The indigenization of Ghanaian Pidgin English

Received: 9 February 2023 Revised: 3 July 2023 Accepted: 8 July 2023 DOI: 10.1111/weng.12635 ORIGINAL ARTICLE The indigenization of Ghanaian Pidgin English Kofi Yakpo Linguistics, School of Humanities, The University of Hong Kong, Hong Kong SAR Abstract In the world Englishes literature, ‘indigenization’ is short- Correspondence Kofi Yakpo, School of Humanities, The University of Hong Kong, Pokfulam, Hong Kong SAR. Email: kofi@hku.hk hand for the localization of Outer Circle Englishes in former exploitation colonies like Ghana. However, the localization of Ghanaian English has been continually reversed by ‘corrective’ realignment with world standard English through Funding information Research Grants Council, University Grants Committee of the Government of Hong Kong, Grant/Award Number: GRF 17608819 institutional regimes. By contrast, the localization of Ghanaian Pidgin English has proceeded unhampered by standardization. This article provides a first analysis of the copula system of Ghanaian Pidgin English, showing that it owes much to patterns found in Akan and other languages of southern Ghana. In this domain, Ghanaian Pidgin English has indigenized and differentiated itself from its sister languages. I propose a consistent and expansive definition of indigenization as ‘the areal alignment of a latecomer with a linguistic ecology, causing its divergence from related varieties elsewhere.’ This study of indigenization shifts the focus from standardized Englishes to contact Englishes. The latter remain unfettered by institutional intervention and are therefore better suited to illustrating the natural dynamics of indigenization than standardized Englishes. 1 INTRODUCTION Over 70 indigenous languages with genetic and areal links have been spoken on the territory of modern Ghana for centuries (Eberhard et al., 2023). Ghanaian Pidgin English and Ghanaian English are, by contrast, latecomers to the multilingual Ghanaian linguistic ecology. Both languages have only been used by wider sections of the Ghanaian population since the mid 20th century. The most direct ancestor of Ghanaian Pidgin English was a Krio-influenced variety This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2023 The Authors. World Englishes published by John Wiley & Sons Ltd. World Englishes 2023;1–21. wileyonlinelibrary.com/journal/weng 1 of West African Pidgin brought to Ghana by African labor migrants from Nigeria, Liberia, and other parts of the West African coast following WWI (Kropp Dakubu, 1997, p. 157).1 Knowledge of English, in turn, was limited to a tiny indigenous elite in precolonial times (for Ghana, see Kropp Dakubu, 1997, p. 155), grew very slowly during British colonial rule, and only increased significantly after the expansion of schooling in the wake of Ghana’s independence in 1957 (Sey, 1973). However, the two languages have been characterized by diametrically opposed social dynamics and correspondingly fundamental differences in their linguistic dynamics. As an Outer Circle English, Ghanaian English belongs to the global cosmos of standardized Englishes. Like elsewhere, standardization, institutionalization, and scholarization have led to ‘structural arrest’ (Yakpo, 2020a, p. 133). The localization of English has been halting and is continually reversed by normative realignment with world standard English. By contrast, the localization of Pidgin, which remains unstandardized, has been unimpeded, rapid, and comprehensive. The local adaptation of Ghanaian Pidgin is a typical instance of linguistic indigenization. I propose a definition of linguistic indigenization as ‘the areal alignment of a latecomer with a linguistic ecology, causing its divergence from related varieties elsewhere.’ In doing so, I argue for a uniform use of the concept of indigenization. This contrasts with the selective use of the term, which is usually limited to the localization of Outer Circle Englishes and other European colonial languages in former exploitation colonies. Mufwene (2009) provides a rebuttal of the selective application of the term by showing that Inner Circle Englishes like North American English also owe their distinctness to indigenization. This study expands the term indigenization into the opposite direction. It aims to show that English-lexifier creoles and pidgins are far better examples of indigenization than Inner and Outer Circle Englishes because their localization is not constrained by standardization. The copula system is an ideal domain for investigating the indigenization of Ghanaian Pidgin. Copulas have been the subject of a number of comparative studies within the family of African Caribbean English Creoles, to which Ghanaian Pidgin belongs (Faverey et al., 1976; Holm, 1999; Michaelis & The APiCS Consortium, 2013). Huber (1999, pp. 234– 236) dedicates a section to copula expression in Ghanaian Pidgin and Yakpo (2022) discusses aspects of the system in comparison with other varieties of West African Pidgin and Krio. No previous study has, however, dealt with Pidgin copulas at a similar level of detail as the present one, nor interpreted the system in the context of the broader areal typology and language ecology of Ghana. The lingua franca function of Pidgin has only been unfolding very gradually in Ghana (Yakpo, 2020b). Until today, the main vernacular and lingua franca of Ghana is Akan, which has therefore emerged as the primary adstrate of Pidgin. This study takes this development into account by privileging Akan as a standard of comparison with Pidgin. The second most important adstrate of Pidgin is Gã, which provided an input into Pidgin earlier on but has since been outpaced by Akan. Reference to Gã will therefore also occasionally be made. This study is based on a transcribed corpus of primary field data of natural conversations in Ghanaian Pidgin totaling 60,238 words. All data was gathered by me and my research assistant Sabinus Chiravira during field research in Accra, Ghana in 2016, 2017, and 2019. The data was transcribed and analyzed through the Fieldworks Language Explorer (FLEx) software (SIL Language Technology, 2022). Field data was complemented by comparative data from consultations and grammaticality judgements with speakers of Akan and Gã via social media, video conferencing, and face-to-face conversations in 2021. The corpus was provided by 25 speakers in the age range of 15 to 36, of which onethird (eight) were women. All speakers had attended, or were attending secondary school, and about half had attended tertiary institutions. The sociolinguistic profile is therefore typical of speakers of the variety termed ‘Student Pidgin’ in the literature (see section 2). All examples stem from field data unless a source is provided. In the few cases where I rely on my speaker intuitions, I mark the examples as ‘own knowledge.’ I proceed as follows. In section 2, I discuss the notion of indigenization further. Section 3 provides a typological overview of systems of nominal and locative predication. In section 4, I conduct a point by point comparison of Ghanaian Pidgin and Akan nominal and locative predication. In section 5, I summarize the analysis and discuss the features and sources of the Pidgin copula system from the perspective of indigenization. Section 6 concludes this study. 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO 2 2 3 INDIGENIZATION AS A TYPE OF AREAL ALIGNMENT In the world Englishes literature, ‘indigenization’ characterizes the process of adaptation and localization of English varieties spoken in the Outer Circle (Hoffmann & Siebers, 2009; Kachru, 2005; Moag, 1982; Wren, 1976). The term has been used in a similar way to refer to the localization of French colonial varieties on the African continent (Mufwene, 1998; Zabus, 2018). In the linguistic-structural sense, the term describes the imposition of phonological, morphosyntactic, pragmatic, and lexical-idiomatic features of indigenous languages onto the local English variety by multilingual speakers. However, Ahulu (1994) argues against Ghanaian English as a distinct variety, noting that the language shares the same nonstandard usages with other postcolonial varieties. The term indigenization has also been criticized for being selectively applied to Outer Circle Englishes, but not to Inner Circle Englishes, nor English-lexifier pidgins and creoles (Mufwene, 2009, 2015). The latter observation cuts across the focus of this study. In West Africa, a whole group of Englishes has undergone indigenization in a far more swift and comprehensive manner than any colonial standard variety of English in the region. This group of Englishes is West African Pidgin and Ghanaian Pidgin is one of its varieties. Ghanaian English and Ghanaian Pidgin were both