LEGON JOURNAL OF THE HUMANITIES

Volume 30.2 (2019)

COLLEGE OF HUMANITIES UNIVERSITY OF GHANA

ISSN: 0855-1502 E-ISSN: 2458-746X LEGON JOURNAL OF THE HUMANITIES

VOLUME 30.2 (2019)

Editor Gordon S.K. Adika

Assistant Editor Gladys Nyarko Ansah COLLEGE OF HUMANITIES UNIVERSITY OF GHANA

ISSN: 0855-1502 E-ISSN: 2458-746X

CREATIVE COMMONS LICENCE Attribution-Non commercial-No derivates 4.0 International (CC BY-NC-ND 4.0)

This issue was produced with financial support from the Carnegie Corporation of New York through the University of Ghana Building A New Generation of Academics in Africa (BANGA- Africa) Project.

Legon Journal of the Humanities is indexed in the database of Open Access Journals (DOAJ). It is also accessible on the platform of African Journals Online (AJOL)

Page ii Legon Journal of the Humanities Vol. 30.2 (2019) Legon Journal of the Humanities is published by the College of Humanities, University of Ghana

EDITORIAL BOARD

Nana Aba A. Amfo ………………………………………………...... ………Chair Professor of Linguistics Department of Linguistics School of Languages, College of Humanities University of Ghana, Legon, Ghana [email protected]

Gordon S. K. Adika ………………………………………………...... ………Editor Associate Professor, Language Centre College of Humanities University of Ghana, Legon, Ghana [email protected]

Gladys Nyarko Ansah …………………………………...... ……Assistant Editor Senior Lecturer, Department of English School of Languages, College of Humanities University of Ghana, Legon, Ghana [email protected]

Esi Sutherland-Addy ………………………………………………...... ……Member Associate Professor Institute of African Studies, College of Humanities University of Ghana, Legon, Ghana [email protected]

Susanne Gehrmann …………………………………………………...... …...Member Professor of African Literatures and Cultures Department of Asian and African Studies Humboldt University, Berlin, Germany [email protected]

Sabelo J. Ndlovu-Gatsheni ……………………………………………...... Member Professor, Department of Development Studies Professor and Head, Archie Mafeje Research Institute (AMRI) University of South Africa, Pretoria, South Africa [email protected]

Legon Journal of the Humanities Vol. 30.2 (2019) Page iii Cas Wepener ………………………………………………………...... ……Member Professor of Religions and Head Department of Practical Theology University of Pretoria, South Africa [email protected]

Joshua Amuah ……………………………………………...... ……………..Member Senior Lecturer and Head Department of Music School of Performing Arts, College of Humanities [email protected]

EDITORIAL ASSISTANT Kofi Asante Twumasi

PAST EDITORS : 1974-1977: Lawrence A. Boadi 1987-1994: John N.D. Dodoo 1994-1999: Alex K. Dzameshie 1999-2003 : E. Kweku Osam 2004-2009 : Gordon S. K. Adika 2009-2015 : Helen A. Yitah 2016-2018 : Augustine H. Asaah

Page iv Legon Journal of the Humanities Vol. 30.2 (2019) ADVISORY BOARD Moradewun Adejunmobi, Professor, African American and African Studies, University of California, Davis, USA. Emmanuel K. Akyeampong, Professor of History; Professor of African and American Studies, Center for Government and International Studies, Harvard University, Cambridge-MA, USA. Li Anshan, Professor of International Relations; Director, Institute of Afro-Asian Studies and Center for African Studies, Peking University; Vice-President, Chinese African Studies, China. Hyun-Chin Lim, Professor Emeritus of Sociology, Founding Director of Asia Center, Seoul National University; President, Korean Social Science Research Council, Korea. Dorothy Odartey-Wellington, Associate Professor of Spanish, School of Languages and Literatures, University of Guelph, Canada. David Owusu-Ansah, Professor of History; Executive Director, Faculty Access and Inclusion, James Madison University, Harrisonburg-VA, USA. Nicola Piper, Professor of International Migration, Department of Sociology and Social Policy, Director of Human Rights and Democratization (Asia Pacific), The University of Sidney; Convener, Sidney Asia Pacific Migration Network (SAPMIN), Australia. Laud Ato Quayson, Professor and Director, Center for Transnational and Diaspora Studies, University of Toronto, Canada. João José Reis, Professor, Department of History, Federal University of Bahia, Salvador, Brazil. Maria Koptjevskaja Tamm, Professor of General Linguistics, Department of Linguistics, Stockholm University, Sweden.

Legon Journal of the Humanities Vol. 30.2 (2019) Page v Email: [email protected]; [email protected]

For further details (e.g. guide for contributors, copyright, etc.) kindly go to the journal’s website: http://coh.ug.edu.gh/ljh

Page vi Legon Journal of the Humanities Vol. 30.2 (2019) Table of Contents

Religious pluralism and interfaith coexistence: Ecumenicalism in the context of traditional modes of tolerance...... 1 David Owusu-Ansah Emmanuel Akyeampong Ethno archaeological clues to stone exploitation in Ancient Dangme-land, Greater- Accra Region...... 19 Fritz Biveridge Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana 43 Kwabena Sarfo Sarfo-Kantankah Joseph Arko Pub interaction in South Western Nigeria 71 Ganiu Abisoye Bamgbose Temitope Michael Ajayi A reconsideration of grammatical categorization in English...... 87 John F. Wiredu The syntax of Dagbani personal pronouns: an analysis...... 109 Kwame Abukari Definiteness in the Zarma Determiner phrase...... 141 Waheed Ayisa Jayeola Documenting Simpa: Advances in language documentation...... 167 Nana Ama Agyeman Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana...... 191 Bernard Afiik Akanpabadai Akanbang Gordon Dugle Millicent Awialie Akaateba Humanities and sciences as complementary aspects of an Afrikan=Black whole: Evidence from Archeoastronomy...... 215 Obádélé Bakari Kambon Yaw Mankatah Asare Book Review: Review of Philosophical Foundations of the African Humanities through Postcolonial Perspectives edited by Helen Yitah and Helen Lauer...... 243 Dannabang Kuwabong

Legon Journal of the Humanities Vol. 30.2 (2019) Page vii

Owusu-Ansah, D. & Akyeampong, E./Legon Journal of the Humanities Vol. 30.2 (2019) DOI: https://dx.doi.org/10.4314/ljh.v30i2.1

Religious Pluralism and Interfaith Coexistence: Ecumenicalism in the Context of Traditional Modes of Tolerance1 David Owusu-Ansah Professor of History & Associate Provost for Diversity James Madison University, United States of America E-mail: [email protected] Emmanuel Akyeampong Gurney Professor of History and African and African American Studies Oppenheimer Faculty Director Harvard University, United States of America Email: [email protected]

Submitted: January 3, 2019/ Accepted: September 9, 2019/ Published: December 30, 2019 Abstract In many parts of the non-western world, religion is singled out as the cause for violent clashes. At the 2007 TrustAfrica workshop in Dakar, the conference of religious leaders, scholars, and experts from 12 African countries and the Diaspora explored this concern under the theme "Meeting the Challenges of Religion and Pluralism in Africa." It was observed that religiously justified conflicts were often the repackaging of community concerns regarding issues of social, economic, and political injustices, inequities and exclusions. Consequently, a project on “religious pluralism and interfaith coexistence in Ghana” was funded in 2008 as part of the efforts to examine the role of local traditions as a foundation to interfaith dialogue. Earlier in 2005, a similar project on the theme of Islam and tolerance, with funding from Harvard and Michigan State University, was conducted in Ghana.2 Several issues emerging from the field conversations are pertinent to topics of how history shaped Muslim relations with their non-Muslim hosts in West Africa. It is often argued in the literature that Islam’s inherently adoptive attitude toward African religious culture made it possible for Muslims and their non-Muslim hosts to co-exist. However, this research contends that, in the case of Ghana, it was the traditional local culture as defined by indigenous religious values that shaped and moderated the environment that sustained peaceful interreligious relations. The authors express concern that as the country experiences rapid urbanization, Westernization, proliferation of charismatic churches and aggressive Christian evangelization, the traditional values that underpinned pluralism and peace in historic times might be threatened (George Bob-Milliar and Karen Lauterback, 2018). Keywords: Traditional cosmology, spiritual knowledge, power, ecumenism, the common good, religious tolerance, pluralism, syncretism, peaceful national coexistence

1 David Owusu-Ansah presented a version of this paper at the 12 Annual-Lecture in Honor of Professor Nehemia Levtzion held at the Levtzion Center for Islamic Studies, Hebrew University, Jerusalem, on 20 March 2017. 2 Interviews used as source material for this paper came from our fieldwork in Ghana in 2005 through 2008. Rebecca Tandoh and our colleague the late Dr. Mark Sey were members of the research team.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 1 Owusu-Ansah, D. & Akyeampong, E./Legon Journal of the Humanities Vol. 30.2 (2019) Introduction and Review of Literature In the age of Al-Qaeda in the Maghreb, Boko Haram and al-Shabaab terror operations on the continent of Africa, it is important that the literature on religious co-existence and the peaceful Muslim and non-Muslim interactions in the history of West Africa be revisited. In his book titled Muslims and Chiefs in West Africa (1968), Nehemia Levtzion focused not so much on the spread of Muslim traders in the region, but rather on the activities of Muslim clerics whose ventures into the non-Islamic territories of the Volta Basin started the process of Islamization of the area and beyond. Composed of the Dagomba, Gonja, Gurma, and Nanumba now in territories of modern Ghana, and the Mossi, Gurma and Grunshi in modern Burkina Faso, Levtzion, similar to scholars such as Lamin Sanneh (1976) and Ivor Wilks (2002), examined the history of clerical services that Muslims performed for non-Muslim rulers. Over time, local lords such as those in Dagomba and Gonja came to “accept” Islam but the majority of their rural communities remained traditionalists. From the writings on the Islamization of North Africa and Egypt to the scholarship that examined the role of Muslim clerics in traveling trading caravans across the Sahara, and to the works on conversion and the Islamization of sub- Saharan West Africa, urban centers were presented as characterizing the seats of traditional rulers and therefore of commerce. Such locations favored traders and clerics who spread the influence of Islam. The rural regions where the majority of the non-Muslim locals resided were places where puritan Muslims retreated when they became dissatisfied with the quality of urban Islam, and in some cases, these were sites from where jihads were planned and staged. This bifurcation is evident in the volume co-edited by Levtzion and Humphrey Fisher under the title of Rural and Urban Islam in West Africa (1987). While it is important not to portray traditional society as unchanging, it is important to note that African villages remained predominantly traditional and retained the purity of the local cosmology to which were attributed all the spiritual agencies that determined the factors of production.3 There was indeed the belief that the effective combination of hard work, fertility of land and even the availability of capital and labor were not sufficient guarantors of productive outcomes. In the Akan forest environment, which includes territories in modern Ghana and parts of the Cote d’Ivoire, balance was sought between the sacred and profane

3 An important observation needs to be made here that we are not portraying traditional society as unchanging. We have seen works that have examined how the bori cult in Nigeria, for example, responded to social change, including Islam and colonial rule. See for example, Adeline Masquelier, “Narratives of power, images of wealth: the ritual economy of Bori in the market,” in Modernity and its Malcontents: ritual and power in postcolonial Africa, edited by Jean Comaroff and John Comaroff, University of Chicago press, 1993.

Page 2 Legon Journal of the Humanities Vol. 30.2 (2019) Owusu-Ansah, D. & Akyeampong, E/Religious Pluralism and Interfaith Coexistence: Ecumenicalism in the Context of Traditional Modes of Tolerance existence and the knowledge of those who were able to tap into spiritual power was of great importance for the collective good of the community. It was crucial to maintain relations with the spirit-world because of the various benefits inherent in such association. The forest shaped Akan cosmology and conceptions of power, as Thomas McCaskie has insightfully demonstrated in several important works (McCaskie, 1983 and 1992; Akyeampong and Obeng, 1995). While much emphasis in the literature had focused on the impact of urban Muslims in converting West African communities, it is the position of this paper that it was rather the strength of the rural and indigenous cosmology that often-shaped inter-religious relations (Akyeampong and Owusu-Ansah, 2014; Owusu-Ansah, 1991) – for the color of water is really a reflection of the soil over which it flows. Scholars such as Peel and Horton (Horton, 1971; Horton and Peel, 1976) saw this African rural character to be something with which Islam was familiar. Cosmologically, the presence in Islam of a powerful God whose blessings could be tapped through the services of clerics was attractive and one of the compelling reasons for the conversion of local rulers. However, as Levtzion often demonstrated in his writings, conversion was often a rational decision and therefore the Muslim presence in West Africa, even when it led to conversion, did not necessarily lead to the abandonment of the indigenous practices. Ivor Wilks on his part attributed the legacy of tolerance and accommodation – the gradual Islamization and the retention of vestiges of indigenous practices that were characteristic of Islam in the Volta Basin – to the teachings of al-Hajj Salim Suware (Wilks, 2002). The 15th century Mande scholar advocated that true conversion occurred in God’s own time and not by the force of the sword – an accommodative situation that often led to the mixing of Islamic practices with the indigenous religious culture (Bravmann, 1983). Syncretism, if embraced in a pluralist culture, will presuppose that citizens, who are committed to different and not necessarily agreeable religious doctrines, will arrive at an accommodation for their common good. However, as Charles Stewart observed in his essay titled “Syncretism and its synonyms” (Steward, 1999), the term has a history of being used in pejorative ways to connote the absence of purity, or to describe the concept of cultural mixing and hybridization. Thus, for Islamic purists such as Askia Muhammad of Songhay, or Uthman dan Fodio of Hausaland, syncretism was a condition that was unacceptable and needed correction via jihad – a process that Claude Meillassaoux observed to have been linked also to razzia or slave raiding in the Western Sudan (Meillassoux, 1993). Given the history of jihad as the effective mode of Islamization, it is difficult to accept the argument that the teachings of Salim Suware were the only sources by which cases of peaceful

Legon Journal of the Humanities Vol. 30.2 (2019) Page 3 Owusu-Ansah, D. & Akyeampong, E./Legon Journal of the Humanities Vol. 30.2 (2019) propagation of Islam in West Africa might be attributed (Hanretta, 2005).4 In other words, research on peaceful religious coexistence should not always be assessed on the ability of the hegemonic religious cultures of Islam and Christianity to make accommodative adjustments. Indeed, it is prudent to ask whether there were values within the indigenous religious environment that also embraced pluralism and contributed to peaceful coexistence. The Color of Water: Contextualizing Ghana’s Cultural

Tradition of Religious Tolerance In two ethnographic projects (2005 and 2008) that Owusu-Ansah and Akyeampong were a part of, the research explored the values of ecumenism in the context of the Ghanaian traditional culture. The first project, “Diversity and Tolerance in the Islam of West Africa,” was a United States Department of Education funded research for Harvard and Michigan State University in 2005.5 The second was the TrustAfrica-sponsored research that focused on interviewing leaders and attendants of indigenous shrines. By this second project, the topics for the 2005 research conversations were expanded to cover issues on tolerance and religious pluralism in Ghana. The conversation topics covered a larger spectrum of beliefs. Muslims, Christians, and Indigenous Practitioners have a record of peacefully coexisting in the country. From these field inquiries, it was concluded that, indeed, in the case of Ghana, it was rather the indigenous cosmology that laid the foundation for religious ecumenism. The earliest Muslim travelers like Ibn Battuta from North Africa or the “central lands of Islam” to visit Muslim societies in West and East Africa, whose very travels had been facilitated by the Muslim leaders and traders in these societies, were critical of practices they considered un- Islamic (Bovill, 1995). Christian missionaries in 19th and early 20th century Gold Coast (e.g., Basel missionaries), wedded to the belief in Acts 4:12 that there is no other name on earth that can save but the name of Jesus, encouraged early Gold Coast converts to physically move away from their villages and towns to reside in Christian Salems or residential quarters, where they could practice their new faith in purity. Clearly, the inspiration for religious pluralism in the Gold Coast or Ghana did not originate in these two universal faiths. In comments on religious pluralism in Senegal, the country’s former leader President Leopold Senghor was purported to have estimated that Senegalese were 4 In a wonderful review of the Nehemia Levtzion and Randel Pouwels’ edited book, The History of Islam in Africa, Ohio (2000), Sean Hanretta cautioned scholars to be mindful of attributing all cases of peaceful co-existence and the absence of jihad to the influence of Suwarian traditions. Hanretta called on researchers to avoid the “pitfall in the literature whereby every failure to wage jihad is explained by references to the influence of Suwari, whether that influence is really attested or not.” 5 For detailed information about the project and related interviews in the context of “Diversity and Tolerance in the Islam of West Africa,” go to Michigan State University Digital Online Library: www.aodl.org.

