2007), and sociopragmatics (Alghamdi 2019;Mahzari 2017) within Arabic and across Arabic and English. The only paremiological study found to be most related to the one at hand, albeit across Romanian and Russian, is Gheltofan (2015) who identified implicit and explicit antonymic or contrastive structures at a paremiological level under the label "paremiological antonyms" and their discourse categories based on their lexicosyntactic frameworks.
As far as this paremiolinguistic study is concerned, no sole work was conducted on the textual functions of antonyms in Arabic paremiographical discourse. There is a sharp lack of interest in the lexicosyntactic configurations of antonyms and the lexicosyntactic frames in which these antonyms tend and prefer to co-occur. This serious lacuna gives rise to the topic and makes it worth investigation, especially in a comparative context across Arabic dialects: Egyptian and Saudi Arabian. The rationale behind undertaking this inquiry consists in approaching antonymy afresh as a key information structuring mechanism for organizing proverbial thoughts in memorizable and easily retrievable patterns.
Therefore, this study sets out to evidentially champion the structural aspects of proverbs across Egyptian and Saudi Arabic dialects, by typologically comparing the discourse functions of canonical antonyms in Egyptian and Saudi proverbs and by synergizing and supporting the structural paremiological classifications proposed by Mieder (2004) and Coinnigh (2014) with the discursive categories and syntactic frames of antonyms developed by Jones (2002). The objective is twofold.
(1) To identify and quantify the syntactic frameworks of canonical antonymy and their relevant discourse functions across Egyptian and Saudi proverbs.
(2) To exemplify and qualify the syntactic frames and their pertinent textual functions across Egyptian and Saudi proverbs.
To fulfil these two objectives, the study raises the following questions and pursues a quest for answers. (1) What are the typical syntactic frames that host canonical antonyms in Egyptian and Saudi proverbs?
(2) What are the representative discourse functions that canonical antonyms seek to perform therein?
(3) What are the lexical-semantic implications of antonym discourse functions for proverb translation?
In its quest to achieve the objectives and answer the questions above, this study puts forward a hypothesis that Arabic paremiographical antonyms shall contribute, besides the discourse functions of antonyms developed and standardized by The Comparative Lexical Relations Group (Complexica Project)¹ across English (Jones 2002), Swedish ), and Japanese (Muehleisen and Isono 2009), new data-driven and data-based discourse functions which might be retrieved and replicated with different datasets across languages and can rigorously update the current relevant literature and enrich its state of the art. The members of the group have, since 2002/2005, aspired to expand their typological studies on the functions of antonyms in discourse to a wider range of languages and linguistic contexts. So, the present study seeks to provide taxonomic evidence for such and more antonym categories in an un(der)studied cross-varietal Arabic paremiographical context. We tend in particular to prove that antonymic pairs with gradable canonicity exist in Arabic paremiography, thereby contributing new data-driven and data-based categories and validating the cross-linguistic and possibly universal prevalence of the phenomenon.

Background
This study concentrates on canonical antonymy. By "canonical," scholars mean that some word pairs are said to be better or more conventional antonyms than others (Jones et al. 2012;Murphy 2003). For example, "large/small" and "big/little" are perceived to be more typical cases of canonical antonymy than "large/little" and "big/small." Antonymous pairs such as the former two are typically called "canonical antonyms" (Murphy 2003), while the latter two, having low innate opposition, are considered less canonical antonyms. When the pairs have no innate opposition, they are dubbed "noncanonical oppositions" (Davies 2012). Paradis et al. (2009) regard members of canonical pairs as the ones which freely elicit each another and whose elicitations are shared by a majority, if not a totality, of native speakers. They remain in canonical opposition even out of context. Murphy (2003) and Paradis et al. (2009) argue that a canonical pair that holds semantic opposition is "conventionalized" as a pair in language, contrary to noncanonical oppositional pairs (Davies 2012(Davies , 2013.
Canonical antonymy has been the locus of lexicosemantic studies from Aristotle to date. Hassanein (2020b) tabulates all given typologies of the so-called canonical antonymy. Jones et al. (2012: 6) mention that the modern thought about antonymy dates back to the categories of propositional opposition created by Aristotle, who developed a diagrammatic representation ("the square of opposition") for universal and particularized affirmations and negations (e.g., Every S is P; No S is P) and introduced terminologies, particularly "contradictory" and "contrary," which have iteratively been adopted in contemporary linguistic, particularly the structuralist, approaches. Lyons (1968Lyons ( , 1977Lyons ( , 1995 and Cruse (1976Cruse ( , 1986Cruse ( , 2000 are among the most prominent semanticists who divide canonical antonyms or opposites into four types: (1) Complementaries: opposites dividing conceptual domains into two mutually exclusive compartments, as in "alive/dead." (2) Contraries: gradable antonyms showing degrees of some lexical-semantic field, as in "fast/slow." (3) Reversives: directional opposites denoting change in opposite directions between two states, as in "fall/rise." (4) Converses: relational opposites yielding the same proposition if they are reversed, as in "buy/sell." Canonical antonymy has been examined by the traditionalists as a paradigmatic syntax-free relation based on an axis of choice or selection, but by the modernists (e.g., Fellbaum 1995;Jones 2002;Mettinger 1994) as a syntagmatic syntax-based relation according to an axis of chain or combination.  and Davies (2012) mention that Jones's (2002) seminal study on antonym classification is the most comprehensive, which is why his typology has rigorously been retrieved and replicated by many subsequent studies on Swedish   (Hassanein 2018(Hassanein , 2020a, and Persian (Mohammadi et al. 2019). Insightful perspectives on the canonicity of antonymy can exclusively be found in Jones et al. (2012: 43-70) and van de Weijer et al. (2014).

Methodology
This section introduces two cross-dialectal datasets gathered for a comparative analysis and pathways set as a toolkit for classification of antonym functions and frames across the datasets in comparison.