introduced from outside, have only been adopted by larger speaker populations from the 1960s onwards, and were until recently learned exclusively outside of the home. Both serve as lingua francas rather than vernacular home languages. Further, Ghanaian Pidgin is linked with Ghanaian English in a complex anti-register relationship (Yakpo, 2020b, pp. 70–73). The two languages are, of course, also genetically related. This is, however, where the similarities end. Ghanaian Pidgin has expanded exclusively through emergent processes, driving the language along a path of natural acquisition by speakers, as well as regional (Hampel, 2020) and lectal differentiation. It first spread to the urban Ghanaian working class, military, and police force during the first half of the 20th century (Dadzie, 1985). In a second phase, Pidgin was adopted from the late 1960s onwards by adolescents and young multilingual, predominantly male adults of the Ghanaian educated class in government boarding schools and universities (Amoako, 2011; Dako, 2002; Huber, 1999; Osei-Tutu, 2014). The latter process led to the differentiation of a distinct youth sociolect of Pidgin, a register known in the literature as ‘(Ghanaian) Student Pidgin’ (see above-cited sources). Due to its ties with the urban elites, the capital Accra, and the wealthier, more heavily populated coastal region of the country, Student Pidgin is exerting a strong influence on the older working-class variety (referred to as ‘Town Pidgin’ in the literature; see sources above). Student Pidgin has become the dominant target variety for the burgeoning youth population of Ghana’s cities and towns and is being used by a growing proportion of young women (Dako, 2013). Student Pidgin is therefore the variety of choice for this study. By contrast, Ghanaian English has almost exclusively spread through scholarization and administrative action. The role of English as an everyday lingua franca is still very limited although it has been adopted by sections of the highly educated elite for use among themselves. Pidgin is, by contrast, mostly used among peers, siblings, in small business transactions and popular music, and as a lingua franca in unregulated domains. The linguistic outcomes differ accordingly. Ghanaian English has retained an external center, has remained grammatically, lexically, and symbolically parasitic on Inner Circle Englishes. Ghanaian Pidgin, in turn, has become autonomous vis-à-vis its most direct ancestors (presumably an earlier variety of Nigerian Pidgin, and ultimately, Krio). It has been indigenized substantially by its speakers since its arrival a hundred years or so ago. In the same time span, Ghanaian English has, by contrast, remained barely distinguishable from varieties in neighboring countries and the United Kingdom except, perhaps, in some salient phonological and lexical features (Bobda, 2000); nonstandard morphosyntactic features remain highly variable (Huber, 2012, p. 390). Ghanaian English and Ghanaian Pidgin are therefore both subject to areal diffusion from other languages spoken in Ghana (Osei-Tutu, 2016), but the indigenization of Ghanaian English is halting and fractured. The language is continually thrown back into the fold of world standard English by the gravitational pull of scholarization and institutional 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO action at the national level as well as the international socioeconomic and cultural hegemony exercised by speakers of Inner Circle Englishes. Indigenization implies some degree of autonomy vis-à-vis an ancestor or input language. The point when that autonomy has been reached and the language deserves the label ‘indigenous’ is somewhat fuzzy and needs to be determined on a case-by-case basis. When ‘indigenization’ is applied consistently and expansively, the term can characterize the areal alignment of a language in the wake of colonization, migration, and acculturation. Movement of a people with their language or of a language through its acquisition by (immobile) speakers is quintessential to indigenization, for it may imply that the latecomer language to an ecology is typologically rather different from the languages already in place. The refashioning of features that accompanies the indigenization of a newly arrived language without a close genetic and areal relationship to resident languages may therefore lead to typologically noteworthy and crosslinguistically unusual outcomes (van Sluijs et al., 2016). Claims of indigenization as defined above therefore need to be buttressed by careful comparative-typological analysis. This is what I attempt here, concluding that the Pidgin copula system has indigenized considerably by aligning itself with the systems of Ghanaian adstrate languages. By indigenizing, Ghanaian Pidgin has differentiated and distanced itself from its relatives spoken in other parts of West Africa. 3 TYPOLOGICAL AND AREAL ASPECTS OF COPULA EXPRESSION Prototypical copulas are semantically vacuous elements that occur in intransitive clauses with non-verbal or less verby predicates. Crosslinguistically, such intransitive clauses may contain predicates that code events, properties, classes (nouns), and locations (Stassen, 1997). The use of copulas is characteristic in the expression of +TIME STABLE class membership (nominal predication), and by comparison with nominal predication, –TIME STABLE location-existence (locative predication) (Givón, 1979; Pustet, 2003). The distributional reach of prototypical nominal and locative copulas varies crosslinguistically. Standard Average European languages (Haspelmath, 2001) like English employ a single copula to link nominal and locative predicates to a subject but also properties, resultant states, passives, and progressives. By contrast, the Macro-Sudan languages of West Africa and beyond (Güldemann, 2008) feature split copula systems (Stassen, 2013). In Ewe (Ghana, NigerCongo, Kwa) nominal predicates co-occur with the copula nyé (1). The locative-existential verb lè is employed with locative predicates (2).1 (1) Séná NAME nyé núfíálá. COP teacher ‘Sena is a teacher.’ (Ewe; own knowledge) (2) Séná NAME lè xĬ be.at house mè. inside ‘Sena is indoors.’ (Ewe; own knowledge) In the African languages covered in this study, the locative-existential verb takes adverbial complements. Location is the conceptual basis on which location itself, existence, existential condition, and quality (location in a certain state) are predicated (Ameka, 1995, pp. 158–160). This also holds for creoles of the Atlantic basin with Macro-Sudan substrates and European superstrates other than English (e.g., Berbice Dutch Creole, see Kouwenberg, 1994: 125). APiCs feature 76 shows that split systems are also encountered in all Afro-English creole languages of West Africa and the Americas (Michaelis & The APiCS Consortium, 2013). The APiCs features do not, however, cover variation beyond the nominal-locative split. Additional variation in numerous West African languages is induced by changes in tense-aspect-mood (TAM), best captured by the variable 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO 4 5 ± FACTATIVE (Welmers, 1973, p. 348): States of affairs that are not marked for a TAM category acquire default interpretations in accordance with their lexical aspect class. Unmarked dynamic or inchoative verbs acquire a perfective sense, while stative verbs and copulas are interpreted as imperfective or current state. When a clause featuring a copula is specified for tense-aspect-mood other than +FACTATIVE, forms other than the basic nominal or locative copula are obligatory, or preferred. Hence, a speaker of Ewe uses the nominal copula nyé ‘COP’ in a clause like (1) above to express an imperfective (or present or current state) state of affairs. When the state of affairs is anterior or posterior to reference time, the lexical (serial) verb (trĭ) zù ‘become’ is preferably recruited (3). (3) Gbè day èká, èví sìà â-(trĭ) zù one child this 3SG.SBJ:POT-turn become teacher núfíálá. ‘This child may/will be(come) a teacher one day.’ (Ewe; own knowledge) Polarity is a third feature that can codetermine variation. Nominal and locative predicates typically communicate the existence and location of things and concepts in the physical and metaphorical worlds. A +AFFIRMATIVE polarity copula clause is therefore presuppositionally more natural than a −AFFIRMATIVE one (Miestamo, 2005, pp. 195–200). The expression of negative polarity therefore leads to further lexical variation. In Gã (Ghana, Niger-Congo, Kwa), a locative predication with positive polarity is formed with the locative-existential verb yĬ (4). A locative predication with negative polarity makes use of the negative locative-existential verb bĬ (5). In Gã (and Akan, see (12)), both locative-existential verbs are isomorphic with the verbs of (non-)possession; hence the alternative translations in (4) and (5). (4) Kòjó NAME shíà. yĬ be.at/have house ‘Kojo is home/has a house.’ (Gã) (5) Kòjó NAME shíà. bĬ be.at/not.have house ‘Kojo is not home/doesn’t have a house.’ (Gã) A fourth source of formal variation is linked to overlaps in form and function between nominal copulas and elements that encode focus. The isomorphy of focus markers and copulas is a widespread areal pattern in the Macro-Sudan (Stassen, 1997, p. 76). Campbell (2017, pp. 365–366) describes the Gã focus marker nì as having anaphoric discourse deixis. It identifies a referent from a discursively preconstructed (known) domain (6). Fronting of the focused constituent tsĬĭlĬ ‘teacher’ is mandatory in these structures. Clauses containing nì can only deictically identify by default third-person referents, hence the ungrammaticality of the independent pronoun in (6). (6) TsĬĭlĬ teacher nì *(lĬ). FOC 3SG.INDP ‘S/he is a teacher.’ Lit. ‘A teacher is what she is (not some other profession).’ (Gã) A final aspect of typological divergence between European and many Macro-Sudan languages concerns the predication of properties. Ewe, for example, has a rather small adjective class whose members function as adverbial complements of the locative-existential verb lè (Ameka, 2001), see (7). A somewhat larger number of property concepts is lexicalized as stative verbs (8). 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO (7) XĬ-kpĭ-kpĭ lè Berlin house-see-RED.NOM LOC PLACE m-é lè NEG 1 -3 SG . SBJ be.at soft-ADV NEG2 bĬbĬ-è ò. ‘Finding a house in Berlin is not easy.’ (Ewe; own knowledge) (8) Àsì lá m-é kúráá ò. bĬbĬ price DEF NEG1 -3SG.SBJ soft at.all NEG 2 ‘The price is not cheap at all.’ (Ewe; own knowledge) In many Macro-Sudan languages, the use of prototypical copulas is therefore largely limited to the expression of nominal predication with the values +TIME STABLE, +FACTATIVE, and positive polarity. Even forms expressing locative –TIME STABLE, +FACTATIVE are semantically rather rich. Many languages also show overlaps of copula and focus marking functions. Further, very few properties occur in structures involving copulas. In most contexts, forms are used that also fulfil other lexical and pragmatic functions. I therefore avoid the term ‘copula’ from now on unless the form in question is prototypical. 4 NOMINAL AND LOCATIVE PREDICATION IN GHANAIAN PIDGIN AND AKAN I now compare and contrast Ghanaian Pidgin and its principal adstrate Akan, and where relevant, draw attention to similarities and differences with English. I first turn to the basic ±TIME STABLE split in section 4.1, then to variation induced by changes in ±FACTATIVE tense-aspect-mood in section 4.2. In section 4.3, I cover the predication of properties. Table 1 and Table 2 summarize the variation of forms occurring with nominal (+T) and locative (–T) predicates. I employ the abbreviations +T, –T, +F, –F for the four values. Pidgin (Table 1) makes three contrasts, namely +T+F, +T−F, and –T±F. The potential fourth contrast −T+F vs. −T−F is inactive (−T±F), which is why dé ‘be at’ may be used for both values. Further, the +T−F contrast is not very entrenched in Pidgin because it applies in a rather restricted number of contexts. TA B L E 1 +T −T Ghanaian Pidgin. Akan. TA B L E 2 +F −F +F −F bì ‘COP’ bí ‘COP.NFACT’, +T yĬ ‘COP’, né ‘FOC’ yĭ ‘make’ mék ‘make’ −T wĬ ‘be at’, bá ‘come’, kĭ ‘go’, nní ‘not be at’ tè ‘stay’, gyìnà ‘stand’ dé ‘be at’ Akan (Table 2) marks all four contrasts through separate variants. Noteworthy is the existence of a –T+F form variant conditioned by –AFFIRMATIVE polarity (wĬ ‘be at’ vs. nní ‘not be at’). A further difference with Pidgin is the existence in Akan of the focus marker né ‘FOC’ (first line left) that also introduces nominal predicates with the values +T+F. Both Akan and Pidgin contrast sharply with Ghanaian English, which behaves no differently from other standardized Englishes in this respect. English only has a single copula namely be and its allomorphs. There is no formal distinction between a copula with nominal (I am a student) and locative (I am in school) complements. Further, English employs affixal inflection and suppletion for person-number, as in w-as vs. w-ere, while Pidgin and Akan forms are morphologically invariant. English TAM-conditioned suppletion is triggered by specific TAM readings, as in ±PRESENT (I am vs. I was) and is therefore functionally quite different from the ±FACTATIVE split in the African languages. Further, the English copula takes hundreds of properties as adjectival complements unlike corresponding forms in Pidgin and Akan. I compare the systems of nominal and locative predication in more detail in sections 4.1–4.3. 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO 6 4.1 7 Variation induced by ±TIME STABLE Ghanaian Pidgin has a typical split system. The nominal copula bì ‘COP’ occurs in +TIME STABLE clauses. Bì is negated ‘symmetrically’ (Miestamo, 2005), by preposing the general verbal and nominal negator nó ‘NEG’ to the nominal copula (9). (9) Mà fĭs 1SG.POSS first ném (nó) bì Thomas. name NEG COP NAME ‘My first name is (not) Thomas.’ (Ghanaian Pidgin) The −T locative-existential verb dé ‘be at’ takes adverbial complements (10). The verb is semantically rich and therefore less of a prototypical copula than bì. It may take direct complements without adpositions and may occur without a complement in existential clauses. Example (10) would therefore be well-formed without the locative complement dĬ klás rúm with the meaning ‘there are (no) people.’ The locative-existential verb is also negated symmetrically with the general negator nó. (10) Pípò (nó) dé people NEG dĬ klás be.at DEF rúm. class room ‘People are (not) in the classroom.’ (Ghanaian Pidgin) A basic ±TIME STABLE split between nominal +T+F and locative −T+F is also found in Akan, realized, respectively, by the nominal copula yĬ ‘COP’ (11) and the locative-existential verb wĬ ‘be at’ (12). The Akan locative-existential verbs are, however, polysemous and also function as the basic verbs of (non-)possession. Sentence (12) can therefore also mean ‘does your mother have a house?’ Pidgin has a separate possession verb, namely gĭt ‘obtain, have’ (< Eng. get) (see (38)). (11) ºĬ-yĬ ĬkyèrεĬkyérεĭfóĭ. 3SG.SBJ-COP teacher ‘S/he’s a teacher.’ (Akan) (12) Wó 2SG.POSS mààmé wĬ fíé? mother be.at house ‘Is your mother at home?’ (Akan) A second difference between the two languages is that Akan alone shows overlaps in the expression of nominal predication and focus. The +T+F nominal copula yĬ ‘COP’ may be replaced by the focus marker né ‘FOC’ (Amfo, 2010). The subject NP is focused in situ, hence the use of independent forms if the focused subject is pronominal (13). Né identifies a referent from a discursively preconstructed, known domain (Amfo, 2010, p. 37), hence just like nì ‘FOC’ in Gã (see (6)). (13) Mé né 1SG.INDP FOC Ĭyàrèsáfóĭ. doctor ‘I am (the one who is) a doctor’ (not somebody else in the reference group). (Akan) Ghanaian Pidgin has no separate form functioning as a focus marker cum copula. In this, it differs starkly from all its West African relatives. The latter feature nà ‘FOC,’ a focus marker with copula functions (for Cameroon Pidgin, see Mbakong Tsende, 1993: 55–56; for Pichi, see Yakpo, 2019: 233–236). Instead, Ghanaian Pidgin features a cleft focus 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO construction employing the basic +T+F copula bì ‘COP’ and an expletive pronoun (see (24) further below). Although the resulting construction looks very English-like, Akan also features cleft-constructions with the +T+F copula yĬ ‘COP’ functioning as a cleft particle (14). In fact, the construction is crosslinguistically very common (Declerck, 1988). (14) ƐĬ-n-yĭ 3SG.SBJ.INAN- NEG-COP:NEG Ĭyàrèsáfóĭ né mé. doctor 1SG.INDP COP. FOC ‘It’s not me who is a doctor.’ (Akan) The +T+F nominal copula yĬ is negated symmetrically by preposing the Akan standard negation marker -n- and suprafixing a high tone to the copula (hence yĭ ‘COP:NEG’) (14). Akan also makes use of asymmetrical negation by employing a lexical variant for the −T+F value with negative polarity. The locative-existential verb wĬ ‘be at’ has the negative counterpart nní ‘not be at’ (16). In this, Akan differs sharply from Ghanaian Pidgin, which shows no such polarity-induced variation (see (10), and (18) in section 4.2). (15) ºĬ-n-yĭ ĬkyèrĬkyérĭfóĭ. 3SG.SBJ-NEG-COP:NEG teacher ‘S/he’s not a teacher.’ (Akan) (16) Dààbí, no mé mààmé nní fíé. 1SG.POSS mother not.be.at house ‘No, my mother isn’t at home.’ (Akan) 4.2 Variation induced by ±FACTATIVE There is no generalized ±FACTATIVE distinction in Ghanaian Pidgin. However, in some specific contexts, we do find some conditioned variation with nominal predicates (see (24)-(29)). The clause featuring bì ‘COP’ in (17) and the clause in (18) featuring dé ‘be at’ are both specified for −FACTATIVE by the potential mood marker gò ‘POT.’ Both forms are freely combinable with TAM markers and the negative particle nó, contrary to equivalent forms in Akan (see (31) and (32) below). In other WAPs this is also the case for dé but not for the nominal copula (Ayafor & Green, 2017; Mazzoli, 2013; Yakpo, 2022). (17) À (nó) gò bì 1SG.SBJ NEG POT COP lĭyà. lawyer ‘One day, I’ll (not) be a lawyer.’ (Ghanaian Pidgin) (18) À (nó) gò dé Hǒ mòró. 1SG.SBJ NEG POT be.at PLACE tomorrow ‘I’ll (not) be in Ho tomorrow.’ (Ghanaian Pidgin) Ghanaian Pidgin is an aspect and mood-prominent language like Akan and the other Kwa languages of Ghana (Ameka & Kropp Dakubu, 2008) as well as all WAP and Krio varieties (Faraclas, 1987; Yakpo, 2019, pp. 158–159). Clausal time is therefore determined in relation to predicates in the discourse. This is unlike the absolute tense system of English, in which clausal time is anchored in relation to speech time. A −FACTATIVE state of affairs can therefore be specified for past or future tense by adjuncts alone, as in bày dĬ táym ‘by the time’ in (20). The clausal marker ná ‘THEN,’ borrowed from Akan, and its Pidgin calque dĬn (< Eng. ‘then’) (20) may also specify bì ‘COP’ or dé ‘be at’ as non-present. The 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO 8 9 resulting structure in Pidgin strikingly resembles the corresponding one in Akan (see (21) below), and illustrates the extent of borrowing from Akan in some areas of Pidgin grammar. (19) Ná THEN wànà háws 1PL .POSS house bì dís COP this SP SUB some friend SPEC buy=3SG.OBJ ó, wé sàm pàdí bí báy=àm. ‘This was actually our house before, then some friend bought it.’ (Ghanaian Pidgin) (20) Mék yù nó shívà, bày SBJV 2SG.SBJ NEG fear dĬn à dé THEN 1 SG . SBJ by dĬ táym yù DEF time 2SG.SBJ POT gò kách Kùmásé, catch PLACE dĭ. be.at there ‘Don’t worry, by the time you reach Kumasi, (then) I’ll be there.’ (Ghanaian Pidgin) In Akan too, the unmarked nominal copula yĬ ‘COP’ and the locative-existential verb wĬ ‘be at’ retain their default +FACTATIVE tense-aspect value of imperfective or current state in relation to event time. Given the right pragmatic context, (11) and (12) above could therefore also be interpreted as ‘she was a teacher’ and ‘was your mother at home?’ Akan speakers can nevertheless explicitly anchor stative verbs in the future or past with the clausal particle ná ‘THEN’ alone. States of affairs specified by ná are then still construed as +F. (21) ºĬkyéná, wó-bĭ-dúrù Kùmásé nó, tomorrow 2SG-FUT-arrive PLACE mè-wĬ ná DEF THEN hĭ dèdàw. 1SG.SBJ-be.at there already ‘Tomorrow, (when) you arrive in Kumasi, I’ll be there already.’ (Akan) However, overt TAM specification for −FACTATIVE induces lexical variation. The +T−F value is realized by the verb yĭ ‘make’ (22). Note that yĭ ‘make’ solely differs from yĬ ‘COP’ by a lexical high tone. (22) Àfé year bààkõĭ àkyí nó, one back DEF 1SG.SBJ-FUT-make teacher mè-bĭ-yĭ ĬkyèrεĬkyérεĭfóĭ wĬ China. LOC PLACE ‘In a year’s time, I’ll be a teacher [doing teacher] in China.’ (Akan) The use of ‘make’ with diverse complements, ranging from nominals to property words and ideophones is common in many languages of the West African littoral and very likely to be areal (more in section 4.3). Hence, there is little doubt that low-toned yĬ ‘COP’ (+T+F) and high-toned yĭ ‘make’ (+T−F) are diachronically related. In fact, the lowtoned copula yĬ appears to be a lexicalized variant of yĭ ‘make,’ tonally inflected for continuative aspect. The Akan continuative is formed by the suprafixation of a low tone, but inflection for this TAM category is defective and lexically limited to a few verbs (see Boadi, 1966). Synchronically, the two forms are differentiated by their lexical tone, have different functions, and can therefore be seen as distinct lexemes/function words. The polysemy of Akan yĬ ‘COP’ vs. yĭ ‘make’ (see (22)) has provided the model for interesting instances of ‘selective polysemy copying’ (Johanson, 2008) and semantic blending in corresponding Pidgin forms. One of these is the alternation between bì ‘COP’ and mék ‘make,’ which mirrors the alternation between Akan yĬ ‘COP’ and yĭ ‘make.’ In Pidgin, mék may express a +TIME STABLE state of affairs when its nominal complement is seen to be inhabited by a transient property, as signaled by the adjunct fĬ tĭn yíĬs in (23). Such uses instantiate the +T−F value. But in a crucial distinction to Akan, the use of mék in this way is not categorical in Pidgin, and the use of bì ‘COP’ is always possible, even when there is overt TAM marking (see (17)). 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO (23) À kápèntà mék 1SG.SBJ make fĬ tĭn yíĬ-s carpenter PREP ten fĬ Tógó. year-PL PREP PLACE ‘I was [worked as] a carpenter for ten years in Togo.’ (Ghanaian Pidgin) A second instance of selective polysemy copying from Akan is the use of the high-toned non-factative copula bí ‘COP.NFACT’ instead of the low-toned general nominal copula bì ‘COP’ in specific morphosyntactically conditioned contexts. Bí ‘COP.NFACT’ is found in −NEUTRAL focus clauses (24) and WH-clauses where the copula is stranded (25). In the corresponding Akan structures, a high tone is placed over the focused verb by a fully productive phrasal tone rule (Marfo, 2005), again leading to a contrast between a high-toned and a low-toned yĭ vs. yĬ (26). In Ghanaian Pidgin, forms other than the copula are, however, not affected by the rule. The high-toned alternant bí ‘COP.NFACT’ is limited to these specific constructions and therefore covers a functionally more restricted space. Note that example (25) is also well-formed if the non-factative nominal copula bí is substituted by the locative-existential verb dé. In the latter case, the speaker shifts the semantic focus to the transient quality of the subject. (24) Ì 3SG.SBJ bì só COP like.that that guy dát gáy bí/dé ó. COP. NFACT/be.at SP ‘That’s how that guy is (it is so that guy is).’ (Ghanaian Pidgin) (25) Háw ì fĬ bí? how 3SG.SBJ MOD COP.NFACT ‘How should it be?’ (Ghanaian Pidgin) (26) Sàá that nà Ĭ-yĭ! FOC 3SG.SBJ.INAN-COP.FOC ‘That’s how it is!’ (Akan) Beyond that, the non-factative nominal copula bí is employed in two idiomatic contexts. In (27), it occurs with the idiomatic sense ‘be really good.’ Two characteristics underline the idiomaticity of the construction. For one part, imperfective marking, as in (27), is normally incompatible with a copula in Pidgin. Secondly, the vowel of bí COP.NFACT is often lengthened for emphasis and the copula naturally co-occurs with intensifiers, in this case kĬkĬ ‘only, really,’ sourced from Gã. (27) Dá that múvì dè bí(í) kĬkĬ. movie IPFV COP. NFACT INT ‘That movie is really good.’ (Ghanaian Pidgin) The structure in (27) above again constitutes an interesting case of selective copying and semantic blending with idiosyncratic results in Pidgin. The Akan source structure in (28) involves the high-toned property verb yĭ ‘be good,’ which is a segmental and tonal homonym of yĭ ‘make.’ Both Akan forms therefore contrast tonally with low-toned yĬ ‘COP.’ But in this instance, Pidgin speakers have equated the tonal contrast between yĬ ‘COP’ and yĭ ‘be good’ with an opposition between the +F nominal copula bì ‘COP’ (see (9)) and the −F nominal copula bí ‘COP.NFACT’ (see (27)). Speakers have also innovated with respect to TAM marking since the Akan stative verb yĭ ‘be good’ remains unmarked for a current state reading, while Pidgin bí in (27) is marked for imperfective aspect. (28) Sàá that sìní yĭ páá. movie be.good INT ‘That movie is really good.’ (Akan) 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO 10 11 The second idiomatic usage of the +T−F nominal copula bí ‘COP.NFACT’ is the phrasal expression ì bí làyk sé ‘it seems that’ (29), which partly calques the use of high-toned yĭ ‘make’ in the corresponding Akan structure in (30). (29) Ì làyk sé bí shì 3SG.SBJ COP.NFACT like QUOT lĭf. 3SG.SBJ.F leave ‘It seems (it is like) that she has left.’ (Ghanaian Pidgin) (30) ƐĬ-yĭ mè sĭĬ 3SG.SBJ.INAN-make 1SG.OBJ QUOT Ĭ-à-kĭ. 3SG.SBJ-PRF-leave ‘It seems to me (it makes me) that she has left.’ (Akan) Turning to locative predication, we have seen that such states of affairs may be specified for −FACTATIVE TAM in Ghanaian Pidgin without occasioning the replacement of the locative-existential verb dé (see (18)). Conversely, the Akan locative-existential verb wĬ ‘be at’ may not be overtly specified for tense, aspect or mood. A −T−F state of affairs is expressed by the motion verbs bá ‘come’ and kĭ ‘go’ (Ellis & Boadi, 1968) when the subject is construed as willful (31). When the subject is inanimate, the expression of −T−F is achieved by way of stative verbs like tè ‘stay’ and gyìnà ‘stand’ (32). (31) Mĭ-bá Nkràn 1SG.SBJ:FUT-come PLACE Ĭkyéná. tomorrow ‘I’ll come to [‘be in’] Accra tomorrow.’ (Akan) (32) Lĭrì-hyĭn nó bé-gyìná lorry-vehicle DEF FUT-stand hĭ à-twĬn wò. there CONSEC-wait 2SG.OBJ ‘The lorry will be (standing) there waiting for you (when you arrive).’ (Akan) 4.3 The predication of properties Many basic properties (color, dimension, value) are lexicalized as stative verbs in Ghanaian Pidgin (33), just like in all Afro-European creole languages of the Atlantic basin, other varieties of WAP, and en gros, the Macro-Sudan languages (Kouwenberg, 1996). (33) Shì blák. 3SG.SBJ.F be.dark ‘She is dark-complexioned.’ (Ghanaian Pidgin) Beyond that, a large range of properties, including basic ones like ‘good’ (34), can also be introduced with the +TIME STABLE nominal copula bì ‘COP.’ My idiolect of Pidgin probably allows any English-sourced property word to appear in a predicate adjective construction like (34). The difference between a clause like (33) above, where the property is expressed by a verb, and one like (34), where it is expressed as an adjectival complement appears to be stylistic to me. Due to its typological distance from English and its proximity to constructions in Ghanaian languages (see (8) and (36)), the verbal option in (33) evokes associations of authenticity and linguistic competence (cf. Osei-Tutu, 2018). Other WAP and Krio varieties except Pichi (Yakpo, 2019, pp. 264–266) appear to be just as liberal in the use of predicate adjective constructions featuring the nominal copula bì ‘COP.’ Since Pichi has not been in contact with English for one and a half centuries, this geographic distribution suggests influence from English in Ghanaian Pidgin. Note, however, that the structural template of the predicate adjective construction is found in all languages of southern Ghana (see (37) below), even if it is lexically restricted to a smaller range of properties. 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO (34) Ì gò 3SG.SBJ POT páá. bì gúd COP good INT ‘It would be really good.’ (Ghanaian Pidgin) The −TIME STABLE locative-existential verb dé ‘be at’ is also found with temporally more transient properties (35), a feature that also characterizes other African Caribbean English Creoles (for Suriname, see Migge, 2000). (35) Tĭl=àm sé à dé bízì smĭl. tell = 3SG.OBJ QUOT 1SG.SBJ be.at busy small ‘Tell her I am a bit busy.’ (Ghanaian Pidgin) In Akan, most properties are probably lexicalized as stative verbs, as in wà ‘be long’ in (36). Akan nevertheless has quite a few high-frequency property concepts that appear in predicate adjective clauses featuring the +T+F nominal copula yĬ ‘COP.’ The group even comprises basic adjective classes of color (Ĭ-yĬ tùntúm ‘it is dark’), value (Ĭ-yĬ pápá ‘it is good’), and dimension, see (37). In spite of the structural overlap with Pidgin (and congruence of both languages with English), the liberty of the Akan nominal copula to collocate with properties is nowhere near that of Pidgin. (36) ºĬkwán road nó wà. DEF be.long ‘The road is long.’ (Akan) (37) Àbòfrá yí child this yĬ téntén páá. COP long INT ‘This child is really tall.’ (Akan) Ghanaian Pidgin also offers the coding of −FACTATIVE TAM of property predicates by way of mék ‘make,’ besides nominal predicates (see (23)). For all speakers of the Student Pidgin variety that I consulted (eight altogether), the resulting structure acquires a resultant state sense as explained by the meta-comment of speaker Derek Asante Abankwa in (38). Mr. Asante Abankwa acquired Pidgin during his adolescence in the port city of Tema as well as during studies at the Kwame Nkrumah University of Science and Technology in Kumasi, the second largest city of Ghana. (38) ‘Shì mék blák,’ dĬn 3SG.SBJ.F make dark THEN sĬm táym, dĬn some time THEN DEF dĬ chík gĭt kálà ó. girl get color SP ‘“She became dark”, (means that) after some time, the girl got color.’ (Ghanaian Pidgin) The use of ‘make’ with property words is a clear case of areal diffusion to Ghanaian Pidgin since it is not attested in any other WAP or Krio variety. In Akan, some physical properties that can be construed as resultant states can only occur as complements to yĭ ‘make,’ but not yĬ ‘COP,’ e.g. dén ‘hard’ (39). Gã employs the verb fèé ‘make, do’ to the same effect (40) and with a much wider distribution than the corresponding Akan form (Campbell, 2017, pp. 148–150). Hence congruent influence on Ghanaian Pidgin is very likely in this case. (39) Né hõĭ 3SG.POSS body yĭ dén. make hard ‘S/he is strong.’ Lit. ‘Her/his body makes hard.’ (Akan) 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO 12 13 (40) ÀmĬ-fèé klálò. 3PL .SBJ-make ready ‘They’re ready/have readied themselves.’ (Gã) Turning back to Ghanaian Pidgin, we find that some speakers choose to express a transient body state like ‘be hungry’ in an argument structure where the body state hángà ‘hunger’ is coded as an agentive subject and the emoter mí ‘1SG.OBJ’ as an experiencer object (41)(a). In the alternative argument structure that is probably more common in Student Pidgin the body state is coded as the dynamic verb hĭng ‘be hungry’ and the emoter as an experiencer subject (41)(b). Reflexes of both structures are equally common in related varieties (for Pichi, see Yakpo, 2019, p. 355). (41) (a) Hángà (dè) kách mí. hunger IPFV catch 1SG.OBJ ‘I’m (getting) hungry.’ (Ghanaian Pidgin) (b) À dè hĭng. 1SG.SBJ IPFV be.hungry ‘I’m hungry.’ (Ghanaian Pidgin) The structure in (41)(a) above mirrors that of Akan in (42) below, which features an experiencer construction with the verb dè ‘hold, seize’ (for the full range of meanings of the verb, see Christaller, 1933, pp. 68–69). Note that dè is marked for continuative aspect via a low tone. By contrast, the Pidgin equivalent kách ‘catch’ in (41)(a) may either be zero-marked for perfective aspect to express the present state (‘I’m hungry’) or by imperfective dè to express the entry-into-state (‘I’m getting hungry.’). Speakers therefore resort to the different grammatical means at the disposal of the two languages to render the (entry-into the) state of hunger. (42) ºĬkĭm hunger dè mè. hold.CONT 1SG.OBJ ‘I’m hungry.’ Lit. ‘Hunger holds me.’ (Akan) Overall, we see many parallels between Ghanaian Pidgin and Akan (and other adstrates) in form, semantics, and the idiomatics of property predication, but also quite a bit of micro-variation and innovation in the realm of Pidgin TAM marking. These aspects and others are explored further in section 5. 5 THE INDIGENIZATION OF THE COPULA SYSTEM OF GHANAIAN PIDGIN I now revisit the main points of the analysis in the preceding sections. In 5.1, I compare the distribution of the features conditioned by the values ±TIME STABLE, ±FACTATIVE, as well as the features associated with property predication. In 5.2, I classify these features according to the process through which they were likely to have been transferred to Ghanaian Pidgin. I conclude that the copula system of Pidgin has undergone a significant amount of areal alignment with its adstrate(s). As a result, Ghanaian Pidgin appears more indigenized in the realm of copula expression than related varieties in West Africa. 5.1 Feature distribution Table 3 lists the distributions of features discussed in section 4. (Non-)occurrence in the three languages is indicated by ‘yes’ or ‘no.’ Red-shaded areas show features shared between Ghanaian Pidgin and Akan, green shows features 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO TA B L E 3 Distribution of features in Ghanaian Pidgin, Akan, and English. Feature Pidgin Akan English Examples COP /‘be at’ split yes yes no (9)–(12) b ‘be at’/‘have’ isomorphism no yes no (12), (16) c COP/ FOC isomorphism no yes no (13)–(14) d COP = cleft particle yes yes yes (13), (24) 2 ±FACTATIVE a aspect prominence and relative tense yes yes no (19)–(21) b COP /‘make’ isomorphism yes yes no (22)–(23) c restricted COP variation yes no no (24)–(27), (29) d ‘be at’ variation no yes no (31)–(32) 1 ±TIME STABLE a 3 Property predication a property verbs yes yes no (33), (36) b COP occurs with almost any property yes no yes (34) c ‘be at’ occurs with properties yes no no (35) d ‘make’ occurs with properties yes yes no (38)–(40) e experiencer constructions yes yes no (41)–(42) Red: Pidgin and Akan; green: Pidgin and English; blue: only Pidgin; grey: Pidgin, Akan, and English; no color: not shared with any other language. shared by Pidgin and English, grey shows features shared by Pidgin, Akan, and English, and blue marks features unique to Pidgin. Unshaded areas are only found in a single language. Relevant examples are added in the rightmost column for reference. (1) ±TIME STABLE. Ghanaian Pidgin shares the basic time stability COP/’be