Page 4 Legon Journal of the Humanities Vol. 30.2 (2019) Owusu-Ansah, D. & Akyeampong, E/Religious Pluralism and Interfaith Coexistence: Ecumenicalism in the Context of Traditional Modes of Tolerance 90% Muslims, 10% Christians, and 100% animists.6 A 2010 research at the Pew Foundation Forum on Islam and Christianity in Sub-Saharan Africa, lent credence to the Senghor claim of local cultural contents. In fact, the research observed that when respondents in Sub-Saharan Africa were asked to describe their religious beliefs and practices, a paradox emerged. “The survey findings suggest that many Africans are deeply committed to Islam or Christianity and yet continued to practice elements of traditional African religions” (Pew Research Center, 2010). It is from such observations that the strict interpretation of census information represented in graphs, even in the Pew research, could be misleading at first glance.7

In the last national census of Ghana in 2010, Christians from all denominations were recorded at the 60th percentile, 16 percent Muslims, and the remaining Indigenous and “others”. A more recent estimate divides the country’s population into about 70 percent Christians, almost 18 percent Muslims and the rest as Indigenous and others (Ghana Statistical Service, 2011; See also, worldpopulationreview.com/ countries/ghana-population). Yet, the local soil in which Islam and Christianity were sown contribute enormously to the accommodative interfaith relations in 6 For a more detailed conversation on Christian-Muslim relations in Senegal and the thoughts of President Senghor therein, see Souleymane Bachir Diagne, “Religion and the Public Space in Senegal: The Evolution of a Project of Modernity,” presented at the Institute for the Study of Islamic Thought in Africa (ISITA), Working Paper Series, Northwestern University, March 2009 7 Website for the Pew Forum graph is at: https://www.pewforum.org/2010/04/15/executive-summary-islam-and-christiani- ty-in-sub-saharan-africa/.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 5 Owusu-Ansah, D. & Akyeampong, E./Legon Journal of the Humanities Vol. 30.2 (2019) this country where traditional leadership (the stool in the south, the skin in the north) continues to be understood as an embodiment of religious and political power. Power is perceived as multi-sited, as distinct from authority that was politically sanctioned. Traditional leaders explored and engaged multiple sources of religious power and sought to harness these for the interest of polity and chief (Akyeampong and Owusu-Ansah, 2014). Therein lies the rationale for the evolution of an institution like the office of the Nsumankwaa or traditional royal physicians in the Akan chiefdom. The continuing saliency of chieftaincy in colonial and independent Ghana has ensured the survival of indigenous religions and traditional values, including that of religious coexistence (Amponsah, 2018). In addition, as the first head of state of an independent Ghana, Kwame Nkrumah, sought to create a Ghanaian nation-state, where the three major religious traditions were incorporated to make familiar and acceptable at state functions the performance of prayer through libation, by a Muslim Imam, and by a Christian minister. Based on the preponderance of evidence, we concluded our research with the hypothesis that the greater the level of syncretic pluralism, the higher the possibility of religious tolerance. We present here three separate ethnographic accounts from the field as examples of indigenous ecumenism that continues to define religious tolerance in the modern state of Ghana. Field Interviews and Observations

Summary of Interviews with Archbishop Dr. Peter Kwasi Sarpong8 In several publications spanning almost six decades, Catholic Archbishop of Kumasi Dr. Peter Kwasi Sarpong observed that the African traditional religious environment contributes greatly to peaceful ecumenical relations that allow for religious unity (Sarpong, 1974, Osei-Bonsu, n.d.). While he admitted that the Ghanaian environment has not been free of conflict, he attributed the occasional clashes to those he referred to as “small fundamentalist groups who insist on the exact interpretation of the words of either the Bible or the Quran.” He further observed that “indigenous religion has not been a source of religious conflict in Ghana because it does not have a ‘Book’ that says categorically this or that.” Rather, he pointed to traditional cultural practices, which are always imbued with religious value. In addition, shared family and community activities such as the performance of rite of passages have public spaces and built-in ecumenism and therefore venues for peaceful interactions. Such performances do link to spiritual agents that are believed to contribute to the common good. 8 Emmanuel Akyeampong conducted these interviews with Archbishop Peter Akwasi Sarpong (dd. 15 June 2005, Santaase, Kumasi). For the full interview, see aodl.org/islamictolerance/asantehistory//object/3C-18C-10/.

Page 6 Legon Journal of the Humanities Vol. 30.2 (2019) Owusu-Ansah, D. & Akyeampong, E/Religious Pluralism and Interfaith Coexistence: Ecumenicalism in the Context of Traditional Modes of Tolerance In other words, religiosity was defined not in terms of affiliation toany specific Christian denomination, Islamic sect, or any particular local shrine. Rather, religiosity was presented as relating to the sense of believing in and associating with spiritual agents that are understood to be capable of contributing to a peaceful and prosperous existence. Religion therefore serves the practical needs of a people and therefore it is not surprising to find members of the traditional extended family who share the common compound but still affiliate to different religious faiths. The Archbishop illustrated this with a story of his own childhood experiences where his family kept a shrine but also switched from one religious’ faith to another – including association with Muslims: Personally, I am an Asante… and my father’s village is adjacent to my mother’s village. If you go to my village and somebody did not tell you that this building is in Maase [village in the Offinso District] and the next one is Mpenne, you will not know where the boundary begins… Interestingly, in my mother’s house we had Muslims who lived there and in my father’s, [also] there were Muslims. Every day, at my father’s house, the Muslims were teaching the Muslim children the Quran, and suddenly I was even reciting the Quran [as a child]. So, there was such a good [relationship] and that was my experience. Even now when the chief Imam celebrates the Ramadan [fast] and they kill a cow or a sheep, they present me with some and whenever I go on pastoral visits to villages, I have the great satisfaction when Muslims come to meet me, not just to greet me, but they attend my mass and after mass they will come to receive the blessing from me… You talked or mentioned the Nsumankwaahene: Whether at a formal occasion or informal occasion, when the Nsumankwaahene sees me, whether he is engaged in formal activities or not, this man will bow to me [in recognition]. He calls me “Nyame Komfo” [the fetish priest of the Christian God]. Many of the things I say in my books are not taken from [other peoples’] books. I am a typical village boy… I [started] schooling by accident and I lived the life of nature and the true African boy at a village about three miles from Maase. [In addition], in my household, I had five fathers, which means that my biological father had four brothers [and I call all of them father]… The interesting thing in the setup of my household was that my father’s eldest brother (Kwadwo Gyinaye) was a [traditional] fetish priest, then the other brother (Kwaku Agyin) was a Catholic, then came my father (Kwabena Marfo, popularly known as Kwabena Donkor…) who was born during the Yaa Asantewaa War in 1900 became a Catholic three months following my birth. His younger brother (Amo) was a Catholic and the last brother (Kwaku Badu) was not anything or could have been anything. Later on, the first brother Kwadwo Gyinaye the traditionalist became a Catholic, and Amo, the good catholic, became

Legon Journal of the Humanities Vol. 30.2 (2019) Page 7 Owusu-Ansah, D. & Akyeampong, E./Legon Journal of the Humanities Vol. 30.2 (2019) a traditional priest. And I can tell you that he became a zealot Catholic who used to call himself “Adikan abedi akyire” [the first who has become the last] and he called himself “Paul”. So, as I was saying, many of the things I have written about in my books (if I am not exaggerating) they have come from the experiences of my childhood. So, here I was in the middle of religious tolerance (Akyeampong, aodl. org/islamictolerance/asantehistory//object/3C-18C-10/). What Archbishop Sarpong talked about as domestic ecumenism and pluralism in Ghanaian traditional society does not mean a lack of a definite knowledge of a monotheistic God. His central thesis was the argument that the European missionaries, who propagated the gospel, did not appreciate the culture within which they worked. He therefore called on the Church in its modern setting to develop an African theology by which Christian values would be explained appropriately to the people. Had that been done, he argued, the common religious values inherent in the local culture would have been tapped to strengthen the Christian message in Africa. A Day at the Antoa Anyaman River: Injunctions, Adjudication, and Reconciliation: Field Observations from Antoa Shrine9 A shrine is the most traditional religious object and site in Ghana. In the precolonial era, every important household in the village setting had a shrine similar to that commented upon in the narrative by Archbishop Sarpong. Additionally, and across the country, there were popular and powerful shrines that people from all walks of life visited for spiritual consultation – for divination and to seek amulets and charms defensively for protection and offensively to attack enemies. Since the shrines counseled living together in harmony, it was imperative that the traditional communities adhered to the moral codes of conduct associated with any given shrine with which they were attached. The day spent at the Antoa Anyaman River Shrine provided the best example of how deities adjudicated conflicts and intolerance. From the research and visit to Antoa, a village not far from Kumasi, we came to know that the shrine has an attending priest but the river “belongs” to the local chief and therefore the elders of the village’s royal family serve as custodians and on the village council to the shrine. Thus, all cases brought before the shrine are first introduced to these elders, who adjudicate cases in the name of chief and deity. The elders discuss the cases that are brought before them, ask a lot of questions of the consulting clients, provide advice and punishment, and recommend fines, but it is the priest of the shrine who performs all the rites associated with ritual cleansing.

9 Field observations by David Owusu-Ansah and Emmanuel Akyeampong (dd. 12 July 2008). For the full interview and observations from Antoa, see 2008 Interview with Antoahene Nana Kwame Owusu-Agyeman I at aodl.org/islamictolerance/as- antehistory//object/3C-18C-17/.

Page 8 Legon Journal of the Humanities Vol. 30.2 (2019) Owusu-Ansah, D. & Akyeampong, E/Religious Pluralism and Interfaith Coexistence: Ecumenicalism in the Context of Traditional Modes of Tolerance The process of arbitrating conflict at the shrine is similar to those that are conducted at any traditional home or village in Ghana. Household or village elders constitute as council to address disagreement, adjudge wrongdoers and assign fines, but alternatively, a party that feels aggrieved may invoke the name ofa deity to declare absolute innocence. Such an act, as understood in Akan culture, immediately brings the deity into the adjudication process (Hagan, 1968).10 The consequence of guilt is costly and in the olden days might have even resulted in the punishment of death. This was the case because the invocation of the deities in the declaration of injunction was always predicated on the challenge that the full force of the deity must be used to punish those who make unfounded accusation or false declarations of innocence. The deities are moral arbiters – they detest lies, cheating, dishonesty, and hold the position that there could be no reconciliation and forgiveness without truth telling and the full admission of guilt by offending parties. Thus, it is believed that, where the name of a deity is invoked in a conflict, ultimately the guilty party will be inflicted with sickness or misfortune as sign of guilt. This can only be removed with a visit to the deity where the appropriate rituals are performed following the peaceful adjudication of the original conflict to the satisfaction of all parties. During the day of our visit, we noted that visitors to the shrine included persons from all religious backgrounds. The cases brought before the elders of the shrine for adjudication were very similar to those that would have been brought before a court of law. These included cases about marital issues and disagreements, ordinary theft, land litigations regarding disputes over ownership and boundaries, conflicts over rent payments between renters and property owners, unfounded accusations for which the accused was seeking redress and damages, and even efforts to redress unfulfilled promises that lovers made in the state of passion. What was impressive was the rapidity with which people agreed to tell the truth. The guilty parties, prior to being taken to the riverside for cleansing, were individually paraded through the streets of the village in public disgrace. This whole process provided opportunity for moral education and conflict reconciliation. Thus, even though many Ghanaians identified themselves as Christians in the national census, their fears and modes of resolving conflict could very well be determined by their traditional cultural values and anxieties. The public perception of the tendency of indigenous deities to mete out instant justice encourages confession by even the uninitiated when guilty to avert the wrath of the deity. These values in the culture do reflect a history of tolerance that is based on the idea that a sinful society is a community at risk of self-destruction. 10 The ultimate act of calling on the sacred (ntam bo) in the declaration of one’s innocence in Akan society is the swearing on the Asante sacred Golden Stool.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 9 Owusu-Ansah, D. & Akyeampong, E./Legon Journal of the Humanities Vol. 30.2 (2019) Conversation with Baffour Amisare II, Chief and Custodian of Tano Shrine at Tano Oboase11 “Abosom” or deities are attached to every important stool in Akan societies; this point was repeatedly affirmed in our conversations with traditional elders regarding the role of shrines in the history of Ghanaian culture. Among the Akan of Ghana, the stool symbolizes political authority because it is believed to encase the ancestral spirits of previous rulers (Sarpong, 1971; Quarcoo, 1990). It provides legitimacy to political rule. Thus, the history of deities and their importance to society are intrinsically linked to that of prominent traditional families or settlements. Both deity and the traditional state prefer tranquility. The role of deities as protectors of the traditional state and as agencies for the assurance of stability and peace are therefore a part of the national history on tolerance. The narratives about relationships between deities and states are preserved in the context of various local oral traditions. Examples of such accounts relate to migration stories, the discovery of better settlement sites, as explanations of religious and/or political changes in the affairs of the community, in interpretations for the outcomes of past military encounters in the earlier state formation period, and/or as explanations in customary precedence.12 Of the several interviews conducted at traditional palaces and shrines across the country, the conversation with significance on the subject of religious ecumenicism was held on 15 July 2008 with Tano Oboase chief Nana Baffour Amisare II. Present at this palace interview were the following traditional officials: Nana Adomako Akyeampong (Nifahene),13 Tano Komfo Oppong Kyekyeku (traditional priest), Okyeame Nana Yaw Boadi (Linguist), and Tano Oboase Tour Guide Mr. Osei Tano Brempong. It is essential to report that during this period of interviews, Dr. Emmanuel Akyeampong and our late colleague Dr. Yaw Bredwa-Mensah were also engaged in research at Bono Manso, a site known for its links to the powerful Bono state. Later in the week, we joined Dr. Bredwa-Mensah to visit the historic village of Tano Oboase and toured the ruins of the historic Tano site from where several Akan groups trace their ancestral migrations. In the conversation with Nana Amisare II and his elders, we learned about how Tano Oboase became part of the Asante nation, how the royal house of Tano 11 Interview by David Owusu-Ansah (dd. 15 July 2008). To access the full interview with Tano Oboasehene Nana Baffour Amisare II and his elders, go to aodl.org/islamictolerance/asantehistory//object/3C-18C-18/. 12 Several such interviews were recorded in both the 2005 and 2008 field interviews referred to earlier in the essay. Specifically, see Emmanuel Akyeampong and Rebecca Tandoh (dd. 23 April 2006), Interview with Opanyin Kwabena Buor, Head of the Boa- hen Amantuo family. The narrative told the story of the role of the Mampong deity, Mpra, in facilitating an historic migration and in determining the selection of the site for settlement. Subsequently, Mpra contributed to establishing Mampong as a prominent Asante state. See, kora.matrix.msu.edu/files/60/396/3C-18C-F-106-14_Opanin Kwabena Buor Mampong.pdf. 13 Nifa is the Akan word for “right” as distinct from benkum or left. The Nifahene in the traditional Akan chieftaincy structure was the right wing/flank command of militia that protected the king when Akan states or Oman went to war. In his Asante in the nineteenth century, historian Ivor Wilks referred to these ahenfo or chiefs as captains who participate in the critical decision making processes of the Akan state.