Datasets
Classical, Modern Standard, and Vernacular Arabic corpora are extremely sparse, if not scant. Insofar as Arabic is concerned and to the best of our knowledge, there is no Arabic paremiographical corpus available for mining and extracting canonical antonyms with the automatic corpus-based method. The lack of a pertinent corpus leaves us totally incapable of using corpus tools to identify and quantify highly cooccurring canonical antonym pairs. Hence, we are left with no choice but to resort to the well-founded and often-cited dictionaries of proverbs in the two dialects in comparison. Were the proverbial corpus even available, the automatic method for an Arabic corpus would be problematized with a number of limitations. First, the automatic method fails to extract antonymous pairs, whereby oppositeness is built between Xantonyms construed in the text and Y-antonyms ellipted in the text but recovered from the context. AlHedayani (2016: 39) supports this argument with her statement: "However, in a corpus study where co-occurring pairs of antonyms are searched for, such sentences (that is, sentences including ellipted Yantonyms) will not be retrieved." Second, the automatic method is likely to extract lexically elocuted opposites, but unlikely to spot semantically intended ones that the former pragmatically instantiate, a distinction drawn by Hurford et al. (2007) between a word meaning and speaker meaning. Last, the automatic method is unable to log noncanonically contrasted pairs, "noncanonical oppositions" (Davies 2012: 42), hosted by the same syntactic frames, particularly the ancillary ones (cf. Jones 2002: 47). Hsu (2015: 58) sums up the grounds for choosing the manual extraction of mining contrastive constructions over the automatic: (a) human judgment is very needed in some cases (i.e., of less canonical and noncanonical oppositions) and (b) some antonym functions are frameless but have formal features necessitating manual intervention.
Given the unavailability of an Arabic paremiographical corpus and the reasons mentioned above, we are decided to follow Davies (2012) and Hsu (2015) in taking a manual data-analysis stance, not arguing, nevertheless, against the usefulness of the automatic method for data extraction. Using spreadsheet, we have created two paremiographical datasets from two linguoculturally divergent dialects. The first dataset has manually been gathered from the Book of Al-ʔAmtha:l Al-c A:mmiyya ("Vernacular Proverbs") written in Egyptian Arabic by Aẖmad Taymu:r Pasha (2014), explained and arranged alphabetically. According to critical reviews, "Vernacular Proverbs" is a leading antiquarian Book regarded as numero uno in the Egyptian paremiography and comprised of 3,000 proverbs that encapsulate the wisdoms and traditions of eras and generations in a fixed succinct and terse but compendious and sententious form. It has also become the most dependable source of Egyptian lifestyles, customs, traditions, and thoughts and is loaded with explanatory details of the proverbs in their sociocultural contexts.
The second dataset has manually been collected from a ten-volume Book of Al-ʔAmtha:l Ash-Sha c biyya fi: Qalb Al-Jazi:ra Al-c Arabiyya ("Folk Proverbs in the Heart of the Arabian Peninsula") written in Saudi Arabic by c Abdulkari:m Al-Juhayma:n (1982), also in an alphabetical order. "Proverbs in the Heart of the Arabian Peninsula" (2nd ed.) chronicle 10,000 Arabian proverbs and is considered the most comprehensive Book on Arabian paremiography. In a prologue to the second edition of the Book, the author claims to have collected the proverbs as vocally and typically articulated and uttered by the emitters, the folk inhabitants of the Arabian Peninsula, strictly of Najd, the geographical central region of today's Saudi Arabia. He has arranged the proverbs alphabetically, from A to Z, not only in the proverbs per se but also in the initial word of each proverb, excluding proverbs which include a (porno)graphic language against public decorum and socioreligious norms. He also provides an explicatory sociohistorical context of each proverb (cf. Al-Juhayma:n 1982 [P1]: 9-17). In the epilogue to this multivolume Book, reviewers claim it to be a chronicle of proverbs no less important than ancient ones and a solid base for sociocultural, socioeconomic, sociohistorical, and sociolinguistic research, sourced firstly from folk poetry and secondly from folktales (cf. Al-Juhayma:n 1982 [P10]: 464-484). With particular reference to Arabic paremiological discourse and paremiographical heritage, regional variations between Egyptian Arabic and Saudi Arabic loom larger at different linguistic levels owing to geographical separation and lack of language contactdialectal variations which are still markers of regional and local identities and which feature in syntax, phonology, morphology, orthography, and lexicology (see Omar and Alotaibi 2017 for extensive details).
The Egyptian Arabic (ARE) dataset includes 417 canonical pairs of proverbial antonyms, whereas the Saudi Arabic (KSA) dataset comprises 564 pairs. The two datasets are characterized by codemixing and intertextualizing their proverbs with three distinct varieties of Arabic: Vernacular Arabic (VA), Modern Standard Arabic (MSA), and Classical Arabic (CA). One example from either dataset may suffice to show the codemixing of Arabic varieties (CA, MSA, and VA) into single proverbs. Consider, e.g., the firedescribing Najdi proverb ʔum c A:bis ta:kul al-ratb wa-l-ya:bis ("Mother of c A:bis eats the wet and the dry") which undertakes a dialectal deviation of the CA or MSA glottalized verb taʔkul "eats" into the deglottalized VA form ta:kul and which is intertextually reminiscent of the Qur'anic Arabic verse wa-la: ratb wa-la: ya:bis ʔilla: fi: kita:b ("… or anything, fresh or withered, that is not written in a clear Record" (Qur'an 6:59, Abdel Haleem 2004: 84). Likewise, the Egyptian proverb surtak ʔi:h surtak ʔiyya:k ("What is your chapter? Your chapter is It is You") which codemixes the CA forms surtak/ʔiyya:k "your chapter/It is You" with the VA form ʔi:h "What?" and which is suggestive and indicative of the first chapter of the Qur'an: ʔiyya:k na c bud wa-ʔiyya:k nasta c i:n ("It is You we worship; it is You we ask for help" (Qur'an 1:5, Abdel Haleem 2004: 3). During the extraction of both datasets, we observed a preponderant co-occurrence of less canonical and noncanonical oppositionsa co-occurrence hypothesis that Arabic paremiography tends to structure and organize the proverbs in terms of dichotomous contrasts. These less canonical and noncanonical oppositions, though worth investigation, do not fall within the scope of the present study and thus have not been quantified.