at’ split with Akan (feature 1a). This is a widespread areal feature in West Africa and it is attested in all relatives of Ghanaian Pidgin in West Africa and the Americas. Typologically more interesting is, therefore, the absence of two polysemy patterns. Ghanaian Pidgin does not feature the areal polysemy pattern in which a single verb instantiates location-existence (‘be at’) and possession (‘have’), as in Akan (feature 1b). Secondly, Ghanaian Pidgin does not show a focus marker/nominal copula isomorphism either (feature 1c). The latter feature is attested in Akan and many other languages of West Africa and further afield, including all of Ghanaian Pidgin’s West African relatives. Finally, the function of COP as a cleft particle (feature 1d) is cross-linguistically so common that it occasions the only overlap between the three languages in the table. (2) ±FACTATIVE. Ghanaian Pidgin and Akan are aspect and mood-prominent languages with relative tense (feature 2a). This feature is areally widespread, however, and found in numerous Afro-European creoles as well. The existence of an isomorphism of the nominal copula (COP) and ‘make’ in Ghanaian Pidgin (feature 2b) is therefore a better indicator of areal diffusion from Akan and Gã specific to the Ghanaian linguistic ecology. Contrary to Akan, Ghanaian Pidgin, however, features a semantically and morphosyntactically restricted variation of +TIME STABLE forms induced by the presence of –FACTATIVE TAM marking, rather than a generalized one (feature 2c). A further point of divergence between Ghanaian Pidgin and Akan is the absence of variation of the locativeexistential verb induced by –FACTATIVE (feature 2d) in Pidgin. The –TIME STABLE locative-existential verb dé ‘be at’ 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO 14 15 is therefore free to combine with any TAM marker as well as the negator nó. The presence of feature 2d suggests that dé is more copularized in Pidgin than the equivalent form wĬ in Akan. (3) Property predication. Ghanaian Pidgin and Akan align with each other and contrast with English in the way basic adjectives can be expressed as stative property verbs (feature 3a). By contrast, there is a significant congruence between Pidgin, Akan, and English in the possibility to use their respective nominal copulas (bì, yĬ, be) with at least some properties, which is a widespread cross-linguistic pattern. At the same time, Ghanaian Pidgin shares the use of the nominal copula bì ‘COP’ in predicate adjective clauses with almost any property only with WAP varieties that continue to be in contact with English (feature 3b). Equally, Ghanaian Pidgin allows for the locative-existential verb dé ‘be at’ with properties construed as transient (feature 3c). By contrast, Akan speakers never make use of the equivalent wĬ ‘be at’ for the same purpose. Further, Pidgin and Akan both feature the resultant state interpretation of properties by way of ‘make’ (feature 3d), where Pidgin aligns even more closely with Gã than with Akan. Finally, both Pidgin and Akan, but not English, have experiencer constructions in which the subject instantiates bodily states like hunger and thirst and the object instantiates an animate emoter (feature 3e). Overall, we find significant analogies between Ghanaian Pidgin and Akan, some features that suggest transfer from English, and some features that may suggest congruent transfer from both Akan and English. We also find a smaller dose of unique features in Ghanaian Pidgin. I explore the sources of the features identified in this section further in 5.2. 5.2 Genetic, areal, and innovative features Table 4 classifies the features in Table 3 according to three hypothetical sources of transfer: (a) genetically transmitted to Ghanaian Pidgin from its ancestors; (b) areally diffused from African adstrates or borrowed from English; (c) innovations. Ghanaian Pidgin most likely inherited features 1a, 2a, 3a, 3c, and 3e from its WAP and Krio lineage, where they also occur. But these features have also been areally reinforced in Ghanaian Pidgin via their presence in Ghanaian adstrates. For example, although Akan does not permit the use of the locative-existential verb in predicate adjective constructions, other languages in the linguistic area do, for instance, Ewe (see (7)) and Gã (Campbell, 2017, pp. 146–148). The absence of feature 1b in Ghanaian Pidgin, on the other hand, is due to the more advanced state of TA B L E 4 Classification of Ghanaian Pidgin features according to source of transfer. Features Genetic transmission Areal diffusion and borrowing Innovation 1a, 2a, 3a, 3c, 3e Present in all WAPs and in Krio Reinforced by presence in Akan and other adstrates — 1b Absent in all WAPs and in Krio Reinforced by absence in English — 1c — Reinforced by absence in English Absence in GhaP is unique feature 1d Present in all WAPs but absent in Krio Reinforced by presence in Akan and English — 2b, 3d — Transferred from Akan — 2c Reinforced by presence of general copula variation in all WAPs and in Krio Reinforced by presence of general copula variation in Akan Presence in GhaP is unique feature 2d Absent in all WAPs and in Krio — — 3b — Transferred from English — 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO copularization of dé in Ghanaian Pidgin and its relatives vis-à-vis Akan wĬ, but probably also reinforced by generally advanced copularization in English. Feature 1c, the absence of a COP/FOC isomorphy (the focus particle cum nominal copula nà ‘FOC’ in other WAP varieties), is a unique feature in which Ghanaian Pidgin diverges from all other WAPs and Krio. Given its ubiquity in other WAPs and the presence of functionally equivalent forms in the Ghanaian adstrates, it is likely that nà did not get selected by early Ghanaian Pidgin speakers in the first place. Consequently, an innovative focus strategy involving bì/bí consolidated itself, which also partly draws on indigenous adstrate patterns, and Akan in particular. The presence of feature 1d in Ghanaian Pidgin, Akan, and English may seem trivial because of its broad crosslinguistic distribution. However, Krio varieties only use the focus particle nà ‘FOC’ not bì ‘COP’ as a cleft particle. WAP varieties in contact with English (Naijá and Cameroon Pidgin) use both, and Pichi, which is not in contact with English, uses only nà. Ghanaian Pidgin therefore probably owes feature 1d to both genetic transmission from its Naijá lineage, and reinforcement through the presence of 1d in Akan and English. Features 2b and 3d, structures involving mék ‘make,’ have no parallel in other WAP and Krio varieties. Both constitute features that clearly result from areal diffusion to Ghanaian Pidgin from Akan and other Ghanaian adstrates. Feature 2c is unique, an innovation that sets Ghanaian Pidgin apart from all other WAP and Krio varieties. The restricted variation of bì ‘COP’ vs. bí ‘COP.NFACT’ is, however, closely modelled on a small spectrum of the generalized variation found in Akan and other WAPs and therefore partly replicates genetic and areal models. By contrast, 2d is a very stable genetic feature that has not been influenced by areal diffusion from Akan. This is interesting because new forms (mék and high-toned bí) have emerged in the domain of nominal predication in Ghanaian Pidgin, and verbs like ‘go’ and ‘stand’ could have been recruited by analogy with Akan to cater for the –T–F value. Finally, 3b almost certainly results from transfer from English, since extensive property predicating functions of the nominal copula are only encountered in WAPs in contact with English. Overall, areal diffusion to Ghanaian Pidgin from adstrates appears to be more significant than that from corresponding adstrates to the copula systems of other WAPs. For example, to my knowledge Naijá does not use ‘make’ with nominal and property lexemes although speakers of Yoruba, one of its main adstrates, employ ‘make’ for these purposes (Sachnine, 1997). One possible cause is that Ghanaian Pidgin is characterized by a shallower ‘social entrenchment’ (Yakpo, 2022). Ghanaian Pidgin does not function as a vernacular (home) language, contrary to all other WAPs. Ghanaian Pidgin also shares many typological features with adstrate languages like Akan and Gã due to the historical links between African(-descended) populations on both sides of the Atlantic. For example, the predominantly isolating morphological structure of Ghanaian Pidgin and the adstrates goes along with fuzzy boundaries between lexemes and function words (Ansaldo et al., 2018), as we have witnessed in the case of copular expression. At the same time, the lexical similarity of Ghanaian Pidgin and English also provides opportunities for borrowing from the lexifier because it facilitates change of phonologically similar Ghanaian Pidgin lexemes and function words in the direction of their English etymons (Yakpo, 2017, 2023). In sum, areally diffused and contact-induced innovative features in Table 4 are a reliable diagnostic of the considerable indigenization of the copula system of Ghanaian Pidgin. Features borrowed from English that are equally common to other WAP and Krio varieties are, however, not a diagnostic of indigenization since they have not contributed to the differentiation of Ghanaian Pidgin from its West African Pidgin relatives. 