Page 10 Legon Journal of the Humanities Vol. 30.2 (2019) Owusu-Ansah, D. & Akyeampong, E/Religious Pluralism and Interfaith Coexistence: Ecumenicalism in the Context of Traditional Modes of Tolerance Oboase became responsible for attending to the shrine. We also learned that the river-deity is called “Taakora”14 – for the fact that the spirit of the river does not condone evil and untruthfulness. In a chronological fashion, it was stated that it was rather the spirit of the river-deity that possessed one Afua Akoma, who was the ancestral grandmother of the Tano royal family, indicating how nature-deities predated the founding of settlements. Since she could not effectively attend to the deity during her menstrual cycle, she transferred those priestly duties to her male counterpart, the chief. Thus, technically, the Tano Oboase chief is a traditional priest who “lifts the shrine” when he is in the company of the King of Asante (Asantehene). On any other occasion, a fulltime Tano priest (bosomfo) performs the day-to-day responsibilities of attending to the shrine. In addition to describing his duties as chief and attendant of the shrine, Nana Amissare II also identified himself as a Christian and a member of the local Catholic Church. However, he did not see his roles as chief and custodian of the shrine to be in conflict with his membership of the Catholic Church and his faith as a Christian. Asked, if on the other hand the Church has any concern that, a member of its congregation is not only a traditional chief but also an officiant of an indigenous shrine, Chief Baffour Amisare II responded as follows: Look, I go to Church and I pay my membership dues. The Church says “thou shall have no god besides the Lord” and we know that the deities believe in God and we believe in this one God (authors’ emphasis). I have chosen on my own accord, however, not to attend the communion but in my heart, I do not see any conflict or disagreement between serving the shrine and worshiping God at the Church because the deity derives its power from the same God. So, I am not concerned about what humans will tell me regarding a possible conflict between these two ways of serving God. I go to Church and nobody chases me out from worshiping there. Moreover, on the occasions when I perform our duties regarding the shrine, they (the people from the community) come to join us. Besides, no Pastor of the Church has ever assumed the leadership of the Church in this community without coming first to greet me before he takes his post (Owusu-Ansah, Field interviews, 2008). Even though the chief admitted that such clerical visits could be viewed as only a courteous protocol extended to him as head of the village community than an acceptance of his primary role as custodian of the shrine, the chief also rationalized that were there not a Tano River Shrine at the settlement, there would not have been any village for a Church to be established and therefore a priest to be assigned. Indirectly, therefore, those clerical visits were also in recognition of the river (water) that made community life possible, and the shrine that reflected the relationship between the river deity and the community. 14 “Taakora” is composed of the words, “Tano” (the name of the River) and “akora or akokora” (the Old man or that which is ancient). It is an important attribute that elderly persons adhere to truth for the good of the society.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 11 Owusu-Ansah, D. & Akyeampong, E./Legon Journal of the Humanities Vol. 30.2 (2019) These arguments are less about any specific debate about who recognizes whom. Rather, they relate to religious tolerance – a subject that is often treated and defined by scholars of religious conversion in such terms as syncretism and eclecticism. The most appropriate topic of discussion in this case of religious encounters is embedded in the phraseology of the “weight of tradition” – a subject that Akyeampong evaluated in the portrait of Asantehene Agyemang Prempeh I, c. 1888-1931, as a Christian and a modern man, but also as a traditional ruler. Akyeampong’s essay recounts the story of an Asante king, who is transformed into a “modern” man while in British exile on the Island of Seychelles. The king learned to read and write the English language; he accepted the teachings of the Anglican Christian denomination; and yet, he held on to his traditional responsibilities as king (both in exile and upon his return). He attended to the traditional stools of the ancestors; abided by the sanctions of the most sacred object of the state – the Golden Stool; he practiced polygamous marriage upon his return from exile and was unwilling to abolish the royal harem, and yet remained Christian. Akyeampong’s portrait of this modern Christian king is about a person who did not abandon traditional ideas, principles, or ideals totally. To ask whether this king was indeed a Christian would be equal to questioning whether the more than 60 percent of the Ghanaian population that responded in the 2010 census to be Christians had ceased belief in and practices of elements of traditional religion. What is clear, however, is the high degree of religious tolerance that Prempeh demonstrated through his affirmation of traditional obligations while he attended Church and managed matters of the traditions of the Asante nation (Akyeampong, 1999). The successful Asante northern military expeditions of the mid-18th century, which led to the incorporation of territories in the Volta Basin, also brought Asante into contact with Mande Muslim communities whose history of commercial travels to the fringes of the Akan forest dated back to the heydays of the Trans- Saharan trade. Consequently, Asante officials took advantage of Muslim services that included their knowledge in the making of prayers and Islamic talisman. Even today, a particular group of Muslims still visits the palace of the king and makes prayers regularly. All Asante rulers since 1903 see themselves as Anglicans, and, yet, see no conflict whatsoever in continuing the tradition of having Muslims and traditional priests of the Nsumankwaa perform prayers and rites at the palace (Akyeampong, Interviews with Nsumankwaahene aodl.org/islamictolerance/ asantehistory//object/3C-18C-B). In fact, the Asante kingdom at the height of its power in the early 19th century either controlled or held influence over territories larger than the modern nation of Ghana. Thus, it can be argued that the manner, by which traditional authorities in the precolonial era related to Muslim merchants-

Page 12 Legon Journal of the Humanities Vol. 30.2 (2019) Owusu-Ansah, D. & Akyeampong, E/Religious Pluralism and Interfaith Coexistence: Ecumenicalism in the Context of Traditional Modes of Tolerance clerics, as well as to European traders and Christian missionaries, has contributed in a large measure to determining the politics of inter-religious relations in the country today. Taking Advantage of Religious Tolerance: Councils of Religious Bodies and the National Peace Council – Bringing the Past to Bear on the Present Similar to Archbishop Peter Akwasi Sarpong, whose interview was cited earlier, the foundational role of the indigenous religious culture in ensuring interfaith tolerance in Ghana was equally not lost on a leading Pentecostal pastor, Dr. Mensa Otabil of the International Central Gospel Church. In a 12 July 2006 interview, Dr. Otabil identified the indigenous religion culture as “the greatest buffer between Islam and Christianity” (See aodl.org/islamictolerance/asantehistory//objects/3C- 18C-11/). He expressed concern however that “as elements of the traditional religious culture thin out, [there could be] more confrontations between the world religions in this country.” The observation was similar to that of Archbishop Sarpong, where the values of flexibility and accommodation were ascribed to the local culture rather than to the world religions. In the same conversation, Dr. Otabil made two important observations. First, that conflict among religious groups often arises when fringe movements emerge to challenge the main faith. Second, that in an environment where the influence of the existing religions are more or less in a state of equilibrium, peaceful relations are best preserved and the destabilizing effects of fringe movements are best controlled if an umbrella forum under which disagreements are redressed exist. Thus, in 1984, a draft constitution for such a religious organization was written and promulgated for the creation of such an umbrella group in Ghana – the Forum of Religious Bodies. Reverend Dr. Nathan Samwini, a former director of Christian-Muslim relations at the Ghana Christian Council, noted that the idea of creating the umbrella Forum of Religious Bodies emerged from conversations that dated back to 1978. The primary concern then was how best to bring Muslim organizations in the country into dialogue with Christians (aodl.org/islamictolerance/seculareducation// object/47–1B7-4E/). The ability to construct such an organization was predicated on the fact that other religious councils existed in the country and had performed credibly well. The logical next step was the creation of a forum, at which members from the various religious organizations met from time to time for the purpose of redressing religious concerns amicably and articulating views on matters of national interest with a common voice. Member-organizations included in the religious forum were the Ahmadiyya Mission of Ghana, Federation of Muslim Councils of

Legon Journal of the Humanities Vol. 30.2 (2019) Page 13 Owusu-Ansah, D. & Akyeampong, E./Legon Journal of the Humanities Vol. 30.2 (2019) Ghana, Ghana Christian Council, Catholic Bishops Conference, and the Ghana Pentecostal Council. Prior to the formation of the Ghana Religious Forum, the most vocal and credible non-governmental religious groups that engaged governments on issue of social justice, and/or to redress conflicts resulting from partisan politics were the Christian Council of Ghana and the Conference of Catholic Bishops. Equally critical to maintaining national peace are traditional chiefs whose political and cultural authority are rooted in the veneration of local ancestors and are legitimized by association with shrines. While the institution of chieftaincy and therefore chiefs is recognized politically as important players in redressing conflicts, the acknowledgement of priests of the indigenous shrines as representatives in national ecumenical peace-building organizations has been lacking until recently. We are here referencing the 2011 Parliamentary Act 818 that called for the establishment of National, Regional and District Peace Council Boards that were charged to “facilitate and develop mechanisms for conflict prevention, management, resolution and to build sustainable peace” in the country (Republic of Ghana, “National Peace Council Act, 2011 (Act) 818),” Accra, 2011). Gazetted on 20th May 2011 and its members limited to four-year terms, the composition of the Peace Councils was intentionally inclusive. They comprise representations of the Catholic Bishops Conference, the Christian Council, National Council for Christian and Charismatic Churches, Ahmadiyya Muslim Mission, Ahl-Sunnah Muslim Group, Tijaniyya Muslim Group, Practitioners of African Traditional Religion, A member of the National House of Chiefs, two nominees of the Office of the President (one of whom must be a woman), and two persons who represent “identifiable” groups such as institutions of higher learning, civil society organizations involved in conflict resolution and peace building. The idea of establishing a group that considered the country’s plurality of religious cultures is significant. An important advancement is the representation of a practitioner of the traditional religion and a representative of the Ghana Council of Chiefs – an implicit recognition of the legitimacy of Ghana’s indigenous inclusive values. Concluding Remarks The activities of religious organizations are now an integral part of the history of the country. Similar to chieftaincy, institutions of Church and Mosque assume definite roles and articulate legitimate voices on issues relevant to national peace. These are the examples of the activities of local networks of organizations that TrustAfrica highlighted in its 2006/07 Annual Report. Researchers and scholars were called upon to stress the contributions of these local agencies to sustaining peace and reconciliation. The inclusion of practitioners of indigenous religions as

Page 14 Legon Journal of the Humanities Vol. 30.2 (2019) Owusu-Ansah, D. & Akyeampong, E/Religious Pluralism and Interfaith Coexistence: Ecumenicalism in the Context of Traditional Modes of Tolerance members of the national agency that mediates peace and reconciliation in Ghana, even at the time when national census figures reported the majority of Ghanaians as identifying as members of the world religions, is a significant recognition that religious traditions that were hitherto described negatively as syncretic or viewed as backward by the educated, could be foundational to sustaining pluralistic values that are essential for inclusive democracy. Indeed, it is essential for national governments to continue to recognize the importance of such networks of organizations. Equally important and insightful was Archbishop Sarpong’s observation that the Christian missionaries who propagated the gospel in Ghana in the early decades of European rule missed an important opportunity to appreciate the inclusive pluralistic values of the traditional religion and culture that could have made it more successful in erecting the Church on a local foundation (Cosmas Ebo Sarbah, 2014). In fact, scholars and researchers should continue to explore, elaborate, and articulate the relevance of these religious traditions and experiences that underpin ecumenism and religious tolerance in Ghana. For, it is only under such peaceful conditions that the State, society, and the several religious communities can redress perceived discrimination relative to secular education and urban youth unemployment.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 15 Owusu-Ansah, D. & Akyeampong, E./Legon Journal of the Humanities Vol. 30.2 (2019) References Abamfo, O. A. (2013). Religion and the Inculturation of Human Rights in Ghana. Bloomsbury. Akyeampong, E. (1999). Christianity, Modernity and the Weight of Tradition in the Life of ‘Asantehene’ Agyeman Prempeh I, c. 1888-1931. Africa: Journal of the International African Institute, 69 (2), 279-311. Akyeampong, E. Interview with Baffour Domfe Gyeabuor III, Asantehene’s Nsumankwaahene, aodl.org/islamictolerance/asantehistory//object/3C- 18C-B/. Akyeampong, Emmanuel and David Owusu-Ansah. (2008). List of Field Interviews at http://aodl.org/islamictolerance/asantehistory/. Akyeampong, E. and Owusu-Ansah, D. (2014). Wealth in Knowledge: Spiritual Services and Political Power in Ghana. Paper presented for the African Studies Association Conference Panel on Mystical Agency as a Problem of Knowledge. Indianapolis, 22-23 November. Amponsah, D. (2018). Kwame Nkrumah and the politics of religion in early postcolonial Ghana. Paper presented at the Ghana Studies Association panel Registers of Belief, Creativity and Power in Ghana at the African Studies Association annual conference in Atlanta, GA. 28 November-1 December. Bob-Milliar, G. and Lauterback, K. (2018). A spiritual topography and history of Kwadaso, Kumasi. Paper presented as part the Ghana Studies Association panel Registers of Belief, Creativity and Power in Ghana, African Studies Association Conference, in Atlanta, GA. 28 November-1 December. Diagne, S. B. (March 2009). Religion and the Public Space in Senegal: The Evolution of a Project of Modernity. Presented at the Institute for the Study of Islamic Thought in Africa (ISITA), Working Paper Series. Northwestern University, Evanston, IL. Dumbe, Y. (2011). Islamic Revivalism in Contemporary Ghana. Sodertorn Studies on Religion, 5. Stockholm, Sweden. Fisher, H. (1973). Conversion Reconsidered: Some Historical Aspects of religious Conversion in Black Africa. Africa: Journal of the International African Institute, 43 (1), 27-40. Ghana Statistical Service. (2011). Population and Housing Census: Provisional Results, Accra, Ghana Statistical Service.