Method
A tale of two methods can be told in respect of previous research on antonymy: (I) the (overused) paradigmatic method and (II) the (underused) syntagmatic method. Antonymy is a syntagmatic and paradigmatic relation (Jones et al. 2012;Murphy 2003). The paradigmatic method has been fathered by Lyons (1977) and Cruse (1986) and falls beyond the methodological scope of this study. The syntagmatic method, which exclusively falls into the scope of this study, has seminally been initiated by Deese (1964) as well as Lehrer and Lehrer (1982), relatively confirmed by Fellbaum (1995), syntactically framed by Mettinger (1994), but discoursally categorized by Jones (2002). Using a corpus-based approach, Jones (2002) has managed to shift the focus of studies on canonical antonymy from introspective, subjective, and context-free intuitions to extrospective, objective, and context-based insights. The present study seeks to champion the syntagmatic approach to canonical antonymy since a few studies have shown that canonical antonyms tend to co-occur in quantifiable and classifiable contrastive constructions far more than noncanonical opposites (Davies 2013;Jones 2002;Jones et al. 2007Jones et al. , 2012Mettinger 1994).
This study builds its method on the typological analysis of antonym functions in discourse as most extensively conducted by Jones (2002) and rigorously retrieved and replicated by his fellow members of the Comparative Lexical Relations Group and other scholars across several different languages (as shown above), being the most comprehensive study to date (cf. Davies 2012: 45;Murphy 2009: 27-28) and the largest systemic corpus approach to antonym co-occurrence (AlHedayani 2016: 25). Jones (2002), presumably inspired by his predecessors' scholarly insights into the syntagmatic profiling of canonical antonymy (Fellbaum 1995;Mettinger 1994), explored by quantification and exemplification the syntagmatic co-occurrences of antonyms in syntactic structures, using a corpus-based approach to describe what canonical antonymy actually does in discourse, not how it looks like from a logical perspective. To identify, quantify, and exemplify the textual functions of canonical antonymy, Jones (2002) constructed a database of 3,000 sentences retrieved from a corpus of 280 million words from The Independent. The result was developing eight discourse functions or categories of antonymy in news discourse tabulated below in alphabetical order with typical examples in Table 1.
The categories Jones (2002) identified, quantified, and exemplified are claimed, rather shown, to be more or less common in different genres of texts and talks (cf. Murphy 2009: 28). It is widely acknowledged that canonical antonymy can perform different textual functions in other corpora and may accommodate various genre-based and data-driven syntactic frameworks. Mieder (2004), the world's foremost paremiologist (cf. Puglia 2019), observes the occurrence of English proverbial chunks, antonyms are certainly no exception, into structural patterns on the syntagmatic axisan observation in line with a co-occurrence hypothesis that antonym pairs co-occur in syntactic frames far more frequently than chance would allow (Justeson and Katz 1991: 18). Insofar as the present study is concerned, the discourse functions of antonyms in Arabic paremiography have not been examined before, which constitutes the main rationale for undertaking this inquiry. This study sets out to quantitatively and qualitatively test Jones's (2002) typology of antonymy against the two datasets in comparison so as to support Mieder's (2004) observation and synergize a provisional typology thereof.

Analysis
This section presents both qualitative and quantitative analyses of major and minor discourse functions of canonical antonymy across the Egyptian and Saudi datasets.

Qualitative analysis
This subsection is concerned with classifying and exemplifying the discourse functions of canonical antonyms based on their co-occurrences in structurally parallel and culturally common syntactic frames across stretches of Arabic paremiographical discourse. A by-product of the quantitative approach is that it provides a good potential for examining proverbial antonymy and its ideological repercussions in crosscultural discourse. We identify and typify the quantified categories of canonical antonyms in qualitative terms below, with two illustrative examples of each category, one Egyptian and the other Saudi Arabian, in Arabic, followed by a tentative literal translation to serve the lexicosyntactic purpose of the study. The categories are presented according to frequency, i.e., from the most to the least frequent.

Ancillary antonymy
According to Jones (2002: 45-60), ancillary antonymy solely consists of two pairs of antonyms designated as A-pair and B-pair. The A-pair is an auxiliary catalyst that plays an ancillary role in treating the B-pair (more) contrastively. If the B-pair has no innate contrast, the A-pair generates an instantial contrast. If the B-pair has low contrast, the A-pair sharpens this contrast. If the B-pair has high contrast, the A-pair confirms this patent contrast to the point of antonymity. By doing so, the B-pair is nudged further up the scale of opposition by the A-pair. Ancillary antonymy is said to use no specific frames, but to import them from other categories, as in "X and Y" from coordinated antonymy, "X and not Y" from negated antonymy, "who(ever) X, Y" from subordinated antonymy, and so forth. One typical example from Jones (2002: 45-46) is "I love to cook but I hate doing the dishes" in which the contrastive antonyms love/hate are linked to another contrast between the catchier contrastive non-antonyms cook/doing the dishes.
(1) kiti:r al-haraka ʔali:l al-baraka much the.movement little the.blessing "Much movement, little blessing." (Taymu:r 2014: 371) (2) f-t-taʔanni: as-sala:ma w-f-l-c ajala an-nada:ma In.the.caution the.safety and.in.the.haste the.repentance "In caution there is safety and in haste repentance." Proverb (1) features two pairs in an "X, Y" framework imported from simultaneous antonymy: an A-pair of canonical opposites kiti:r/ʔali:l ("much/little") that imparts contrast to a minimal B-pair of al-haraka/albaraka ("movement/blessing") not regarded as opposites in neutral contexts. Contrary to the old Arabic proverb fi: al-haraka baraka (loosely, "in movement there is a blessing;" freely, "He who travels harvests goods"), proverb (1) is reminiscent of the English proverb "A rolling stone gathers no moss" that describes a much-traveling person as a thriftless rolling stone, who leads a wandering, unsettled life but eventually gains so little. This mere description figures prominently in two other stylistically variable adages: kathi:r al-kara:t qali:l al-bara:t and saba c sana:yi c wi-l-bakht da:yi c . Proverb (2) hosts two pairs of antonyms in an "X and Y" framework borrowed from coordinated antonymy: a canonical A-pair at-taʔanni:/al-c ajala ("caution/haste") that activates and sharpens a latent contrast between semicanonical opposites of a B-pair assala:ma/an-nada:ma ("safety/repentance"). This proverb is an invitation to "make haste slowly" by acting with diligence and caution and be "better safe than sorry." Albeit with different phraseological opposites and syntactical frames, all the proverbs deem it wiser to be so careful than to be so hasty that one may later regret because "haste makes waste."