6 CONCLUSION The genetic continuities remain strong between Ghanaian Pidgin on one side and other WAP varieties and Krio on the other. The copula system of Ghanaian Pidgin nevertheless owes significant parts of its morphosyntactic and semantic structure to areal patterns prevalent in Akan and other languages of southern Ghana. The Student Pidgin register has 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO 16 17 gained ground as a national youth sociolect and this circumstance is very likely to have additionally accelerated the indigenization of the language. English nevertheless remains the socially dominant superstrate and main lexifier of Ghanaian Pidgin. This also facilitates borrowing from English. Innovation is also found, but it relies on some degree of structural transfer from Akan and is markedly less important than areal diffusion from Ghanaian languages and borrowing from English. Pidgin and English are both latecomers to the Ghanaian linguistic ecology and are therefore still undergoing a process of indigenization. However, due to diametrically opposed social dynamics, the indigenization of English has been slow while that of Pidgin has been rapid and far-reaching. Language engineering by global and national elites and scholarization have led to the structural arrest of standardized Englishes, including Ghanaian English. English is therefore continually caught in a cycle of bottom-up areal alignment with Ghanaian adstrates (including Pidgin) at the level of the idiolect and small group networks, and top-down areal disalignment by ‘corrective’ standardization through social institutions (educational and literacy regimes, administrative practices, mass and social media, the formal business sector, international migration to and from Ghana, family language policies). Due to the complete absence of top-down standardization, Ghanaian Pidgin has, by contrast, evolved through bottom-up emergent processes that draw on the individual agency of speakers who interact in the course of everyday social and economic activity. I hope to have shown that these sociolinguistic dynamics have resulted in a far more comprehensive indigenization of Ghanaian Pidgin than of Ghanaian English within the same time span. In arguing for a uniform and expansive use of the term ‘indigenization,’ this study shifts the focus from standardized Englishes to contact Englishes. Ghanaian Pidgin and its West African sister languages remain relatively unfettered by administrative intervention, scholarization, and the socio-economic and cultural hegemony of Inner Circle Englishes. They are therefore much better suited to illustrating the natural dynamics of indigenization than the region’s standardized Englishes. ACKNOWLEDGEMENTS I gratefully acknowledge the support of Sabinus Chiravira and Levi Ekwa Mokake with data collection and transcription. Research for this work was supported by the Research Grants Council, University Grants Committee of the Government of Hong Kong under grant GRF 17608819 to Kofi Yakpo. DISCLOSURE STATEMENT No potential conflict of interest was reported by the author. ORCID Kofi Yakpo https://orcid.org/0000-0002-6585-8687 NOTES 1 ‘West African Pidgin English’ is an earlier name for the linguistic grouping in the literature (Agheyisi, 1971; Dwyer, 1966). Presently, ‘West African Pidgin,’ without the addition of ‘English,’ is gaining ground among linguists (Ogueji & Ahia, 2019) in keeping with practice by speakers to refer to their own and other varieties by the name of ‘Pidgin’ [Píjìn]. The exceptions are Krio (Sierra Leone) and its offshoots in The Gambia and Equatorial Guinea, which are referred to, respectively, as ‘Aku’ and ‘Pichi’ by their speakers. In keeping with these practices, I henceforth use the term (Ghanaian) Pidgin to refer to the language. 2 All Ghanaian languages including Pidgin are tone languages. A high tone is rendered by an acute accent /ó/ and a low tone by a grave accent /ò/. For the transcription of Ghanaian Pidgin, I employ an orthography based on Krio (see Coomber, 1992). The grapheme <> renders the open-mid front vowel [], <> renders the open-mid back vowel [], and <y> stands for the voiced palatal approximant [j]. Akan, Gã, and Ewe examples are transcribed following standard orthographies. 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO ABBREVIATIONS - Morpheme boundary : Separates meanings of unparsed morpheme = Clitic morpheme boundary 1 1st person 2 2nd person 3 3rd person ó High tone ò Low tone WAP West African Pidgin CONSEC Consecutive aspect COP Nominal copula DEF Definite article F Factative tense-aspect-mood FOC Focus marker FUT Future tense INAN Inanimate INDP Independent/emphatic personal pronoun INF Infinitive INT Intensifier IPFV Imperfective aspect LOC Locative preposition MOD Modal particle NEG Negative NFACT Non-factative OBJ Object case PL Plural number PLACE Place name POSS Possessive case POT Potential mood PREP General associative preposition PRF Perfect aspect PRS Present tense PST Past tense QUOT Quotative marker SBJ Subject case (Continues) 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO 18 19 SBJV Subjunctive mood SG Singular number SP (Pragmatic) sentence particle SUB Subordinator T Time stable THEN Non-present tense marker REFERENCES Agheyisi, R. N. (1971). West African Pidgin English: Simplification and simplicity. Stanford University. Ahulu, S. (1994). How Ghanaian is Ghanaian English? English Today, 10(2), 25–29. https://doi.org/10.1017/S0266 078400007471 Ameka, F. K. (1995). The linguistic construction of space in Ewe. Cognitive Linguistics, 6(2–3), 139–182. https://doi.org/10.1515/ cogl.1995.6.2-3.139 Ameka, F. K. (2001). Ideophones and the nature of the adjective word class in Ewe. In F. K. E. Voeltz & C. Kilian-Hatz (Eds.), Ideophones (pp. 25–48). John Benjamins. https://doi.org/10.1075/tsl.44.04ame Ameka, F. K., & Kropp Dakubu, M. E. (Eds.). (2008). Aspect and modality in Kwa languages. John Benjamins. Amfo, N. A. A. (2010). Noun phrase conjunction in Akan: The grammaticalization path. Pragmatics, 20(1), 27–41. https://doi. org/10.1075/prag.20.1.02amf Amoako, J. (Ed.). (2011). Ghanaian Pidgin English: Diachronic, synchronic and sociolinguistic perspectives. Nova Novinka. Ansaldo, U., Bisang, W., & Szeto, P. Y. (2018). Grammaticalization in isolating languages and the notion of complexity. In H. Narrog & B. Heine (Eds.), Grammaticalization from a typological perspective (pp. 219–234). Oxford University Press. Ayafor, M., & Green, M. J. (2017). Cameroon Pidgin English: A comprehensive grammar. John Benjamins. Boadi, L. A. (1966). The syntax of the Twi verb. London: Univ. of London. Bobda, A. S. (2000). The uniqueness of Ghanaian English pronuncation in West Africa. Studies in the Linguistic Sciences, 30(2), 185–198. Campbell, A. A. (2017). A Grammar of Ga. [Doctoral dissertation, Rice University]. https://scholarship.rice.edu/handle/1911/ 102269 Christaller, J. G. (1933). Dictionary of the Asante and Fante language called Tshi (Twi) (2nd edition, revised and enlarged by J. Schweizer). Basel: Evangelical Missionary Soc. http://www.pathawi.net/twibible/sources/ Coomber, A. (1992). The new Krio orthography and some unresolved problems. In E. D. Jones, K. I. Sandred & N. Shrimpton (Eds.), Reading and writing Krio (pp. 15–20). Acta Universitatis Upsaliensis. Dadzie, A. B. K. (1985). Pidgin in Ghana: A theoretical consideration of its origin and development. In F. Ugboajah (Ed.), Mass communication, culture and society in West Africa (pp. 113–121). Hans Zell. Dako, K. (2002). Student Pidgin (SP): The language of the educated male elite. Research Review from the Inst. of African Studies (IAS), Univ. of Ghana, 18(2), 53–62. Dako, K. (2013). Student Pidgin: A masculine code encroached on by young women. In L. L. Atanga, S. E. Ellece, L. Litosseliti & J. Sunderland (Eds.), Gender and Language in Sub-Saharan Africa: Tradition, struggle and change (pp. 217–231). John Benjamins. Declerck, R. (1988). Studies on Copular Sentences, Clefts and Pseudo-Clefts. De Gruyter Mouton. Dwyer, D. (1966). An Introduction to West African Pidgin English. African Studies Center. Eberhard, D. M., Simons, G. F., & Fennig, C. D. (Eds.). (2023). Ethnologue: Languages of the World (26th ed.). SIL International. https://www.ethnologue.com/ Ellis, J., & Boadi, L. (1968). “To be” in Twi. In J. W. M. Verhaar (Ed.), The verb “be” and its synonyms: Philosophical and grammatical studies (pp. 1–71). D. Reidel. Faraclas, N. G. (1987). Creolization and the tense-aspect-modality system of Nigerian Pidgin. Journal of African Languages and Linguistics, 9(1), 45–59. Faverey, M., Johns, B., & Wouk, F. (1976). The historical development of locative and existential copula constructions in AfroEnglish creole languages. In S. Steever, C. Walker & S. Mufwene (Eds.), Papers from the parasession on diachronic syntax. Chicago Linguistics Society. Givón, T. (1979). From discourse to syntax: Grammar as a processing strategy. In T. Givón (Ed.), Discourse and syntax (pp. 81– 112). Academic Press. Güldemann, T. (2008). The Macro-Sudan belt: Towards identifying a linguistic area in northern sub-Saharan Africa. In B. Heine & D. Nurse (Eds.), A linguistic geography of Africa (pp. 151–185). Cambridge: Cambridge University Press. 