Page 16 Legon Journal of the Humanities Vol. 30.2 (2019) Owusu-Ansah, D. & Akyeampong, E/Religious Pluralism and Interfaith Coexistence: Ecumenicalism in the Context of Traditional Modes of Tolerance Hagan, P. H. (1968). The Golden Stool and the Oath of the King of Ashanti. Research Review (University of Ghana), 4 (3), 1-33. Hoexter, M., Eisenstadt, S. and Levtzion, N. (2002). The Public Sphere in Muslim Societies. State University of New York Press. Horton, R. (1971). African Conversion. Africa, 41 (2), 85-108. ___. (1975). On the Rationality of Conversion. Africa. 45, 373-399. Horton, R. and Peel, J. A. (1976). Conversion and Confusion: A Rejoinder on Christianity in Eastern Nigeria. Canadian Journal of African Studies. 10, 481-498. Iddrisu, A. (2013). Contesting Islam in Africa: Homegrown Wahhabism and Muslim Identity in Northern Ghana, 1920-2010. Carolina Academic Press. Kobo, O. (2012). Unveiling Modernity in 20th Century West African Islamic Reforms. Brill. Levtzion, N. (1968). Muslims and Chiefs in West Africa. Oxford. Levtzion, N. (Ed.) (1979). Conversion to Islam. Holmes & Meier. Levtzion, N. and Fisher, H. (Eds). (1987). Rural and Urban Islam in West Africa. L. Rienner Publishers. Masquelier, A. (1993). Narratives of power, images of wealth: the ritual economy of Bori in the market. In J. Comaroff and J. Comaroff (Eds.), Modernity and its Malcontents: ritual and power in postcolonial Africa. University of Chicago Press. Meillassoux, C. (1993). Women in slavery in the Western Sudan. In J.M. Burns and E.K. Ching (Eds.) Problems in African History: the precolonial centuries. New York, Markus Wiener. Meyer, B. (2012). Christianity in Africa: From African Independent to Pentecostal- Charismatic Churches. In The Wiley Blackwell Companion to African Religions, (pp. 153-170). Wiley-Blackwell. Owusu-Ansah, D. (1991). Islamic Talismanic Traditions in Nineteenth Century Asante. Edwin Mellen Press. ____. (2015). ‘Off to Northern Ghana in the Morning’: Ramifications of a Trip, A review essay in honor of Ivor Wilks. Ghana Studies, 18, 177-182. ____. (2005). Fieldnotes: Interview with Rev. Dr. Nathan Samwini (Ghana Christian

Legon Journal of the Humanities Vol. 30.2 (2019) Page 17 Owusu-Ansah, D. & Akyeampong, E./Legon Journal of the Humanities Vol. 30.2 (2019) Council Coordinator of Christian-Muslim Relations), dd. 18 July, Accra. See http:// aodl.org/islamictolerance/seculareducation//object/47-1B7-4E/. Owusu-Ansah, D. and Akyeampong, E. (2008). Fieldnotes: Interviews at Antoa with Elders of the Antoa Shrine, dd. 12 July. Antoa, http://aodl.org/ islamictolerance/asantehistory/. Owusu-Ansah, D., Iddrisu, A. and Sey, M. (2013). Islamic Learning, the State and the Challenges of Education in Ghana. Africa World Press. Pew Research Center. (2010). Tolerance and Tension: Islam and Christianity in Sub- Saharan Africa. Pew Forum on Religion and Pubic Life, www.pewforum. org. April. Sanneh, L. (1976). The Origins of Clericalism in West African Islam. Journal of African History, 17 (1), 49-72. Sarbah, E. C. (2014). Notes from the Grassroots: Paradigm shift in theological formation towards interreligious dialogue in Ghana. Studies in Interreligious Dialogue, 24 (2), 225-236. Sarpong, P. A. (1974). Ecumenical Relations in Ghana. Kampala: Uganda. ____. (2002). People Differ: An Approach to Inculturation in Evangelisation. Accra: Sub-Saharan Press. Stewart, C. (1999). Syncretism and its Synonyms: Reflections on Cultural Mixing. Diacritics, 29 (3), 40-62. Weiss, H. (Ed.) (2002). Social Welfare in Muslim Societies in Africa. Nordiska Africkainstitutet. Wilks, I. (1989). Asante in the Nineteenth Century: The Structure and Evolution of a Political Order. Cambridge University Press. ____. (2011). Al-Hajj Salim Suware and the Suwarians: A Search for Sources. Transactions of the Historical Society of Ghana, 13, 1-80.

Page 18 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) DOI: https://dx.doi.org/10.4314/ljh.v30i2.2

Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana Fritz Biveridge Senior Lecturer Department of Archaeology and Heritage Studies University of Ghana, Legon. Email: [email protected]

Submitted: March 19, 2019 / Accepted: July 2, 2019/ Published: December 30, 2019

Abstract Dangme-land is naturally replete with abundant stone resources whose exploitation facilitated the peoples’ adaptation to their environment. The objective of the researcher in this paper was to investigate for what purposes the ancient indigenous Dangme populations that occupied Sega Hill, Kpone, Shai and Prampram exploited and used stones in the remote past. It also documented the techniques and methodologies utilized by the people to exploit this resource. Data for the study were derived from ethnographic and archaeological investigations and from early European sources relating to the area. The combined evidence revealed that the people of the aforementioned settlements exploited stone for a variety of purposes, such as for the construction of agricultural terraces and protective enclosures against slavers. The ethno archaeological evidence also indicated that the four principal early vocations of the people namely; fishing, crop farming, hunting and trapping were facilitated greatly by the use of a variety of tools manufactured from stone.

Keywords: Ethno archaeology, stone blocks, cultural materials, Dangme, European sources Introduction This research investigated how the indigenous populations of Sega Hill, Kpone, Prampram and Shai (Fig. 1) exploited and used stones in the remote past. Sega Hill, Kpone and Prampram are all coastal Dangme settlements located along the Gulf of Guinea in the Greater Accra Region of Ghana. Shai is also occupied by the Dangme but is located approximately 55 km. north of Accra, the capital city of Ghana. Since the early 1970’s, several comprehensive archaeological and ethnographic investigations have been conducted by some members of faculty of the Department of Archaeology and Heritage Studies (DAHS), University of Ghana, at some prehistoric Dangme settlements such as Hioweye, Sega Hill, Ladoku, Adwuku and Wodoku (Anquandah, 1982, 1985, 1996; Biveridge, 2005). However, none of

Legon Journal of the Humanities Vol. 30.2 (2019) Page 19 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) these studies investigated how the ancient indigenous populations exploited and used stones. The situation has been exacerbated by the fact that ethno historical narratives of the people on the subject are patchy and obscure. It was in the light of this situation that the author initiated this research to investigate and document how these populations exploited this natural resource for their benefit. The study had two overarching objectives. The first was to establish and document how indigenous Dangme populations occupying the above-named research areas exploited and used the abundant stone resources in their environment. The second was to identify the tool kits, methodologies and techniques employed for their exploitation. Archaeological and ethnographic data constituted the primary data sources used to undertake the study. The former was derived from extensive surface surveys and excavations undertaken at the ancient settlement quarters of the above-named settlements except Shai where no excavation was conducted because the author was unable to secure an excavation permit from the Ghana Museums and Monuments Board (GMMB), the state institution responsible for issuing excavation permits in Ghana. Two units each were excavated at Kpone and Sega Hill. At Prampram, one unit was excavated because of financial constraints. The purpose of the excavation was to recover stone objects of archaeological significance. The archaeological research was undertaken in phases, each spanning approximately two weeks and covered one town. Ethnographic data were derived from 37 respondents, mainly clan elders and selected resource persons settled in the research areas. The author also examined historical sources documented by early European writers some of whom visited the region and interacted with the local people. These documents date to between the mid-17th to late 20th century.

Page 20 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana

Figure 1: The research areas (Source: Author) The Dangme can be divided broadly into two groups namely; the inland and the coastal Dangme’s. The former live along the foothills of the Akwapim-Togo Range and their main towns are Yilo Krobo, Manya Krobo, Shai and Osudoku. Shai was the ancestral home of several Dangme groups prior to their forceful transplantation onto the Eastern Coastal Plain by English colonizers in 1892. Crop farming, animal husbandry, potting and trading constitute their main vocations. The latter are settled along the Gulf of Guinea and comprise Prampram, Kpone, Ningo and Ada. Deep sea fishing, trading and crop farming are their primary economic activities. Ethno historical traditions of the Dangme indicate they migrated from external sources before settling at Shai and the Lorlovo Plains in Osudoku (Anquandah, 1996, p. 75). Some places mentioned in Dangme ethno historical traditions as their ancestral homeland include Togo, Dahomey and Yoruba-land (Anquandah, 1982).

Legon Journal of the Humanities Vol. 30.2 (2019) Page 21 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) Historical and linguistic evidence however suggest that they migrated from the Lower Volta Basin in two groups – Se and Le (Boahen, 1977; Anquandah, 1982; Kropp-Dakubu, 1976). Socially, the Dangme are organized into clans based on patrilineal descent, subdivided into localized patrilineages. Some notable socio- cultural events of the Dangme include the Dipo puberty rites and celebration of the following festivals: Homowo (Kpone and Prampram), Ngmayem (Krobo) and Asafotufiami (Ada). All of the above-named Dangme groups speak different dialects of the Dangme language, a Kwa branch of the Niger-Congo family of languages. Early European records (Barbot, 1732; Bosman, 1705; Meredith, 1812) posit that the Dangme played a primal role in the socio-political and commercial affairs of the Gold Coast (now Ghana) during the Atlantic contact and pre-colonial eras. Kpone (Ponnie), Ningo (Lay) and Prampram (Pompena) for example, were important commercial entrepots from where important traditional export commodities notably ivory, salt, gold, and slaves were exchanged and exported enroute to Europe. Novel and exquisite European trade merchandises like glass beads, ceramics, alcoholic and non-beverages, textiles, guns and gunpowder were also channeled through them to inland forest polities between the sixteenth and late nineteenth centuries. A few derelict canons, the architectural ruins of Fort Venon (Prampram), Fort Fredensborg (Ningo) and the foundations of an ancient Danish trade post at Kpone are some vestiges attesting to early Euro-Dangme interaction in the study areas. Stone as used in the text refers to all hardy concretions of earthy or mineral matter such as granite, quartz and gneiss. Dangme-land is naturally replete with stony boulders and outcrops composed mainly of granite and gneiss. Other less occurring mineral types in the study areas are quartz, hornblende, mica and quartzite. Rocks of the Dahomeyan System underlie the research area. According to Oboli (1998), the geology consists mainly of massive crystalline granulites and gneisses which have undergone extensive metamorphism due to great pressure. Other major rock types unique to the system comprise hornblende, pyroxenes, calcium-scapolite and garnet. Migmatites with subordinate quartz, schist, biotite, schist and other sedimentary rock types constitute the remnants (Kesse, 1985). Much of the coastline consists of flat sandy patches interspersed with large rocky promontories which become visible for several metres off-shore at low tide. All the coastal settlements named above in the research area are sandwiched between large saline lagoons. From west to east, they comprise Laloi, Gao, Moyo and Sangaw Lagoons. The banks of the lagoons are characterized by dense mangrove swamps which are home to several species of freshwater molluscs and crustaceans. The research area experiences the Dry Equatorial Climate type which

Page 22 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana is characterized by high temperatures ranging from 35° C. in March to about 25° C. in August (Dickson & Benneh, 1973). The area also experiences double maxima rainfall annually and the rainfall pattern is generally sporadic and unreliable, averaging about 800 mm. annually (Dickson & Benneh, 1973). The relatively impervious underlying rocks make rainwater penetration into the ground difficult; it thus runs off as broad sheet flood during the wet season. Vegetation cover differs significantly as one moves inland from the coast. For example, along the banks of the lagoons can be found several thick patches of mangroves and waterlogged swamps but further inland, the vegetation is dominated by open grasslands, scrambling and climbing thickets and shrubs occasionally interspersed with tree savanna. The main tree types are Ceiba pentandra, Antiaris Africana, Albizia zygia, Sterculia tragacantha and Baphia nitida. Relevance of Stone Artifacts for Archaeological Reconstruction Stone has a usage history of over two million years. Its hardy and relatively imperishable nature made it the preferred material for making the first undoubted humanly made tools like choppers, flakes, arrowheads, burins (for hunting); hand axes (for farming) and warfare weapons (arrowheads). Others included stone armbands, anklets and beads (for body decoration/ornamentation) and grinders, querns (for food processing). These devices have been discovered at the lowest occupation levels at some of the earliest known settled villages worldwide. Stones are also known to have been used more extensively than any other material, such as for building fortifications and buttressing house foundations. Their small sizes ensured they remained preserved and almost intact in the event of destructive natural and man-made disasters like bushfires, earthquakes, tsunamis and wars. Their study can thus yield veritable clues about past subsistence practices, how ancient sites were used, socio-cultural patterns and their implications for the users. Flint, quartzite, granite and quartz constituted the preferred rock types used by early man to make tools because of their extra hardy nature which minimize disintegration and breakage upon impact with other hardy materials like bone. Flint can be dated chronometrically by the Thermoluminsence (TL) dating method. Thus, by typological analysis and extrapolation, other artifacts recovered from the same stratum can be relatively dated and the synthesis of human needs can be recognized in the archaeological record. The earliest stone tool types included choppers and pebbles. They were formed using Oldowan technology which involved knocking one rock particle against another to remove unwanted flakes from the main rock body to have the preferred working edge as well as achieve the desired shape and size. According to