Coordinated antonymy
Coordinated antonymy joins or disjoins a pair of opposites by using junctive and disjunctive coordinating and correlative conjunctions to expound inclusivity, exhaustivity, or negativity of the compartments of a certain domain. Common syntactic frames that host coordinated antonyms include "and, or, neither…nor, either…or, whether…or," or any other coordinating or correlative conjunctions. One representative instance from Jones (2002: 71) is "If a school with bad exam results says it is, nevertheless, producing fine people, we can neither agree nor disagree" in which the boldfaced pair of verbs are negated to signal inclusivity and intimate neutrality of scale.
( Proverb (3) features a pair of coordinated antonyms, hna:k/hina: ("there/here"), in a neither-nor syntactic frame that negatively exhausts the lexical field of space.² The proverb is quoted to characterize a person as being spiritless and useless for real tasks, a characterization paralleled by other propositionally similar proverbs, such as la: fu:ʔ wa-la: taht ("neither up nor down") and la: fi:sh wa-la c li:sh ("neither Fi:sh nor c Li:sh")³ in VA and la: salla: alla:h c alayh wa-la: sallam ("May God neither bless nor save him") and la: fi:-l-c i:r wa-la: fi:-n-nafi:r ("Neither in the caravan nor in the battle")⁴. Proverb (4) comprises just the same propositional content and oppositional pair of Proverb (3) but in reverse order, resulting in an attentioncalling chiasm or antimetabole that is more culturally equivalent to the English counterpart "Neither here nor there" (i.e., "of no importance, significance, or relevance").

Subordinated antonymy
Subordinated antonymy, a category foreign to Jones's (2002) typology, has had its inception in Hassanein (2013) and has been rigorously replicated in MSA (AlHedayani 2016) and CA (Hassanein 2018) datasets. The category features a protasis-apodosis or subordinate-superordinate clause structure: one antonym appears in a dependent clause appended sequentially, consequentially, conditionally, or concessively to a main or independent clause in which the other antonym resides, i.e., within an "if X then Y" or "when X then Y frame." Many validated cases feature in AlHedayani (2016: 40-131) who classifies the structure according to the meaning it conveys in Arabic news discoursea meaning of consequence, as in "Then hatred prevails and fear spreads when love is absent", wherein hatred spreads when love is absent in that the absence of the latter causes the spread of the former. Subordinating conjunctions work actively but distinctly here. Proverb (5) opposes a pair of canonical antonyms, tizra c uh/tiʔla c uh ("sow/reap"), within a whatever X then Y frame. The proverb states that one reaps whatsoever one sows and shall bear the consequences of one's own actions, a restatement of an MSA proverb ka-ma: tazra c tahsud. It is a VA remodeling of a CA counterpart, ka-ma: tadi:n tuda:n, which puts a pair of case antonyms, tadi:n/tuda:n ("convict/be convicted"), into voice (active vs. passive) opposition and which is representative of the Biblical proverb "As you sow, so shall you reap" and the Qur'anic verse "Is favor not the only recompense of favor?" (55:60). The last two proverbs also employ a subordinate framework "As X, Y". Proverb (6) features a canonical opposition between a pair of antonyms, shabb/sha:b ("grow up/grow old"), typical of physical development from childhood to adulthood to old age. The profound meaning of this adage is that habit ever remains (at-tab c yaghlib atta-tabbu c ), old habits die hard, the leopard cannot changes its spots, and what has always been will always be, a meaning oxymoronically signified in Wordsworth's paradoxical line "Child is father of the man" in his poem "My Heart Leaps Up" also known as "The Rainbow."

Negated antonymy
Negated antonymy records the co-occurrence of a canonically antonymous pair within a framework negating one antonym, nominals in the first place, in favor or emphasis of the other, the most favorite of which is X and not Y (amounting to Jones's (2002: 88) frame X not Y). At deep structure, the category is plausibly overlapping with replacive antonymy (X instead of Y) and preferentially comparative antonymy (X (more) adj-er than Y), in which one antonym is chosen over or weighed against the other antonym, respectively. Jones (2002: 88) provides a typical example in "However, the citizen pays for services to work well, not badly" that uses not badly to affirm well and counteract any suggestion that the services should work badly. (7) at-tasha:sh wa-la: al-c ama: the.dim-sightedness and.not the.blindness. "Dim-sightedness, and not blindness."  (7) hosts two semicanonical antonyms, at-tasha:sh/al-c ama: ("dim-sightedness/blindness"), in an X and not Y frame, in which the X-antonym is prioritized and chosen over the Y-antonym. The locus of this proverb is that it is better to have less than one expects than to have nothing at all, as also packed elsewhere (Taymu:r 2014: 454) in nuss al-c ama: wa-la: al-c ama: kulluh ("Better sand-blind than stoneblind," "Half a loaf is better than no bread," and "In the country of the blind, the one-eyed man is king."). Purblindness is considered a bliss in contrast with stone-blindness, a situation in which there is a latent call for people to be content with having something rather than nothing. Proverb (8) retrievably replicates the same negated framework X and not Y, hosting a morphologically related pair of antonyms, fa:l/la: fa:l ("omen/not omen"), wherein the negative marker la: ("not") negates the stimulus X-word fa:l "omen" to constitute the response Y-word la: fa:l "not omen." This VA proverb is said or cited when someone hears from someone else something portending evil and replies in a preventive manner: "God/ Heaven forbid." The central theme here is to presage good omen from God rather than bad omen from someone. The word fa:l (deglottalized) is a vernacular reformulation of the standard word faʔl (glottalized), which is intrinsically evocative of optimism. Unlike the antonym pair members that are lexically opposed and negated in proverb (7), those counterparts in proverb (8) are morphologically negated with the negative particle la: ("not") and thus opposed. The latter members, fa:l/la: fa:l "omen/not omen," involve a morphological process of antonym formation that creates the opposite of an existing root/base by adding a negative prefix morpheme to it and also a semantic process by producing a new opposite sense (cf. Jones 2002: 90 (58c), Hurford et al. 2007: 228, Hsu 2015.