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO YAKPO Hampel, E. (2020). Regional variation in Ghanaian Student Pidgin: Use and attitudes. Sociolinguistic Studies, 14(3), 299–320. https://doi.org/10.1558/sols.38791 Haspelmath, M. (2001). The European linguistic area: Standard average European. In Language typology and language universals (pp. 1492–1510). De Gruyter. Hoffmann, T., & Siebers, L. (Eds.). (2009). World Englishes – Problems, Properties and Prospects: Selected papers from the 13th IAWE conference. John Benjamins. https://doi.org/10.1075/veaw.g40.21muf Holm, J. (1999). Copula patterns in Atlantic and Non-Atlantic Creoles. In J. R. Rickford & S. Romaine (Eds.), Creole genesis, attitudes and discourse: Studies celebrating Charlene J. Sato (pp. 97–120). John Benjamins. https://doi.org/10.1075/cll.20. 10hol Huber, M. (1999). Ghanaian Pidgin English in its West African context: A sociohistorical and structural analysis. John Benjamins. Huber, M. (2012). Ghanaian English. In B. Kortmann & K. Lunkenheimer (Eds.), The Mouton World atlas of variation in English (pp. 382–393). De Gruyter Mouton. Johanson, L. (2008). Remodeling grammar: Copying, conventionalization, grammaticalization. In P. Siemund & N. Kintana (Eds.), Language contact and contact languages (pp. 61–79). John Benjamins. Kachru, B. B. (2005). Asian Englishes: Beyond the canon. Hong Kong University Press. Kouwenberg, S. (1994). Berbice Dutch Creole. Mouton de Gruyter. Kouwenberg, S. (1996). The relation between adjectives and verbs, with special reference to Berbice Dutch Creole. In P. Christie (Ed.), Caribbean language issues old and new. Papers in honour of Professor Mervyn Alleyne on the occasion of his sixtieth birthday (pp. 27–40). University of the West Indies Press. Kropp Dakubu, M. E. (1997). Korle meets the sea: A sociolinguistic history of Accra. Oxford University Press. Marfo, C. O. (2005). Akan focus and topic constructions and the prosody-syntax interface. Journal of West African Languages, 32(1/2), 45–60. Mazzoli, M. (2013). Copulas in Nigerian Pidgin. [Doctoral dissertation, University of Padova]. Mbakong Tsende, A. (1993). Predication et marqueurs aspecto-temporels du Pidgin English camerounais. Journal of West African Languages, 23(2), 45–58. Michaelis, S. M., & The APiCS Consortium. (2013). Predicative noun phrases and predicative locative phrases. In S. M. Michaelis, P. Maurer, M. Haspelmath & M. Huber (Eds.), The atlas of pidgin and creole language structures (APiCs). Oxford University Press. https://apics-online.info/parameters/76 Miestamo, M. (2005). Standard negation: The negation of declarative verbal main clauses in a typological perspective. Walter de Gruyter. Migge, B. (2000). The origin of the syntax and the semantics of property items in the Surinamese plantation Creole. In J. McWhorter (Ed.), Language change and language contact in Pidgins and Creoles (pp. 201–231). John Benjamins. Moag, R. F. (1982). The life cycle of non-native Englishes: A case study. In B. B. Kachru (Ed.), The other tongue: English across cultures (pp. 233–253). Pergamon Press. Mufwene, S. S. (1998). Indigénisation, français en Afrique et normes: Quelques réflexions. In Une ou des normes? Insécurité linguistique et normes endigènes en Afrique francophone (pp. 49–59). Institut d’études Créoles et Francophones. https:// yorkspace.library.yorku.ca/xmlui/handle/10315/7996 Mufwene, S. S. (2009). The indigenization of English in North America. In T. Hoffmann & L. Siebers (Eds.), World Englishes – Problems, properties and prospects: Selected papers from the 13th IAWE conference (pp. 353–368). John Benjamins. https://doi. org/10.1075/veaw.g40.21muf Mufwene, S. S. (2015). Colonization, indigenization, and the differential evolution of English: Some ecological perspectives. World Englishes, 34(1), 6–21. https://doi.org/10.1111/weng.12129 Ogueji, K., & Ahia, O. (2019). PidginUNMT: Unsupervised neural machine translation from West African Pidgin to English (arXiv:1912.03444). arXiv. https://doi.org/10.48550/arXiv.1912.03444 Osei-Tutu, K. (2014). Exploring meaning in students’ pidgin (SP). [Doctoral dissertation, University of Ghana]. Osei-Tutu, K. (2016). Lexical borrowing in Ghanaian Student Pidgin (GSP): The case of Akan loanwords and loan translations. In N. Faraclas, R. Severing, C. Weijer, E. Echteld, W. Rutgers & R. Dupey (Eds.), Celebrating multiple identities: Opting out of neocolonial monolinguism, monoculturalism and mono-identification in the Greater Caribbean (pp. 47–54). University of Curaçao and Fundashon pa Planifikashon di Idioma. Osei-Tutu, K. (2018). I get maf wey you get mf: Pronunciation and Identity in Ghanaian student pidgin. American Language Journal, 2(3), 8–25. Pustet, R. (2003). Copulas: Universals in the categorization of the lexicon. Oxford University Press. http://gen.lib.rus.ec/book/ index.php?md5=C0D66EF4014F8C23C6648247588EA42B Sachnine, M. (1997). L’expression de la qualité en yoruba. Linguistique Africaine, 261–282. Sey, K. A. (1973). Ghanaian English: An exploratory survey. Macmillan. SIL Language Technology. (2022). Fieldworks language explorer (FLEx) (9.0). Stassen, L. (1997). Intransitive predication. Clarendon Press. 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 20 21 Stassen, L. (2013). Nominal and locational predication. In M. S. Dryer & M. Haspelmath (Eds.), The World Atlas of language structures online. Max Planck Institute for Evolutionary Anthropology. https://wals.info/chapter/119 van Sluijs, R., van den Berg, M., & Muysken, P. (2016). Exploring genealogical blends: The Surinamese Creole cluster and the Virgin Island Dutch Creole cluster. Lingua, 178, 84–103. https://doi.org/10.1016/j.lingua.2015.12.004 Welmers, W. E. (1973). African language structures. University of California Press. Wren, R. M. (1976). The indigenization of English: Rhetoric in modern Nigerian literature. Black Orpheus: A Journal of African and Afro-American Literature, 3(4), 44–56. Yakpo, K. (2017). Unity in diversity: The homogeneity of the substrate and the grammar of space in the African and Caribbean English-lexifier Creoles. In C. Cutler, Z. Vrzic & P. Angermeyer (Eds.), Language contact in Africa and the African diaspora in the Americas. In honor of John V. Singler (pp. 225–250). John Benjamins. https://doi.org/10.1075/cll.53.10yak Yakpo, K. (2019). A grammar of Pichi. Language Science Press. https://doi.org/10.5281/zenodo.2546450 Yakpo, K. (2020a). Social factors. In E. Adamou & Y. Matras (Eds.), The Routledge handbook of language contact (pp. 129–146). Routledge. https://doi.org/10.4324/9781351109154-10 Yakpo, K. (2020b). Sociolinguistic characteristics of the English-lexifier contact languages of West Africa. In N. Smith, T. Veenstra, & E. O. Aboh (Eds.), Advances in contact linguistics: In honour of Pieter Muysken (pp. 61–84). John Benjamins. https://doi.org/10.1075/coll.57.02yak Yakpo, K. (2022). Social entrenchment influences the amount of areal borrowing in contact languages. International Journal of Bilingualism, 26(2), 140–162. https://doi.org/10.1177/13670069211019126 Yakpo, K. (2023). Two types of language contact involving English Creoles: Why Krio (Sierra Leone) has evolved more towards English than its relative Pichi (Equatorial Guinea) towards Spanish. English Today, 39(1), 12–23. https://doi.org/10.1017/ S0266078421000146 Zabus, C. J. (2018). Le palimpseste africain: Indigénisation de la langue dans le roman ouest-africain europhone. Éditions Karthala. How to cite this article: Yakpo, K. (2023). The indigenization of Ghanaian Pidgin English. World Englishes, 1–21. https://doi.org/10.1111/weng.12635 1467971x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/weng.12635 by Cochrane Germany, Wiley Online Library on [09/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License YAKPO