Legon Journal of the Humanities Vol. 30.2 (2019) Page 23 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) Joukowsky (2015) and Nicholas and Schick (2010), pebbles and choppers constituted the principal tool types used by the genus Zinjanthropus and Australopithecus to butcher game during the Early Stone Age (Lower Paleolithic) which extends to about three million Y.A. Zinjanthropus and Australopithecus were hunter-gatherers and their remains have been found along parts of the East Africa Rift Valley at sites like Hadder, East Turkana and Olduvia Gorge. These early tool kits were sometimes carried over considerable distances of up to fourteen kilometers (Fagan, 2016). They thus represent a simple form of curation, retaining tools for future use rather than just utilizing convenient stones as chimpanzees do. With time, stone tools became increasingly complex, evolving into smaller core tools which were sharp-edged and angular to facilitate slitting skins to cut meat and soft plant matter, achieve deep penetration to kill game, and for shaping wood (Fagan, 2016, p. 58). Stone tools may have served additional functions as tools for mashing and hacking. The new tool kits included flakes, scrapers, projectiles and crude axe-like artifacts flaked on either side. According to Joukowsky (2015, p. 309), the consequent interaction of early man and his tools stimulated the increasingly rapid mental evolution of man during the period. It is the view of some anthropologists that sedentism, crop cultivation and the development of several specialists’ craft industries like potting, weaving and spinning followed this transition. About 3,500 Y.A., modern man (Homo sapiens) further developed greater efficiency in stone tool manufacturing using a new technique calledpressure flaking which differed considerably from the traditional old way of striking/hammering a small stone against another to remove chips off it (Sharer & Ashmore, 2005, pp. 346- 356). In pressure flaking, sharper more delicate tools were made by using a pointed tool to scrape, press and remove finer flakes off the core. These technologically superior tool types were smaller and were intended for a variety of purposes. Other materials known to have been used by early man to make tools included wood from which rudimentary clubs, spears and sticks were fashioned. These tools were used in tandem with stone tools for a long time until about 7,000 years Y.A. when man discovered the technological know-how of smelting iron and forging it into various tool types to enable him adapt/survive in his environment (Toth & Schick, 2010, pp. 47-61). In Ghana, several lithic artifacts comprising polished ground stone axes (Nyame akuma), stone beads (“aggrey” beads), stone rubbers, querns, microliths and grinding stones have been recovered together with fauna, molluscs and other artifacts like pottery across all sixteen regions of the country indicating much of modern Ghana was settled during the Late Stone Age. Many of these artifacts

Page 24 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana were found accidentally by expatriate engineers and geologists while exploring, dredging and exploiting mineral resources across the country. Others were found during the construction of roads, railways and other major construction works like dams during the colonial and post-independence eras. One of the earliest recorded references of the recovery of stone artifacts from Ghana was by the Danish missionary Monrad who found polished ground stone axes at an ancient Late Stone Age settlement at Osu, located in the Greater Accra Region in 1822. Another significant recovery was by Kitson, then Director of the Geological Survey Department. Over a three-year period, he recovered thousands of quartz and quartzite stone tools made by these food producing cultures across the country which he described as, “round ended, flattened pieces of weathered clay stones or burnt clay which are either plain or ornamented with longitudinal, transverse, and oblique cuts, making check patterns of various kinds” (Kitson, quoted in Anquandah, 1982, p. 9). He also noted that the local people occupying the areas where the stone artifacts were found had no idea what these objects were used for, supporting the view that the country was occupied long before the present local peoples arrived. Other personalities who made important lithic discoveries and documented their finds during the early twentieth century included N.R. Junner, Thurstan Shaw, Richard Nunoo, A.W. Lawrence, Richard York, Duncan Mathewson and Oliver Davies. Methods of Data Collection Three surface surveys were conducted on foot during the dry season of the summer of 2016 at each of the settlements. This is a period when much of the overlying vegetation is either withered or dead affording maximum exposure of the ground surface in many places to facilitate direct observation of cultural materials on the ground. Except for Shai where much of the overlying soil was composed of dark brown humus, the other settlements were composed of reddish-brown/dark- brown lateritic soils much of which had undergone extensive disturbance primarily because of farming. This observation was based on the fact that from ground surface to about 10 cm below ground surface was loose and friable, while soils underlying this level were hard and compacted. The vegetation was secondary, decimated by regular tree felling for fuel wood to smoke fish which is the principal vocation of the women-folk of Kpone and Prampram. Cultural materials retrieved from the surface surveys are shown below in Table 1.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 25 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) Table 1: Cultural materials retrieved during the surface surveys at Sega Hill, Prampram, Shai and Kpone together with their quantum and percentage values.

Cultural materials Sega Hill Kpone Prampram Shai 1 Alcoholic beverage bottles 165 4 87 13 2 Querns 12 - 3 - 3 Grinders 8 1 6 4 4 Glass beads 96 - 139 17 5 Stone slabs 7,922 27 376 503 6 Local potsherds 765 96 598 324 7 Imported ceramics 98 12 63 16 8 Bones 101 41 121 20 9 Molluscs’ remains 107 52 88 22 10 imported smoking pipes 56 2 29 16 11 Iron slag 34 - 19 4 12 Locally made smoking pipes 4 - - 2 13 Metal objects 63 2 104 29 14 Bauxite beads 7 - - 1 Total 9438 237 1633 971 Archaeological excavations followed the surface surveys. Sega Hill was the first settlement to be investigated followed by Kpone and Prampram. Sega Hill is currently abandoned but Prampram, Kpone and Shai have been continuously settled from pre-European contact times to present. Ethno historical traditions of the people of Kpone, intimate that Sega Hill was their ancestral home before their relocation, which was necessitated by a chieftaincy dispute between two sons of Angmo Keteku, chief of Sega after his demise (Kpone Traditional Council, C.S.O. 21/22/1421. 1985). The author was assisted during this phase of the research by some students from the Department of Archaeology and Heritage Studies, University of Ghana and locals from the towns investigated. Twenty centimeters arbitrary levels were used to control vertical provenience, while ¼ inch mesh was used to sieve soil from each unit. At Sega Hill, the entire site was gridded affording the use of the simple random sampling technique to select grids for excavation. The simple random sampling technique has two main advantages. According to Sharer and Ashmore (2017, p. 133), it “enables a sizeable portion of data to be collected from a given archaeological pool. It also enables each grid in the sampling frame to have a statistically equal chance for selection thus, removing any opportunity for selection bias by the researcher”. The quadrats of grids selected for excavation were P33 for

Page 26 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana Unit 1 (1 m. x 2 m.) and W60 for Unit 2 (1 m. x 1.5 m.). The total area excavated was 8 sq. m. representing 0.93% of the total area gridded. The sterile levels for the two units were 62 cm. and 121 cm. respectively below ground surface. Cultural materials retrieved from the two units at Sega Hill are shown below in Tables 2 and

Table 2: Types and quantum of cultural materials retrieved at Sega Hill - Unit 1. 3. Unit 1 Cultural Material Type Level 1 Level 2 Level 3 Total % values 1 Bottles 17 9 - 26 1.7% 2 Local pottery 106 79 23 208 14.1% 3 Stone slabs 23 9 1 33 2.2% 4 Molluscs 116 87 43 246 16.6% 5 Bones 32 56 14 102 6.9% 6 Grinders 2 1 - 3 0.20% 7 Glass beads 13 7 - 20 1.35% 8 Daub 17 4 5 26 1.76% 9 Imported) smoking pipes 9 4 - 13 0.88% 10 Metal objects 11 7 2 20 1.35% 11 Imported ceramics 13 9 - 22 1.49% 12 Iron slag - 4 9 13 0.88% 13 Querns 3 - 1 4 0.27% Total by strata levels 362 279 98 736 1475

Table 3: Types and quantum of cultural materials retrieved at Sega Hill - Unit 2. Unit 2 Cultural Level Level Level 3 Level Level Level Total % material 1 2 4 5 6 values type 1 Bottle 12 23 12 - - - 47 3.0% fragments 2 Local pottery 29 21 4 41 28 9 132 8.5% 3 Stone slabs 9 7 2 11 4 1 34 2.2% 4 Mollusc 9 11 7 23 19 7 76 4.9% remains 5 Bones 23 19 9 3 7 2 63 4.0% 6 Grinders - 2 2 - - - 4 0.2% 7 Glass beads 18 19 - 7 - - 44 2.8%

Legon Journal of the Humanities Vol. 30.2 (2019) Page 27 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) 8 Metal objects 1 7 3 1 - - 12 0.7% 9 Daub 16 23 17 13 16 2 87 5.6% 10 Imported - 6 3 - - - 9 0.5% smoking pipe 11 Locally - 2 - - - - 2 0.13% smoking pipe 12 Gaming disc - 1 - - 1 - 2 0.13% 13 Imported 11 12 - 9 - - 32 2.0% ceramics 14 Iron slag - - 6 13 9 - 28 1.82% 15 Palm kernel 106 34 19 12 15 9 195 12.6% shells 16 Glass window - 1 1 - - 2 10.12% pane Total by 234 188 84 134 99 30 769 1538 strata level Unlike Sega Hill, Kpone and the ancient settlement quarter at Prampram could not be gridded because of the stony nature of the ground as well as financial constraints. The choice of where to excavate was informed by density of surface artifact scatters. Two units, designated Unit 1 (1 m. x 2 m.) and Unit 2 (1 m. x 2 m.) were opened at Kpone and the total area excavated was 4 sq. m. representing approximately 0.87% of the total area surveyed. The sterile level of both units was 80 cm below ground surface. The types and quantum of cultural materials retrieved at Kpone are shown below in Tables 4 and 5. Table 4: Types and quantum of cultural materials retrieved at Kpone - Unit 1. Unit 1 Cultural Level 1 Level 2 Level 3 Level 4 Total % values Material Type 1 Metal object 2 - 2 3 7 0.75% 2 Local pottery 79 27 19 31 156 16.7% 3 Bottle fragments 2 8 3 - 13 1.39% 4 Glass beads 2 14 3 - 19 2.0% 5 Stone slabs 39 23 29 21 112 12.0% 6 Molluscs 143 211 97 34 485 52.2% 7 Bones 54 35 26 21 136 14.6%

Page 28 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana 8 Querns - 1 - - 1 0.10% Total 321 318 179 110 929 100 Table 5: Types and quantum of cultural materials retrieved at Kpone - Unit 2, Unit 2 Cultural Level 1 Level 2 Level 3 Level 4 Total % Values Material Type 1 Local pottery 121 78 41 36 276 27.1% 2 Stone slabs 21 10 3 7 41 4.03% 3 Molluscs remains 97 111 63 21 292 28.7% 4 Bones 77 34 21 23 155 15.2%% 5 Iron slag 4 12 15 9 40 3.93% 6 Glass beads 2 4 - - 6 0.58% 7 Grinders 1 - 1 - 2 0.19% 8 Imported 3 5 2 - 10 0.98% smoking pipes 9 Palm kernel 24 81 16 23 144 14.15% shells 10 Imported 2 12 - - 14 1.37% ceramics 11 Bottle fragments 13 23 - - 36 3.53% 12 Coin 1 - - - 1 0.098% Total 366 370 162 119 1017 100 One unit designated Unit 1 (1 m. x 3 m.) was excavated at the ancient settlement quarter at Prampram. Unlike Sega Hill, much of the ground surface at Prampram was bereft of surface artifact scatter except for 12 mounds which appeared to be collapsed house structures. The total area excavated was 3 sq. m. representing 1.2% of the area surveyed and the sterile level for Unit 1 was 160 cm. below ground surface. The stratigraphy at Prampram comprised a mix of very large stone boulders in very compacted soils which significantly slowed down the excavation work and made artifact recovery difficult. The types and quantum of cultural materials retrieved are shown below in Table 6.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 29 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) Table 6: Types and quantum of cultural materials retrieved at Prampram - Unit 1.

Unit 1 Cultural Level Level Level Level level Level Level Level Total % material 1 2 3 4 5 6 7 8 value types 1 Bottle 9 13 45 61 33 - - - 161 7.6% fragments 2 Local pottery 39 57 34 121 153 32 16 19 471 22.3% 3 Stone slabs 3 7 9 11 6 3 - 1 40 1.8% 4 Mollusc 145 207 211 57 43 10 21 9 703 33.3% remains 5 Bones 31 12 7 7 9 13 7 6 92 4.3% 6 Palm kernel 99 43 21 52 11 9 14 21 270 12.7% shells 7 Glass beads 13 21 32 12 3 - - - 81 3.8% 8 Querns - - 1 - 2 - - - 3 0.14% 9 Metal objects 4 7 13 31 19 2 3 - 79 3.7% 10 Imported 12 16 9 21 19 - - - 77 3.6% ceramics 11 Daub 5 11 7 14 2 6 2 1 48 2.2% Imported 13 21 10 17 5 - - - 66 3.1% smoking 12 pipes 13 Iron slag - - - - - 3 7 1 11 0.52% 14 Grinders - - 1 - - 2 - - 3 0.14% 15 Writing slate - 1 2 - - - - - 3 0.14% 16 Mirror - - 1 - - - - 1 0.004% fragment 17 Dress button - - 1 - - - - - 1 0.004% Total 373 416 404 404 305 80 70 58 2110 100 The purpose of the ethnographic research was to facilitate the comparative study of stone devices from archaeological and ethnographic contexts. It was also to discover if they shared similar attributes to suggest they were made by the same ethnic group. A total of 37 respondents consisting of 21 males and 16 females, aged 57 – 91 years from different families and occupations were interviewed. They included fishermen, traditional priests and priestesses, fishmongers, traditional medical practitioners, birth attendants, farmers and canoe builders. All the respondents interviewed resided at the ancient settlement quarters of Kpone, Shai and Prampram where traditional norms, values and customs are highly regarded, appraised and practiced despite the strong in-roads made by Christianity and

Page 30 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana globalization. They were thus considered a formidable group from whom data about the past could be derived. Majority of the respondents were not fluent in English and could not express themselves properly in that language. The interviews, which were recorded on tape, were therefore conducted in Ga and Dangme which are the principal medium of communication in that quarter. A five-page questionnaire was also devised to guide the ethnographic investigation and all respondents answered the same questions. The recordings were later played-back and transcribed from the Dangme/Ga languages to English by the author who speaks both languages fluently. Discussion It appears from the combined evidence (archaeological, ethnographic and historical data) that ancient Dangme populations settled in the research areas partly because of their abundant stone resources. Other factors probably included fertile soil and plentiful wildlife resources. Research by several scholars (Anquandah, 1979, 1982, 1986, 1987, 1996; Gblerkpor, 2005; Biveridge, 2005) in the research areas clearly indicate that stone slabs/blocks constituted an integral raw material in Dangme traditional architecture in the past. Notable edifices known to have been built with stone blocks/slabs included agricultural terraces, compound walling systems, house foundations and fortifications. At Sega Hill, 2,929 stone slabs were noted by the researcher. Out of this number, 217 were retrieved from excavations. The bulk were composed of granite and gneiss and were not standardized in shape and size. The blocks appeared to have been hacked from the several rocky outcrops along the beachfront because colourations and joint formations were similar. Two indigenes of Kpone and Prampram, Oglietse (pers. com. 14th June, 2016) and Ablade (26th June, 2016) posited that the smaller slabs were utilized as floor tiles inside rooms and in the vicinity of house compounds whilst the large/thick ones were used as buttresses to support house foundations. This construction style according to the respondents curbed gully erosion around house foundations. Anquandah (1986, p. 67) found similar stone slabs of what appeared to be “a very large dwelling complex of a notable personality with a sizable family” which measured 25 m. (east to west) and 20 m. (north to south) during pre-excavation surveys at Adwuku south west. The complex contained granary platforms, sleeping rooms and courtyards constructed with stone slabs. Anquandah (1987, pp. 233 - 234) asserted: “The entire complex in its original state was well protected like a mini citadel. It had an entrance with a gate on the western front with a staircase of granite slabs leading to the entrance. What emerges is that the surface structural arrangements belong to the 18th – 19th century and they consist of mud structures erected on