Case-marked antonymy
Case-marked antonymy is the second category that is foreign to Jones's (2002) classes and that has originated in Hassanein (2013) and has been rigorously retrieved and replicated in Hassanein (2018). Hassanein (2013: 212) calls it "case antonymy" which consists of a pair of syntactically or semantically case-opposed words, as in subjective vs. objective, active vs passive, agent vs patient, etc. In light of this study, casemarking denotes syntactic functions or semantic roles served by the arguments (Jackendoff 1972: 39) or referring expressions (Hurford et al. 2007: 257) surrounding the verb in a sentence. Like ancillary antonymy, case-marked antonymy favors no specific frames, but accommodates its frames from other categories, particularly coordinated antonymy and comparative antonymy frameworks, as in X and Y and X (more) adj-er than Y, respectively. Hassanein (2018: 37) gives a prototypical instance "The interrogee is not more knowledgeable than the interrogator" in which the case roles make the question-answer adjacency pair applicable to all the interrogators and interrogees about the Hour from the time of interrogation onwards.  Proverb (9) borrows a correlative syntactic frame, neither X nor Y, from coordinated antonymy to host a pair of utterances, t c a:yirni:/a c a:yrak ("taunt me/taunt you"), marking opposition of case roles that represent ways in which the (pro)nouns associated with the verb (e.g., subjects, objects, and complements) are semantically related to the meaning of the verb (cf. Cruse 2006: 68). The tacit subject pronouns ("you/I") and object pronouns ("me/you") following the verb "taunt" are semantically related to the verbs as actor/ experiencer.⁵ The meaning of the proverb is that we ought not to castigate other people for bad characteristics which we have ourselves. More wisely, "people who live in glass houses shouldn't throw stones" simply because they "are all in the same boat." Proverb (10) employs a comparative proposition which accommodates the frame "X (more) adj-er than Y," from comparative antonymy, wherein two nominals, al-masʔu:l/as-sa:ʔil ("interrogee/interrogator"), serve two opposite syntactic roles, ʔism al-maf c u:l/ʔism alfa: c il ("passive participle/active participle"), represented by the two contrastive case roles of recipient and actor, respectively. This proverb is an instance of intertextuality, being interdiscursively grounded in a historical narrative in the collective body of verbal and actional traditions of the prophet Muhammad technically known as "Hadi:th" or "Sunna." Anecdotally, in a gathering between the prophet of Islam and his companions, Gabriel, the angel of revelation, appeared to the prophet in human shape to question him about the date of Resurrection Day (cf. Hassanein 2018: 37). The prophetic reply literally was: "The  5 Unarguably, case or thematic roles are definitionally opaque, cognitively controversial (Rissman and Majid 2019: 1850-1851), terminologically overlapping, and typologically interventional. Hence, this study adopts the three universally relevant clusters of semantic functions (macro-roles) developed by Keizer (2015: 133) because they are presumed to be relevant for all world languages. questionee does not know more about it than the questioner," a reply proverbialized standardly as ma: al-masʔu:l bi-a c lam min as-sa:ʔil and vernacularly as c ilmi: c ilmak (amounting idiomatically to "Your guess is as good as mine"). The motivations for such a response are (a) to prove the questioner more knowledgeable than the questionee, (b) to evade the answer, (c) to prove the questionee more deficient in information than the questioner, and (d) to avoid embarrassment or entrapment (cf. Al-Juhayma:n 1982 [P7]: 163).

Comparative antonymy
Comparative antonymy gauges or measures one canonical antonym against the other in a positive and elative context, employing frameworks such as X (more) adj-er than Y and X (not) as/like Y. The X-antonym is weighed against the Y-antonym in terms of a semantic scale or lexical field. One example from Jones (2002: 78) is "The new bills are more colourful than the old ones" that provides the scale of color against which the new bills can be compared to the old ones.  Proverbs (11 and 12), share the same propositional content, but with semantically and logically distinct modalities. Both replicate an identical pair of canonical antonyms, sudfa/mi: c a:d ("accident/date"), within an X (more) adj-er than Y framework along a graded lexical and semantic field of goodness. Proverb (11) is epistemically too emphatic in its statement that all accidental meetings are better than dates. The epistemic modality of the proverb stands out in the exhaustive quantifier kull ("every or all") that includes all possibilities without exception: "Every accidental meeting is better than a date." Proverb (12) is the closer equivalent of the two adages to the English proverb "An accidental meeting may be better than a date," being equivalently paralleled by an alethic preposition rubba ("perhaps or maybe"), which denotes uncertainty and attenuates the force of the statement. By comparison, the two proverbs are syntagmatically identical (preposition + subject + predicate + complement), but paradigmatically distinct in the choice of epistemic adverbials kull and rubba, which signify certitude and doubt, respectively. In both proverbs in comparison, the componential analysis of the Arabic-English opposed members, sudfa/mi: c a:d "accident/ date," reflects their antonymicity based on the distinguishing semantic feature "±planned" in that an accident is an unplanned and unspecified meeting, whereas a date is a planned and specified appointment.

Distinguished antonymy
Distinguished antonymy denotes a metalinguistic difference between a pair of canonical antonyms within a v X from Y or n between X and Y frame, where v and n are lexical differentials or synonyms thereof. There is an allusion to an inherent semantic distinction between the opposite words. "The reference is metalinguistic because the "speaker" presupposes our familiarity with "opposites" in order to make a more general statement" (Jones 2002: 81). The most favored framework in this category is v X from Y that Jones (2002: 83) exemplifies with "The forces must no longer discriminate between married and unmarried partners" whose verb phrase is used to distinguish between married [partners] and unmarried partners. Proverb (13) builds the free syntactic framework "v X from Y" (loosely, "be distinct are X and Y") to host a pair of antonyms, ar-radi:/al-muna:sib ("bad/good"), the message of which is to advise marriage partners to inspect each other's conduct, lifestyle, and history to decide whether or not they will be a good match by marriage. If they do so, they will be able to "separate the good from the bad" and choose the perfect match, Mr./Mrs. Right. Proverb (14) is not propositionally going far from the preceding proverb. It replicates the same framework "v X from Y" (loosely, "know X from Y") to host a pair of canonical antonyms, c aduww/sidi: ("enemy/friend"), the purport of which is to "separate the sheep from the goats" and "the wheat from the chaff," that is, to distinguish friends from enemies and frenemies. There is an admonishment against people who pretend to be friends but they are the worst foesan admonishment pan-Islamically intertextualized with the Qur'anic verse: "There is [a kind of] man whose views on the life of this world may please you [Prophet], he even calls on God to witness what is in his heart, yet he is the bitterest of opponents" (Qur'an 2:204, Abdel Haleem 2004: 23).