Legon Journal of the Humanities Vol. 30.2 (2019) Page 31 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) granite stone slabs, paved with floor tiles, each measuring on the average 20 cm. x 10 cm. and manufactured by local potters using coarse local clay.” Stone blocks also appeared to have constituted the primary resource for the construction of secret fortified communal sanctuaries in ancient Dangme-land. These structures some of which were monumental such as those at Hiowey and Adwuku Hills were free-standing defensive walling enclosures, erected by laying course by course, well-fitting granite blocks, one atop another between rock boulders till the desired height and lengths were achieved. A small opening at one end served as the entrance. John Agbo Opata (pers. com, 16th July, 2016), an indigene of Doryumu intimated that his great grandfather, one of several Dangmes’ forcefully transplanted from Hiowey Hill by the British in 1892, before his passing, told him that during their sojourn at Hiowey Hill, the entire community regularly undertook stone gathering exercises from their environs which were intentionally piled midway along major pathways leading to their hill top settlements. Stones were also piled across difficult to access locations along the edges of steep cliffs where the community had a clear visible advantage from their secret hill-top encampments. These strategic hide-outs together with their several stone compilations were purposely and readily accessed and hurled against slavers and other enemy groups intent on attacking them. Nii Affum Atteh (pers. com. 2nd May, 2017), a 93 year old indigene of Doryumu and a farmer by occupation also posited that traditional hunters in the past were obliged by custom to make concerted efforts during hunting expeditions to seek, discover and bring to the knowledge of clan elders natural crevices within numerous rock boulders and rock shelters they encountered. According to Atteh, the purpose of discovering protective shelters was to utilize them as safe havens for women and children during crisis periods. Their locations were known to only a few and not disclosed to the community for security reasons. A total of 12 stone grinders/milling stones (fig. 2)) and 6 querns were retrieved from the archaeological record. It is also worth noting that all 27 homes visited to derive ethnographic data still used traditional grinders to process vegetables and other botanical resources despite the popularity of electronic blenders. The bulk of the grinders and querns recovered at Kpone, Prampram and Sega Hill were composed of granite or gneisses. However, a few were composed of quartzite. The grinders appeared to have been hacked from the numerous rock boulders along the beachfront because in terms of colour, linear markings and joint formations, they appeared to be similar. Grinders from the study areas can be classified broadly into two types namely; mobile and stationery grinders. The former were small in size, approximately 25 cm. (breadth) x 7 cm. (height) x 35 cm. (length). They were of no definite shape, though

Page 32 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana 23 (representing 85.1%) were roughly rectangular shaped. Only 7 of the latter were discovered in the research areas. They measured averagely 120 cm. (length) x 12 cm. (width) x 86 cm. (breadth) and were naturally fixated to underlying bedrocks. The latter type also had 6 - 11 segmented depressions, each measuring approximately 3-5 cm deep along the grinding areas indicating about 6 to 11 persons probably used it in tandem in the past to grind. The stationary grinders were probably communally owned because unlike the mobile grinders, they were located in open spaces outside individual homes. Unlike the stone grinders, all the recovered querns were broken almost in half. A close examination with a hand magnifying glass revealed several linear striations along the grinding areas suggesting extensive and continuous usage over a long period of time.

Figure 2: Part of a stone grinder from Kpone (Picture: Author). Archaeological and ethnographic investigations by Anquandah (1986) at Le and Shai revealed that Dangme groups that settled in the area also utilized another type of stone milling device, which is different from that described above to process staples and herbs. He noted that this rare milling equipment had been recognized only because of its survival in the ethnographic record. In a paper titled Accra Plains, Dangmeland: A case study in the eclectic approach to archaeological and historical studies, Anquandah (1987, p. 80) asserted: By the first half of the seventeenth century, several early settlements at Le and Shai had developed into urban complexes characterized by mud and stone houses,

Legon Journal of the Humanities Vol. 30.2 (2019) Page 33 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) agricultural terraces, stone walling and occasional fortifications, and each house had milling equipments locally called blemagiji that was constructed from wood, mud and grinding stones. Anquandah’s (1986) excavations revealed that stone was an important raw material for the production of several accoutrements/tools like grinders and scrappers in the research areas in the remote past. The type of stone exploited however, varied from community to community and depended on the intended purpose. According to Nene Abladu Attiope (pers. com. 26th June, 2016), a respected 82 year old retired fisherman resident at James Town, a suburb of Kpone, tapering “T” shaped stone hooks made primarily from water worn pebbles collected from the beachfront constituted an integral part of the tool kit of fisher-folks who plied their trade along the Kpone-Keta coastal stretch in the past. According to the above-named respondent, beach pebbles were the preferred media because of their durability which allowed for repetitive usage. It was also relatively harder and heavier allowing them to submerge faster compared to other media like wood and bone. Nene Abladu Attiope intimated that the tapering “T” shape was achieved by detaching bits and pieces of flakes off the original pebble by knocking it against another pebble or a hammer head. The purpose was to achieve an almost “T” shaped stone tool with pointed ends and sharp edges at right angles to the long axis. The resulting jagged-edges made a most effective lock-in tool capable of choking, constricting and inflicting injuries inside the throat of fish while it struggled to free itself. The device was secured to vegetal twinges made out of the fibrous branches ofUlva lobata, a soft but strong seaweed belonging to the phylum Chorophyta. Locally called womogmo, it grows naturally and is abundant along the eastern coastal belt where it is generally considered a nuisance/impediment to fishing. Nene Abladu Attiope further intimated that after harvesting, the branches were tied together at vantage points to achieve a desired length after which it was sun-dried for about two weeks to make it strong. The twinge enabled one end of the “T” shaped stone trap to be lowered into, or raised from the sea by the fisherman who held onto the other end. The soft fibrous tissue also allowed for tautness to facilitate drawing upwards after it had caught fish. The inedible offal’s of molluscs were oftentimes tied to the ends that submerged in water and served as bait to trap some deep sea fish species in the past (Nene Abladu Attiope, pers. com. 29th May, 2017. Another respondent Nii Tetteh Angmo (pers. com. 27th May, 2017), a 78 year old indigene of Kpone but currently residing at Prampram, during their childhood years, he and other apprentice colleague fishermen used sharp-edged pointed stone lances secured by vegetal twinges to long wooden poles (fig. 3) to spear freshwater

Page 34 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana fishes in theGao and Laloi Lagoons. To make the tool more effective, the edges of the lancets were sometimes grounded and polished on stone blocks to produce finer and tougher cutting edges than could not be achieved by chipping only.

Figure 3: Artistic illustration of the stone lance described by Nii Tetteh Angmo (Drawing: author). Nii Tetteh Angmo (pers. com. 12 July, 2016) also informed the author that they employed two methods in the past to secure their canoes on-shore after berthing to prevent them from being washed off-shore during rain and wind storms. The first involved securing one end of a rope to a canoe and securing the other end to weighty stone boulders submerged in the ocean close to the coastline (fig. 4). The other assured method involved securing one end of a rope to a canoe and tying the other end around any of the many rocky boulders or coconut trees located off-shore.

Figure 4: A stone anchor from Prampram beach. Note the smooth horizontal indentation/notch on the upper right side caused by regular weighty suspension of rope attachments along that segment (Picture: Author).

Legon Journal of the Humanities Vol. 30.2 (2019) Page 35 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) Another respondent Mr. Attiapa Otetse (pers. com. 27th July, 2017), a retired school teacher and part-time hunter from Shai informed the author that hunters in this home area in the past utilized stone blocks as anchors to prevent loss of traps and game. According to the above named respondent, some animals like deer, giant rat and grasscutter sometimes struggled to disengage/free themselves from traps they had been caught in. In the process, these animals occasionally dragged the traps away from the place where the hunter originally set them to die elsewhere, leading to loss of both trap and game. At the ancient settlement quarters of Kpone and Prampram, rocks with natural notches/indentations were sometimes utilized as watering and feeding troughs for chickens, goats and sheep (fig. 5). Out of 35 household visitations (Kpone -21 and Prampram – 14), 23 used these stone watering troughs. The study also revealed that in some households, disused stone grinders were oftentimes not discarded but used for this purpose.

Figure 5: One of several stone troughs used to hold water/food for domesticated animals. The use of stone blocks as whetting edge tools is common in the research area despite the commonality of hand operated mechanical devices for that purpose. Nuumo Tettey Abladu Anson (pers. com. 3rd July, 2016), a subsistence farmer by occupation intimated that it was common practice among local croppers to use large stone blocks to sharpen the cutting edges of their cutlasses, hoes, axes and

Page 36 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana other tools which required sharp cutting edges to facilitate their effective use. Stone blocks utilized for this purpose were mostly located on their farmsteads. However, a few were also discovered inside or close to house compounds. The Dangme were thoroughly proficient, highly skilled and adept at constructing agricultural terraces attested by the several well laid out courses of free standing dry stone walling along the slopes and below their hilltop encampments at Shai and Sega Hill. Each course consisted of hundreds of stone blocks of no identifiable shape or size. A few of the terraces, especially those along the very edgy slopes sometimes consisted of more than one course and still stand intact after several centuries of constant exposure to exfoliation and the other vagaries of tropical weather conditions. This method and design of constructing agricultural terraces appeared to have been embraced by other Dangme groups like the Shai settled in the interior. In a paper presented by Anquandah (1987: 19), he noted that “during the 19th century construction of agricultural terraces with dry stone walling and the process of urbanization was initiated in the Shai areas and further expanded in the following centuries”. The Dipo puberty rite is the most important socio-religious event in Dangme- land and constitutes an integral aspect of traditional Dangme culture. Oral traditional data gathered by Marijke Steegstra (2005) indicates that Dipo had its antecedence in antiquity and that the various rites associated with its practice were initiated by the female deity Nana Kloweki. In the past, in some Dangme communities like Krobo, failure to successfully pass the cleansing/purification rites and rituals resulted in the young girl and her family being bastardized or ostracized from the community. Celebrated annually between March and June in most Dangme towns, Dipo is a “rites des passage” for female adolescents. In theory, “Dipo” acted as a check against teenage pregnancy prior to marriage and ensured that families exercised considerable control on the matrimonial process to guarantee social stability in the community. It is common belief among Dangmes that successfully passing the Dipo puberty rite automatically transitions the tutee spiritually and physically from one stage of life to another. During the period of seclusion from their families and society which lasted about two years in the past, tutees were also taken through a process of traditional nurturing (social and cultural history education), child nurturing and upbringing, home keeping and cooking. Other areas of nurture included family and matrimonial education, introduction to crafts such as potting and bead production, Hayo dancing and Klama song recitations. During the final supplicatory rites, the tutees are made to sit three times on a flat almost circular sacred ritual stone called tegbete (also called the Kloweki rock) by the Dipo priestess (se woyo). It is worth noting that the tegbete stone is itself seated on a

Legon Journal of the Humanities Vol. 30.2 (2019) Page 37 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) number of smaller stones laid out in a circular formation inside the Kloweki shrine which houses the Dipo deity. Inability of a tutee to rise from the tegbete indicated she was either not a virgin or was secretly pregnant at the time of undertaking the rite. When that occurred, special pacification rites needed to be performed on her behalf. According to Dangme custom, one’s inability to perform the required propitiation and appeasement rites could invoke curses from the deity. Perhaps, my encounter with aunty Ayah Amouku, a revered 77-year-old traditional medical practitioner and birth attendant can be described as the most illuminating and informative interaction during the ethnographic research. An indigene of Kpone with over 45 years experience in ethno-medical practice, her in-depth knowledge of medicinal herbs was mind boggling. She is credited with over 65 successful childbirths and renowned for her curative prowess at remedying most gynecological, prenatal and postnatal disorders in the community. Ayah Amouku (pers. com. 21st July, 2016) recounted that as a teen undertaking a five- year apprenticeship under one Nene Korkor Abadge at Osudoku, she witnessed several minor surgical operations involving the use of specialized stone blades which were used to create incisions across specific areas of the body into which herbal concoctions with medicinal properties were inserted to effect cure of some physical ailments She indicated that in the past, a variety of stone blades were also used in achieving facial and body scarification marks. She intimated that prior to cauterizations, the surgical blades, which measured approximately 8 cm x 4 cm were either first steamed in boiling water or overhung over smokeless burning charcoals for several minutes, purposely to sterilize them, after which they were wrapped in clean damp white calicos to cool. She further intimated that prior to the popularization of Western orthodox surgical procedures, most traditional medical practitioners in Dangme-land used stone blades to undertake minor surgical procedures and that unlike metal blades, incisions achieved with stone blades healed faster and left no physical scars on the affected area after healing. Stones are also important utility items used inside the shrines of some revered deities in the research area. At Kpone for example, a shrine dedicated to the Gao and Aya deities were externally enclosed with free-standing stone walling from the basal to about the shoulder level and overlaid with woven palm fronds. After making two schnapp bottles offering to the priest, the author was allowed an inside view of the shrine on condition that no snapshots would be taken. Stones were also used as devices to vertically support a number of local earthen pots and bowls laid out in a circular formation before the fetish. Nii Atobie Oti (pers. com. 3rd July, 2017), one of two wolomo’s (traditional priests) interviewed asserted that for purposes of purity and sanctity, the pots together with their contents which among

Page 38 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana other things contained water, herbs and tree barks were never to be placed on the ground. The other derivatives of the pot were not divulged because the author was neither an acolyte nor an inductee of the cult. Another stone of interest inside the shrine was a large rock-bed on which otoo (meal of mashed yam mixed with palm oil and boiled eggs) and occasionally animals were offered to the deity to appease the gods, atone for sins and sometimes reverse individual and community curses. In spite of the popularity of kerosene, gas and electric cookers in the research area, several families and commercial food vendors continue to use traditional stone hearths to prepare foods like kenkey and banku. According to all the respondents, the use of these traditional cookers extended into antiquity, long predating the arrival of Europeans. The cookers usually comprised three free-standing stones of equal height arranged in a triangular formation to support a cooking pot. Each of the stones had flat bases which enabled them stand upright. Fired firewood and charcoal were passed between two of the stones to heat the pot (fig. 6).

Figure 6: A stone hearth cooker in use at Kpone (Picture: Author)

Conclusion Stone probably constituted the most important natural resource which enabled various Dangme populations settle in the research areas to adapt to their environment. Several tool types made out of stone also facilitated the exploitation of a variety of resources in the study areas. For example, fishing, crop farming,

Legon Journal of the Humanities Vol. 30.2 (2019) Page 39 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) tapping and hunting, the four major vocations of the people all depended on the use of stone tools. Stones were also exploited for other purposes including in traditional religion, medicine and architecture. The piling of stones at vantage points around their encampments is an indication that stone constituted a cheap but effectual pre- emptive weapon used by the people to defend and sometimes attack slavers who actively operated in the region during the Trans Atlantic slave era (1620-1807). The exploitation of stone for the above purposes has diminished significantly over the last century due primarily to Christianity, globalization and modernism. For example, the use of stone as fish and game entrapments is currently non-existent in the study areas. However, the use of stones as devices for sharpening cutting tools like cutlasses and axes, for pulverizing medicinal plants, in traditional worship and for cooking is still a prominent practice among the indigene population.

Page 40 Legon Journal of the Humanities Vol. 30.2 (2019) Biveridge, F./Ethno archaeological Clues to Stone Exploitation in Ancient Dangme-land, Greater Accra Region, Ghana References Anquandah, James. (1979). Accra plains archaeological and historical project. Nyame Akuma.No. 15 (23-37) Anquandah James. (1982). Rediscovering Ghana’s past. London: Longman Group. Anquandah, James. (1986). Ethnoarchaeological clues to Ghana’s great past. Universitas 8, 113-130. Legon Anquandah, James. (1987). Accra Plains Dangme-land- A Study on Ethno archaeology and Iron Age Sites. Nyame Akuma, 21, 27-34. Anquandah, James. (1996). Accra plains Dangme-land: A case study of the Eclectic Approach to archaeological and Historical Studies. FASS Bulletin. Barbot, John. (1732). A Description of the Coast of North and South Guinea, Churchills collection of voyages and travels. London. Biveidge, Fritz. (2005). An archaeological investigation of the Kpone coastline and its implications for understanding the Coastal Dangme. (Unpublished Master of Philosophy Thesis). Department of Archaeology and Heritage Studies, University of Ghana, Leegon. Boahen, Adu Albert. (1977). Ghana before the coming of Europeans. Ghana Social Science Journal, 4, 94. Bosman, William. (1705). Accurate description of the Coast of Guinea. London: Frank Cass. Dickson, K. B and Benneh, K. (1973). A new geography of Ghana. London: Longman Group. Fagan, Brian. (2016). World prehistory, A brief Introduction. New Jersey: Pearson Prentice Hall. Gblerkpor, William. (2005). An Archaeological Investigation of the Krobo Mountain Dry-Stone Terraces. (Unpublished Master of Philosophy Thesis). Department of Archaeology and Heritage Studies, University of Ghana, Legon. Joukowsky, Martha. (2015). A Complete Manual of Field Archaeology. New Jersey: Prentice-Hall, Inc. Englewood Cliffs. Kesse, G.O. (1985). The mineral and rock resources of Ghana. Rotterdam: A. A. Balkema Publishers. Kpone Traditional Affairs. (1985). (C.S.O. 21/22/1421. 1985. Kpone Traditional Council) Kpone. Kropp-Dakubu, M.E. (1976). The contribution of linguistics to the study of

Legon Journal of the Humanities Vol. 30.2 (2019) Page 41 Biveridge, F./Legon Journal of the Humanities Vol. 30.2 (2019) Ghanaian history. Proceedings of the Seminar on Ghanaian Historiography and Historical Research. 20-22 May, 1976 (Ed.) J.O. Hunwick. (26-42). Meredith, Henry. (1812). An account of the Gold Coast of Guinea. London: Frank Cass. Nicholas, Toth and Schick, Cathy. (2010). The evolution of humanity. In C. Scarre (Ed.). The human past: World prehistory and the development of human societies. London: Thames and Hudson. Oboli, H.O.N. (1998). A new outline Geography of West Africa. London: Harrap Books. PRAAD (Public Records and Archives Administration Department) Certificate of Authentication No. 000372. Accra. Sharer, R. J. and Ashmore, Wendy. (2017). Archaeology: Discovering our past. California: Mayfield Publishing Company. Steegstra, Marijke. (2005). Dipo and the politics of culture in Ghana. Accra: Woeli Publications Services. Personal Communication Ayah Amouku. 21st July, 2016 Attiapa Otetse. 27th July, 2017 John Agbo Opata. 16th July, 2016 Mooko Ablade. 26th June, 2016 Nene Abladu Attiope. 26th June, 2016 Nene Oglietse. 14th June, 2016 Nii Afum Atteh. 2nd May 2017 Nii Atobie Oti. 3rd July, 2017 Nii Tetteh Angmo. 27th May, 2017 Nuumo Tettey Abladu Ansaon. 3rd July, 2016

Page 42 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019 DOI: https://dx.doi.org/10.4314/ljh.v30i2.3

Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana1 Kwabena Sarfo Sarfo-Kantankah Senior Lecturer Department of English University of Cape Coast, Cape Coast, Ghana Email: [email protected] Joseph Arko Lecturer Department of English University of Cape Coast, Cape Coast, Ghana Email: [email protected]

Submitted: October 31, 2018/ Accepted: June 23, 2019/ Published: December 30, 2019

Abstract Due to its supervisory responsibility over the executive and public institutions, including ministries, departments and agencies (MDAs), Ghana’s legislature has a mandate to check and fight corruption and make its practice a high risk, low-gain activity. The commitment, urgency and the willingness to fight corruption can be seen not only in creating the legislative framework for addressing corruption, promoting governmental accountability and acting as a check on the executive, but also in how the subject of corruption is broached in parliamentary debates and discussions, an area which relatively lacks scholarly attention. First, using an interpretive content analysis, this paper studies the approaches to the fight against corruption in Ghana by examining the Parliamentary Committee report on the National Anti-Corruption Action Plan (NACAP) and, the debate thereof, in order to uncover the focus of the NACAP. Second, the paper examines Parliamentarians’ use of evaluative language in parliamentary debates to enact otherness and how this othering affects the fight against corruption in Ghana. The paper finds that the NACAP suggests a three-pronged approach to the fight against corruption, viz: ethical/educational, multi-stakeholder and legal approaches. However, the implementation of these approaches has been ineffective due to certain Ghanaian cultural practices. Again, the paper finds that as a result of othering between government and opposition MPs,the fight against corruption is only incidental in parliamentary debates, indicating a low parliamentary commitment to the fight against corruption. Thus, the paper suggests that the fight against corruption must be taken away from the corridors of the Parliament of Ghana.

Keywords: fight against corruption, parliamentarians, National Anti-Corruption Action Plan, content analysis, evaluative language, othering, Ghanaian parliamentary discourse 1 The paper is a follow-on to Sarfo-Kantankah’s (2018) ‘Corruption is a big issue: a corpus-assisted study of the discursive construction of corruption in Ghanaian parliamentary discourse’, Legon Journal of the Humanities, 29(1), pp.226-258.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 43 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) Introduction Ghanaian Members of parliament (MPs) recognise that there is pervasive corruption in Ghana (Sarfo-Kantankah, 2018). In other words, Ghanaian MPs, to a very large extent, understand and are able to ‘communicate the sense in which corruption can be thought of as a danger to’ Ghana’s socio-political and economic development (Bukovansky, 2006, p. 199). It has been found that Ghanaian MPs discursively construct corruption as a huge systemic social canker that hinders the socio-politico-economic development of Ghana, suggesting ‘that stronger measures and more formidable parliamentary commitment are needed to fight corruption’ (Sarfo-Kantankah, 2018, p. 226). It is interesting to know that the views of the Ghanaian MPs echo the United Nations Office on Drugs and Crime’s (UNODC) (2018, no page) statement that: Corruption is a complex social, political and economic phenomenon that affects all countries. Corruption undermines democratic institutions, slows economic development and contributes to governmental instability. Corruption attacks the foundation of democratic institutions by distorting electoral processes, perverting the rule of law and creating bureaucratic quagmires whose only reason for existing is the soliciting of bribes. Economic development is stunted because foreign direct investment is discouraged and small businesses within the country often find it impossible to overcome the ‘start-up costs’ required because of corruption. With such a strong conviction of the problems of corruption, Ghanaian MPs will be expected to fight corruption in an uncompromising manner. However, it does not appear that MPs are genuinely committed to the fight against corruption. The purpose of this paper is, therefore, to examine and critique, through an interpretive content analysis, the current measures laid down by the Ghanaian Parliament in the National Anti-Corruption Action Plan (NACAP) for the fight against corruption. The study also examines the use of evaluative language to enact otherness in parliamentary debates on corruption in order to investigate how otherness affects MPs’ attitudes and commitment to the fight against corruption. Thus, the paper is divided into two complementary parts. The first part examines the current methods for fighting corruption as found in the Parliamentary Committee report on NACAP. The purpose is to demonstrate that there are legislative measures for fighting corruption. The second part, through the principle of evaluation, looks at parliamentary debates in order to explore how parliamentarians broach the fight against corruption and the attitudes they demonstrate. This is necessary because the commitment to implementing the NACAP largely depends on the attitudes of the MPs towards the fight against corruption. The assumption here is that while Parliament has legislative tools to combat corruption, MPs’ attitudes indicate low

Page 44 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana commitment towards the fight against corruption. Therefore, the paper seeks to respond to the following questions: 1. What are the proposed approaches to the fight against corruption, as contained in the NACAP, and why have those approaches largely failed? 2. How do the MPs enact othering in their debates to circumvent their commitment against the fight against corruption? In our view, it is not enough to have a legislative framework aimed at fighting corruption. What is needed is a strong political will and commitment to implement such framework. And understanding the commitment of MPs towards fighting corruption requires an assessment of the attitudes portrayed by MPs during their debates. It is in this sense that this paper is significant, especially considering that it engages the concept of othering, through evaluation, to uncover MPs’ attitudes towards the fight against corruption. Whereas various studies have suggested ways of fighting corruption, it is unclear how many studies, if any, have done so from the point of view of the attitudes of Ghanaian MPs. We are interested in deploying the concept of othering because it allows us to see what engages MPs’ attention in their discourse of the fight against corruption. The concept of othering helps us to observe that, instead of focusing their attention on the fight against corruption, government and opposition MPs engage in the vilification of each other, thereby trivialising the issue of the fight against corruption. The fight against corruption There is no gainsaying that corruption has ravaged most African countries and, therefore, there is the need for strong measures to combat it. Studies have proposed several ways in which countries can successfully fight against corruption. Corruption can be fought through democratic and economic restructuring, strengthening state institutions (Börzela, Stahn and Pamuk, 2010); civic education (Marquette, 2007); enhancing transparency and accountability in state and economic institutions, and social empowerment (Johnston, 1998); and anti-bribery legislation (Cleveland, Favo, Frecka and Owens, 2009). Others include pressure exerted by international bodies such as the European Union and the World Bank (Gadowska, 2010) and strengthening global integrity and ethics (UNODC, 2018a). Also needed are transnational and multilateral cooperation (Bukovansky, 2006), comprehensive and multidisciplinary approach – comprising technical assistance and building institutional capacities (UNODC, 2003, p. 1). This is important because ‘corruption is no longer a local matter but a transnational phenomenon that affects all societies

Legon Journal of the Humanities Vol. 30.2 (2019) Page 45 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) and economies’ (UNODC, 2003, p. 1), and ‘[s]uccessful international cooperation is required to eradicate the corrosive effects of corruption’ (UNODC, 2018b, no page). Agbele (2011, p. 11) has stated that since the beginning of the fourth Republican and multi-party parliamentary democracy in Ghana, various governments have declared their commitment to fight corruption, yet they have remained mere rhetorical gimmicks. Agbele thinks that legislative reforms are cosmetic with huge implementation gaps, even though he admits that some success has been recorded through ‘greater public awareness, media pluralism, some sort of changes in the governance structure’ and ‘improvement in the legal framework’ [sic]. In its press release of 25th January 2017, after the release of the Corruption Perception Index (CPI) 2016, the Ghana Integrity Initiative (2017) recommended among other things that, to fight corruption in Ghana, the new political and institutional leadership must demonstrate commitment to the fight against corruption by strengthening anti-corruption legislation [emphasis ours] and closing identified gaps; passing outstanding laws such as the Right to Information (RTI) and the Public Officer’s Code of Conduct and sanctioning corrupt officials. According to GAN Integrity (2016, p. 1) Ghanaian anticorruption law is primarily contained in the Criminal Code, which criminalizes active and passive bribery, extortion, willful exploitation of public office, use of public office for private gain and bribery of foreign public officials. The Public Procurement Act, the Financial Administration Act and the Internal Audit Agency Act have been introduced to promote public sector accountability and to combat corruption. The government has a strong anti-corruption legal framework in place but faces challenges of enforcement. Gifts and other gratuities offered to civil servants in the aim of influencing their duties are illegal, nonetheless, facilitation payments are not defined in law. The statement raises the question of why the anticorruption laws face such implementation challenges. Could it be the result of how Ghanaian law makers see corruption? The view of this paper is that an understanding of MPs’ attitudes towards the fight against corruption may unearth some of the underlying challenges associated with the enforcement of anticorruption laws. A strong Parliament is key to the fight against corruption (Kyei-Mensah- Bonsu, 2013). The Parliament of Ghana is one of the main bodies entrusted with the power to fight corruption. Due to its supervisory responsibility over the executive and public institutions, including ministries, departments and agencies (MDAs), Ghana’s legislature has a mandate to check and fight corruption ‘and make its practice a high risk, low-gain activity’ (Kan-Dapaah, 2015, p. 17). Through its