Simultaneous antonymy
Simultaneous antonymy, in which a pair of antonyms denote the same referent at the same time and the order of antonyms differ in languages due to sociocultural values, is said to be commoner in some languages than in others (cf. AlHedayani 2016: 8), as in Swedish more than in English ) and in Japanese (Muehleisen and Isono 2009) more than in Arabic (AlHedayani 2016). The paired antonyms are directly equated with each other in contexts which designate the applicability of X and Y to the same referent, as in "Mr Amato's weakness is his strength," in which weakness and strength operate along different semantic scales in that Mr Amato's weakness is presumably a personal weakness, but his strength is a political strength (Jones 2002: 99). Simultaneous antonymy favors an asyndetic and equative framework "X, Y," according with AlHedayani's (2016: 151) finding that simultaneity is expressed in Arabic with three different grammatical forms, two of which are equation and asyndeton. Proverb (15) is wisely explicit about God's omnipresence, employing an equative (cf. Hurford et al. 2007: 42) or equational (cf. AlHedayani 2016: 151) sentence with an "X, Y" frame equivalent to the English frame "X is Y" and hosting a canonical pair of spatially deictic antonyms, hina:/hna:k ("here/there"). The proverb is quoted in the contexts of separation, exile, and expatriation from homeland to an unknown place. The gist of the proverb is that God whose unsleeping eyes are watching over people in every place is everywhere (here and there) to protect and sustain them. It is a compendious call for having strong faith and trust in the Lord. Proverb (16) weaves a web of intertextual relations to the Qur'anic discourse, conjuring up many of interdiscursive strands, as in "Know too that God is severe in punishment yet most forgiving and merciful" (Qur'an 5:98, Abdel Haleem 2004: 77). However, it does so syndetically by inverting and reordering the two equative adjectival phrases for a proverbially positive purpose, within a simultaneously synchronous frame hosting a canonically opposite pair that characterizes the same referent, God, as being coincidentally ghafu:r rahi:m/shadi:d al-c iqa:b ("forgiving, merciful/severe in punishment"). The purpose of the proverb is that one should not wallow in aspirations and hopes without taking necessary precautions, but, rather, should maintain or redress the balance between fear of and expectation from Godfear of punishment as a restraint and expectation for forgiveness as a motive (cf. Al-Juhayma:n 1982 [P5]: 233). Such proverbs preach God's ubiquity, unity, and ability to punish or forgive.

Transitional antonymy
Transitional antonymy, as Jones (2002: 85) puts it, features "The co-occurrence of an antonymous pair within a framework that expresses a movement or change from one location or state to another," as in "How easy to slip from the legal to the illegal trade." Hassanein (2013Hassanein ( , 2018 broadens its scope to comprise a transition in space (from one place to another), time (from one period to another), and state (from one condition to another). Typical frames are "from X to Y" and "v X into Y." Proverb (17) features a spatiotemporal transition in a "from X to Y" framework hosting a semantically canonical pair of opposites, taʔtaʔ/as-sala:m c alaykum (literally, "knock/peace-be-upon-you;" idiomatically, from A to Z, from beginning to end, or from start to finish"). The substance of the proverb is basically to satisfy someone's curiosity about others' affairs and the inquisitive interest in knowing the whole issue from the outset to the close. Another contextual clue is to praise someone's considerable expertise in a particular field. Proverb (18) uses a "from X (in)to Y" framework to host a lexically canonical pair of conversive antonyms, al-ba:yi c /al-mishtiri: ("seller/buyer"), and highlight a transition to a different place and state of possession directly from the seller to the buyer. The proverb is part of a will made by a deceased man in Wushi:qar town (cf. Al-Juhayma:n 1982 [P8]: 250) in al-Qasi:m region, whose people are well-known for being conservative and proactive. One of their precautionary measures is this adage which refers to the seller-to-buyer delivery of an oblation on the eve of the Greater Bairam to be damage-free and get offered to God on the next day.⁶ Since then, this proverb has been quoted to manage reciprocal transactions and mutual agreements between sellers and buyers to prevent potential losses.

Interrogative antonymy
Interrogative antonymy has first appeared in Jones and Murphy (2005), rigorously retrieved and replicated in subsequent studies, such as Jones (2006), ), Muehleisen and Isono (2009: 2197 who call it "disjunctive antonymy," and Davies (2012: 89) who dubs it "binarized option." Interrogative antonymy makes a choice between two opposite alternatives within an interrogative frame, X or Y?, whereby the occurrence of either alternative excludes the occurrence of the other, as in "Is she a good mommy or a bad mommy?" (Jones and Murphy 2005: 413). This study widens the range of this category to include all antonyms co-occurring within a junctive or disjunctive interrogative frame. Proverb (19) is a rhetorical question raised by the speaker in surprise and not intended for a reply from the interlocutor. A canonical opposition is established between a minimal pair of items, hilm/ c ilm ("dream/ reality"), within a rhetorically interrogative frame "X or Y?." At its surface structure, the question potentiates one answer at the expense of the other owing to the disjunctive coordinator wa-lla: ("or"), although its deep structure votes for c ilm ("reality") rather than hilm ("dream"). The proverb holds when something unexpected transpires or when an undesired person pops up against a speaker's wish or expectationa meaning close to the idiomatic string "I couldn't believe my eyes." Proverb (20) retrieves the same frame "X or Y?" to pose a normal question in which the speaker gives the listener a choice between a good alternative, al-janna ("paradise/heaven"), and a bad alternative, an-na:r ("fire/hell"). The listener, narratively a deranged nomad or Bedouin, chooses a third alternative out of question, c aysh ("bread"), because of starvation. When he was told that there is bread in paradise or heaven, he chose the paradise. Since then the proverb has also been cited to describe a person who is too obsessed with something to see anything else or who thematically deviates from the subject matter of the question to a matter he is preoccupied with (cf. Al-Juhayma:n 1982 [P5]: 240).

Replacive antonymy
Replacive antonymy, a third category foreign to Jones's (2002) typology and named after Davies's (2012: 54) "replacive opposition," features in AlHedayani (2016: 182) with the same label and in Hassanein (2013: 208) as "exchanged antonymy." According to Davies (2012), replacive antonymy lies somewhere between  6 The Greater Bairam (Arabic, c I:d al-ʔAdha:) is the second of the two Muslim feasts held after Ramadan, a feast of sacrifice which is celebrated throughout the Islamic world to mark the end of hajj ('pilgrimage') and on which financially able Muslims have to offer sacrifices by slaughtering religiously permissible and edible cattle (cf. Newby 2002: 95). negation and comparison and involves a co-occurrence of two antonyms within a framework that substitutes and exchanges an antonym for another. Typical frames include "v X with Y" and "X rather than Y." The v is a substitutive verb, as in replace, substitute, exchange, or a synonym thereof, as typified in "Mr Shevardnadze stressed that work for peace rather than war should prevail" which shows a preference for one choice over another in a binary manner (Davies 2012: 54 Proverb (21) features a misplacement whereby a vainglorious and affectatious boy wears summer clothes rather than winter clothes in winter. The misplacement is carried out within the frame "X in place of Y" which hosts the antonymous pair as-sayf/ash-shita: ("summer/winter"). The purpose is to mock the consideration of the boy's act of wearing summer clothes rather than winter clothes in winter as vanity. This misplacement is a norm-breaking and swimming-against-the-stream act. Proverb (22) hosts the opposite pair hala:l/hara:m ("licit/illicit") within an "X rather than Y" frame, in which the speaker implores God to bestow halal wealth rather than haram wealth on him or her. The goal is to entreaty God for halal livelihood with which we become content and which makes us dispense with all haram livings, because earning a haram living or consuming haram foods or drinks is a very strong reason for severing the relationship between the Lord and his servant (cf. Al-Juhayma:n 1982 [P10]: 43). This meaning draws on few divine commandments in the Qur'an (Q, 2:168, 2:172, 7:157, 23:51) and is further broached by the prophet of Islam in one of his authentic traditions (cf. An-Nawaw: 2009).

Quantitative analysis
The statistical analyses of the discourse functions and syntactic frameworks of antonym co-occurrences in the Egyptian and Saudi database proverbs demonstrate that the role canonical antonymy plays therein is highly significant and dominant.
As Figures 1-3 Fellbaum 1995), are added, and Arabic (Hassanein 2018). Coordinated antonymy comes second in order, taking 103 (18%) and 103 (24.7%) across KSA and ARE datasets, respectively, according with findings in previous studies in which it interchangeably ranks first or second in contrast with ancillary antonymy. Subordinated antonymy, first developed in Hassanein (2013) and then revalidated in AlHedayani (2016) and Hassanein (2018), stands out as the third largest category in Arabic paremiography and is assigned 35 (6.2%) in KSA and 41 (9.8%) in ARE. The remaining categories, negated, case-marked, comparative, distinguished, simultaneous, transitional, interrogative, and replacive, are attributed marginal frequency distributions (collectively 23.4% in KSA and 12.4% in ARE) compared with the major ones. A riveting finding is the perfect coincidence between the top-down, bottom-up orders of quantified categories across both datasets, which reveals a correspondence between Egyptian and Saudi cultural heritages. From both the quantification and categorization of paremiographical antonyms in ARE and KSA, it is shown that six of Jones's (2002) major antonym categories (ancillariness, coordination, negation, comparison, distinction, and transition) have been retrieved and replicated with(in) the two datasets in comparison, with preponderances of the first two. Nonetheless, two of his eight antonym categories (extremity and idiomaticity) occupy no or null frequency distribution in the datasets in contrast and are coincidently supplanted with half-Jonesian and non-Jone-   Jones's (2002) categories of extreme antonymy and idiomatic antonymy have gotten no frequency distributions in Arabic paremiographies, which supports Hassanein's (2018, 42) argument that antonym classification is data-based and genre-specific. This argument is further attested by diverse typologies developed by scholars, notably Mettinger (1994), Jones (2002), Davies (2012), AlHedayani (2016), and Hassanein (2013Hassanein ( , 2018. For an informative study on these different typologies, see Hassanein (2020b: 21-34).
As shown in Figures 4 and 5, the coordinative frame "X and Y" is the preponderant syntactic structure that records the highest frequency distributions: 227 (40.2%) in KSA and 123 (29.4%) in ARE. Next to it is the negative frame "X and not Y" which obtains the second highest frequencies in KSA and ARE datasets, 86 (15.2%) and 60 (14.3%), respectively. Third in terms of frequency and productivity is the subordinative frame "X, Y" which is mainly represented by a multiplicity of protasis-apodosis clause structure ("if X then Y") and subordinate-superordinate structure ("when X then Y").⁷ The remaining frameworks occupy lower frequencies, some of which are interchangeably specific to either dataset, as in the frame idha: X, Y ("if X, Y") unique to KSA dataset (6 in number, 1%) and the frame ya: X ya: Y ("either X or Y") peculiar to ARE dataset (5 in number, 1.1%). A null reading (0) is assigned to such cases wherein one data-specific frame has a zero frequency in the other.

Conclusion
This article has set out to identify, quantify, and exemplify the discourse functions canonical antonymy serves in Arabic paremiography, by comparing and contrasting two datasets collected manually from Egyptian Arabic and Saudi Arabic proverbial traditions. Based on their co-occurrences within syntactic frames acting as triggers of canonical (and also noncanonical, cf. Davies 2012: 69) oppositions, antonyms are shown to perform eleven functions in Arabic paremiographical discourse, the most important of which include ancillary antonymy and coordinated antonymy. Previous studies have proven these two major  7 Needless to say, the entire analysis is based on the co-occurrences of canonical antonyms in Arabic frames and the given translations are tentative in that the same frame might serve different discourse functions based on context, e.g. 'X and Y' can perform ancillary, coordinate, case-marking, and distinguishing functions. Besides, a single English translation, such as 'neither X nor Y', amounts to different Arabic frames: la: X wa-la: Y, ma: X wa-la: Y (CA/MSA), and mush X wa-la: Y (VA).
categories to be interchangeably ranked first or second over all the remaining classes across many languages, such as English (Jones 2002(Jones , 2006(Jones , 2007(Jones , 2008 . Nevertheless, ancillary antonymy is the most frequent in this study because cross-categorial pairs have been included, a finding concurring with previous studies by Fellbaum (1995: 294) and Lobanova et al. (2010: 25).
Then it is unsurprising that the frames of the two major categories dominate over those of the remaining ones. "It is a fact of which the linguist must take cognizance, that binary opposition is one of the most important principles governing the structure of languages; and the most evident manifestation of this principle is antonymy" (Lyons 1977: 271). Thus, paremiological antonyms play a quantitatively significant and qualitatively pivotal role in Arabic paremiographical discourse because antonyms evoke each other in individual or collective memory. This may explain why Arabic paremiography richly and basically uses canonical antonyms, besides less canonical and noncanonical oppositions, for building the syntagmatic and paradigmatic structures of Arabic proverbs in focus. The quantitative and qualitative analyses of proverbial antonyms across the two datasets in contrast show a general attitude towards structuring and organizing the informational substance of the proverbs in binary oppositionsa finding in line with Lyons's (1977) observation of a human tendency to organize experience in terms of dichotomous contrasts.    Figure 5: English syntactic frames of canonical antonymy in KSA and ARE datasets. Gheltofan (2015: 73) supports this very argument with instances from Romanian paremiology which reaffirm that in a paremiological context there are two major types of antonyms, viz. canonical and noncanonical, considered concrete and viable guises of both systemic and extra-systemic paremiological antonymy. Davies (2012: 42) contends that the canonical status of oppositions ranges in a gradable cline from canonicity to non-canonicity, with semi-canonicity, we would claim, as a midterm in between. By "canonical," we mean a pair of opposites considered by the native speakers of a certain language, the so-called "clang phenomenon" (cf. Muehleisen 1997: 4), to be conventional antonyms in neutral contexts and out of context, as in "hot/cold" (cf. Murphy 2003). By "semi-canonical," we mean a pair of words are less conventional antonyms, as in "cold/warm." By "noncanonical," we arguably mean a pair of words regarded as contextually bound opposites, nonconventional antonyms in natural language use, as in "watch/listen" in "I don't know whether to watch or listen" (cf. Davies 2012). However, not all the proverbs in the Arabic paremiographies in comparison employ canonical antonymy as a proverb-forming mechanism; there are certainly some which do not and whose quantification is beyond scope and out of question herean argument in compliance with Permakov (1975) who argues that not all proverbs are organized based on semantic contrariety and in agreement with Danilov (1995) who develops two main categories of proverb, oppositive and non-oppositive (qtd. in Gheltofan 2015). The contributions made by the present study to paremiology, paremiography, and antonymy are as follows. Proverbially, it provides cross-linguistic evidence for Mieder's (2004) argument that there are structural patterns of proverb chunks in terms of their syntagmatic co-occurrences, including "Better X than Y" ("Better safe than sorry"), "Like X, like Y" ("Like mother, like daughter"), "No X without Y" ("No gain without pain"), "If X, then Y" ("If you can't beat them, join them"), among many more structural frameworks that are quantified, identified, and exemplified in this study. Structural isomorphism figures prominently on a large scale across languages and dialects, a claim strongly supported by Gheltofan (2015) who found Romanian and Russian structurally identical in their manipulations of paremiological antonyms. Semantically, the study adds three discourse functions to Jones's (2002) typology, (a) subordinated antonymy (cf. Hassanein 2013), (b) case-marked antonymy (cf. Hassanein 2013), and (c) replacive antonymy (cf. Davies 2012). The quantitative and qualitative analyses provide statistically and discoursally significant evidence from Arabic proverbial tradition for these three functions that are unidentified and unexemplified in non-Arabic typologies, but can be retrieved and replicated across other languages and datasets. These three classes which have been retrieved and replicated only in Arabic studies (e.g., AlHedayani 2016; Hassanein 2018) have been found to favor lexicosyntactic milieus in Arabic paremiographical discourse. These new categories prove typologies of antonymy to be genre-specific, data-driven, and hence provisional. The typology developed here is rigorous enough to profile the discourse functions of canonical antonymy across (non-)Arabic dialects. Although the textual functions of antonyms across the two dialects and datasets in comparison are categorially identical, their syntactic triggers and frequency distributions are dialectally similar in some aspects but quite distinct in others. The present comparative lexicosyntactic analysis of paremiographical antonyms across Egyptian and Saudi dialects reveals bicultural homogeneity in employing identical antonym classes for the proverbialization of cultural conceptualizations through what Sharifian (2011) classifies as cultural schemas, cultural categories, and cultural metaphors, and bicultural heterogeneity in using varied lexicosyntactic frames for the triggering of binary oppositions and the chronicling of conceptual propositions among generations. More specifically, this study shows the key role canonical antonymy plays as a linguistic mechanism for intracultural transmission and representation and a cross-cultural tendency for organizing paremiographical discourse in terms of binary oppositions.
In conclusion, the eleven classes replicated in this study account for the key role canonical antonymy plays in cross-dialectal Arabic paremiographical discourse. So a large number of semicanonical and noncanonical contexts of opposition refined and excluded in the study demonstrate that antonyms can also function in unusual, robust, and innovative ways pending exploration and generalization (Davies 2012;Jones 2002). This study has focused on the canonical antonyms and developed a dynamic and flexible classification thereof. Future research is direly needed in order to investigate the semicanonical and noncanonical profiles of antonymy or opposition in Arabic paremiography. The question whether or not the functional categories and syntactic frames quantified, qualified, and exemplified here apply to the semicanonical and noncanonical antonyms arises and is yet to be examined. This question is still falling behind and hence is posed for further study into larger stretches of discourse across other languages. More extensive research should be conducted to provide cross-linguistic evidence and safely generalize the typology developed and the conclusions reached.