Page 46 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana Public Accounts Committee (PAC), for example, Parliament is mandated to examine the audited accounts of public institutions to ensure accountability. PAC is a creation of the 1992 Constitution of Ghana, with a primary function of examining audited accounts of the government indicating funds granted by Parliament to meet public expenditure and such other accounts laid before the House (Parliamentary Centre, 2009). Article 103 (6) gives PAC the powers, rights and privileges of a High Court in terms of enforcing witness attendance, production of documents, inter alia. In fact, it is only Parliament that is empowered by the 1992 Constitution to give approval or otherwise for the imposition of taxes, determine the waiver or variation of taxes and it is only Parliament that can sanction the withdrawal of funds from the Consolidated Fund of the State. Parliament exercises control over the expenditure of funds (Parliamentary Centre, 2009). Article 187(2) of the 1992 Constitution states that: The public accounts of Ghana and of all public offices, including the courts, the central and local administrations, of the universities and public institutions of like nature, of any public corporation or other body or organization established by an Act of Parliament shall be audited and reported on by the Auditor-General. Thus, Parliament has the responsibility to monitor the extent to which governments and public bodies spend funds allocated to them. The foregoing demonstrates the power vested in Parliament for ensuring financial accountability in Ghana. This means that the failure of Parliament to diligently perform these functions has enormous implications for the fight against corruption in Ghana. The implication is that the success, or otherwise, of the fight against corruption largely depends on the level of the commitment of the MPs. In this sense, there is the need to study the attitudes of MPs towards the fight against corruption. Conceptual framework and review: othering Tracing its origins to Hegelian philosophy (Jensen, 2011; Linares, 2016) and drawing on the Lacanian psychoanalytical theory of the ego, Lévi-Struassian ethnography of the other, and Levinasian moral philosophy of ‘the face’ (Gabriel, 2012), othering acknowledges that the recognition and construction of oneself is achieved through the recognition of the other (Linares, 2016). Othering is considered as the transformation of a difference into otherness for the creation of an in-group and an out-group (Staszak, 2008). Otherness is the result of ‘a discursive process by which a dominant in-group (‘Us’, the Self) constructs one or many dominant out- groups (‘Them’, Other) by stigmatising a difference – real or imagined – presented as a negation of identity and thus a motive for potential discrimination’ (Staszak, 2008, p. 44). It is a mechanism that is used to distinguish one group from the other.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 47 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) As Barter-Godfrey and Taket (2009: 167) put it: This [othering] is dependent on naming the other to form ideas of group membership; defining otherness to identify some people as an in-group and others asanout- group; and marking the other so that the out-group members are less favoured when compared to the in-group. Othering allows for the creation of identities, serving as ‘a necessary condition of differentiation in relation’, and ‘ascribing otherness plays a necessary role in structuring individual identity’ (Cooey, 1990, p. 7). Othering is a differentiation strategy for labelling ourselves as more credible than others. Being a ‘process of casting a group, an individual or an object into the role of the ‘other’ and establishing one’s own identity through opposition to and, frequently vilification of the other’ (Gabriel 2012), othering is a potential ground for conflict between opposed groups. Our view is that otherness leads to social representation of opponents in a polarised manner. Such representation has the potential to drive attention away from the substantive social and political issues. The concept of othering has been explored from philosophy (e.g. Husserl, 1979); post-colonial theory (where it was originally coined (Said, 1978; Jensen, 2011)), multiculturalism and critical cultural perspective (Said, 1978; Spivak, 1985; see Linares, 2016); feminism (Montoya and Agustin, 2013); including the language classroom (Linares, 2016); health care and health communication (Johnson, Bottorff, Browne, Grewal, Hilton and Clarke, 2004; Barter-Godfrey and Taket, 2009); and media studies and discourse (Breen, Haynes & Devereux, 2006; Kamenova, 2014) and (critical) discourse analytical perspective (Riggins, 1997). The disciplinary domains where the concept of othering has been employed are numerous. In this study our focus is on political discourse generally and parliamentary discourse specifically. In political (discourse) studies, the othering theory has been considered, either directly or indirectly, from international politics and nationalism (Göl, 2005; van Houtum and van Naerssen, 2002); British politics (e.g. Charteris-Black, 2006, who explores immigration metaphors in the 2005 British election campaign). Others are found in American politics – e.g. Kinder and Kam (2010), who study the ethnocentric foundations of American opinion; Muller (2008), who studies ethnic nationalism in American politics; and Kaplan, Freedman and Iacoboni (2007), who investigate political attitudes and party affiliation among US presidential candidates; and Leudar, Marsland and Nekvapil (2004), who consider membership categorization in political discourse. In Australian politics, Duck, Hogg, and Terry (1995) have examined political identification and the third person effect in the 1993 Australian federal election. Lirola’s (2014) edited eleven-chapter volume, Discourses on immigration in times of economic crisis: a critical perspective, exploring othering/otherness in

Page 48 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana Spain, the US, Britain and Central Europe epitomises the concept of othering in recent times. A number of studies have examined othering in parliamentary discourse. These include: Baker (2015), who studies argumentation, legitimation, and othering in debates on refugees in the Canadian House of Commons; Kirkwood (2017), who looks at the humanisation of refugees in the UK parliamentary debates on the European refugee ‘crisis’; and Sosoni (2017), who investigates othering in the Greek parliament. Other studies are: exploration of the description of others in Western parliaments (van Dijk, 1997); parliamentary debates on immigration in Europe (van Dijk, 2000a); ideologies, racism and discourse (van Dijk, 2000b); while van Der Valk (2003) looks at Right-Wing parliamentary discourse on immigration in France. Lazaridis and Tsagkroni (2015) study the securitisation of migration and far right populist othering in Scandinavian countries. Whereas there have been studies on parliamentary discourse and/or corruption in Africa, hardly do we find studies that have used the concept of othering. For example, Mchakulu (2011) uses issue-specific frame theory for the study of Malawian parliamentary discourse, while Jakaza (2013) employs appraisal theory, pragma-dialectic theory of argumentation and controversy in studying Zimbabwean parliamentary discourse. Mwiinga (2015) employs a critical discourse analytical approach to study Zambian parliamentary discourse; and Peter, Mukuthuria and Muriungi (2016) use the Foucauldian discourse analysis theory to study Kenyan parliamentary debates. Amakali (2018) explores speech acts and their rhetorical purposes in the Namibian parliamentary discourse. In Ghana, Appartaim (2009) and Imbea (2010) study Ghanaian parliamentary discourse from a nativisation perspective; while Jibril and Gyasi (2017) examine English expressions in Ghana’s parliament in order to identify Ghanaian English features. Thus, utilising the concept of othering does not only conceptually add to the various ways in which parliamentary discourse can be studied, but also reveal how Ghanaian MPs’ debating strategies undermine the fight against corruption in Ghana. Our assumption here is that an engagement in otherness in parliamentary discourse is fundamentally a blame game, leading to a relegation of the commitment to the fight against corruption. In other words, by accusing each other as being corrupt, MPs divert attention from the sense of common purpose needed to tackle corruption. In order to reveal the othering features of Ghanaian parliamentary debates on corruption, this study deploys an evaluation analytical approach (cf. Thompson and Hunston, 1999).

Legon Journal of the Humanities Vol. 30.2 (2019) Page 49 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) Methodology

Data Two main datasets are used for this study. The first is the report of the Committee on Constitutional, Legal and Parliamentary Affairs on the National Anti-Corruption Action Plan (2012-2021). In the legislative process, before a bill is passed into law, it passes through five stages, including: first reading, when the bill is presented to parliament for the first time; committee stage, where the relevant committee scrutinises the contents of the bill and submits a report to Parliament for debate at the second reading stage, where Parliament debates the principles of the bill and the report submitted by the relevant committee (Armah, Sarfo-Kantankah, Bauer & Mensah, 2018). Thus, the first set of data comprises the Committee report on the NACAP and the subsequent debate on the report. The second set of data is made up of Hansards of debates relating to corruption. Analytical lenses The study employs two analytical approaches: interpretive content analysis and evaluation. While the former is used to assess the existing approaches to the fight against corruption as put forward in the NACAP, the latter is used to examine how othering is deployed by MPs in their parliamentary debates. Content analysis is the analysis of the content of spoken/written communication, including political speeches, through which the conscious and unconscious beliefs, attitudes, values, and ideas can often be revealed (Fraenkel, Wallen & Hyun, 2012). It is ‘a research technique for making replicable and valid inferences from texts (or other meaningful matter) to the contexts of their use [sic]’ (Krippendorff, 2004, p. 18), providing a systematic and objective means for making inferences from data (Downe-Wambolt, 1992). A useful method for examining trends and patterns in documents (Stemler, 2001), content analysis helps to organize and elicit meaning from the data collected and to draw realistic conclusions from it (Bengtsson, 2016). As noted by Berelson (1952, cited in Weber, 1990, p. 9), content analysis helps to: i. identify the intentions and other characteristics of the communicator; ii. determine the psychological state of persons or groups; iii. detect the existence of propaganda; iv. describe attitudinal and behavioural responses to communications;

Page 50 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana v. reflect cultural patterns of groups, institutions, or societies; vi. reveal the focus of individual, group, institutional, or societal attention; vii. describe trends in communication. Specifically, the current paper does an interpretive content analysis (Bengtsson, 2016), where it explores the underlying meaning and implications of the NACAP document and MPs’ debates around corruption. Thus, the content analysis enables us to discover and describe the focus of the NACAP and MPs’ debates on the fight against corruption. Evaluation refers to statements of value, an assessment of something as either good (that is, desirable) or bad (that is, undesirable) (Fairclough 2003: 172; van Leeuwen 2008: 109-110). According to Thompson and Hunston (1999: 5): evaluation is the broad cover term for the expression of the speaker or writer’s attitude or stance towards, viewpoint on, or feelings about the entities or propositions that he or she is talking about. That attitude may relate to certainty or obligation or desirability or any of a number of other sets of values. It means that evaluation covers some aspects of modality such as certainty and desirability, as well as the use of evaluative adjectives and adverbs. Evaluation can be linguistically (grammatically, lexically and textually) and conceptually constructed or recognised (Thompson and Hunston 1999: 13). Grammatically, evaluation is most obviously realised through relational processes (Fairclough 2003: 172), in the form ‘X is [seen by the writer as] Y’ (Thompson and Hunston 1999: 4). It can also be realised lexically through adjectives, adverbs, verbs and nouns. Conceptually, evaluation can be realised by ‘identifying signals of comparison, subjectivity, and social value’ (Thompson & Hunston, 1999, pp. 13-14). The most obvious expression of subjectivity is the use of the first person pronoun (I, we). According to Hunston (1985; 1989; cited in Thompson & Hunston, 1999, p. 14) value-laden evaluation can be realised by considering ‘what is good’ and ‘what is bad’ from the point of view of goal-achievement: what is good ‘helps to achieve a goal’ while what is bad ‘prevents or hinders the achievement of a goal’. This, in Martin and White’s (2005, pp. 42-43) ‘appraisal’ terms, is ‘judgement’, that is, attitudes towards behaviour, which we admire/praise or criticise/condemn (see Sarfo, 2016, p. 83-85). Thus, by employing an evaluative language analysis, we are able to identify how government and opposition MPs each cast themselves into the role of the ‘other’ and establishing their own identities through opposition to and, frequently vilification of the other (Gabriel, 2012).

Legon Journal of the Humanities Vol. 30.2 (2019) Page 51 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) Analysis and discussion The analysis and discussion section is divided into two parts. The first part examines the content of the Parliamentary Committee report on the NACAP to bring to bear the various approaches engaged in for the fight against corruption. The second part, through the concept of evaluation, examines parliamentary debates in order to unearth the issue of othering among Ghanaian parliamentarians. The two parts are complementary: while the first part, which lays the background for the second, shows that there are legislative measures for the fight against corruption, the second part suggests that Ghanaian MPs’ attitudes betray the nature of their commitment to the fight against corruption. The fight against corruption in Ghana: the NACAP way The most comprehensive expression of the fight against corruption can be found in the National Anti-Corruption Action Plan (NACAP) 2012-2021 and the Report of the Committee on Constitutional, Legal and Parliamentary Affairs on the NACAP. The NACAP and the Committee cite some of the previous measures and policies employed in the fight against corruption as ‘moralising crusades’, ‘public execution of persons’, ‘passage of draconian decrees’, ‘confiscation of properties’ ‘passage of several anti-corruption laws’, ‘financial management reforms’, among others. Corruption was mostly criminalised in order to fight it, especially through the Criminal Code (GAN Integrity 2016, p. 1), with some effort put on moral and value-based crusades. As noted by GAN Integrity (2016, p. 1) the Ghanaian ‘government has a strong anti-corruption legal framework in place but faces challenges of enforcement’. Several anti-corruption legal frameworks have been put in place. These include the Criminal Offences Act, 1960 (Act 29); the Criminal and Other Offences (Procedure) Act, 1960 (Act 30); and the Criminal Procedure (Amendment) Act, 1965 (Act 261) among recent ones such as the Whistle Blowers Act; the Public Procurement Act; Financial Administrative Act; Article 286 (1) of the 1992 Constitution; and the Public Office Holders (Declaration of Assets and Disqualification) Act, 1998 (Act 550). However, all these have faced implementation challenges and have failed in the fight against corruption. In the view of the NACAP (and the Committee), these policies and measures have failed to fight corruption as expected, and corruption is still endemic in the country (Sarfo- Kantankah, 2018). The foregoing indicates that, in the fight against corruption, Ghana has struggled to build a culture of rule of law in the midst of building both a liberal and a socialist economy. A free-market driven economy is likely to breed corruption if

Page 52 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana the rule of law is dysfunctional, for ‘institutional reforms by themselves do not fare as well where corruption problems are worst’ (Johnston, 1998, p. 86). It has been argued that one reason for the failure of anti-corruption approaches is ‘the external imposition of contingent standards on societies that are not fully participating in those standards’ (Bukovansky, 2006, p. 184). Those standards usually conflict with traditional value systems of those societies, for ‘[w]hat is corrupt in one country may not be corrupt in another’ (Casas-Zamora, Walecki and Carlson, n.d). In Ghana, the inability to enforce laws that criminalise corruption may be culturally induced, for ‘[t]he mobilization of law reflects not only class differences but different norms and values as well’ (Silbey, 1989, p. 14). For example, the Ghanaian social structure respects familial and social ties, status, class and organisational positions, which have been said to influence law enforcement. Silbey (1989, pp. 13-14): has argued that: The legal system reflects and reproduces the encompassing social structure [sic]. Law is costly, and the costs are distributed differently according to social class, status, and organizational positions. Whether in the eighteenth or twentieth century, rates of grievance and litigation reproduce patterns of class, ethnic, and gender stratification. There are differential barriers not only to invoking law but also to complying with law…. Although the uses of law may be situationally structured, the responses and behaviors of legal actors cumulate, with the result that they come to reflect a wider array of social forces than the facts of specific incidents. For example, the ways in which law is mobilized and made available by the police is shaped by the community beyond the enforcement agents. This implies that the law may be a respecter of person(s), where the political and social elites may get away with corrupt practices. For example, in Ghana, appointment to political and public offices has mostly hinged on nepotism and cronyism (popularly known in Ghana as ‘whom you know’), ‘political affiliation and patronage’, rather than ‘meritocracy’ (Banaseh, 2017, p. 58). The implication is that the political will to arrest and prosecute corrupt officials may be hindered by the social positions of those officials. The foregoing suggests that Ghana’s ‘integrity system’ is questionable and reform is needed (Eigen, 1998, n.p). The NACAP states that it utilises a three-pronged approach to the fight against corruption – prevention, education, and investigation and enforcement. However, this paper interprets the NACAP approach in three different broad senses. In view of the failure of the old means of fighting corruption, the NACAP aims ‘to contextualize and mobilize efforts and resources of stakeholders, including Government, individuals, civil society, private sector and the media, to prevent and fight corruption through the promotion of high ethics and integrity and the vigorous enforcement of applicable laws’ (n.p.). NACAP’s mission implies a three-pronged

Legon Journal of the Humanities Vol. 30.2 (2019) Page 53 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) approach to fighting corruption: ethical/educational, multi-stakeholder and legal approaches. The ethical approach proposes a value-based system, involving the passage of a comprehensive code of conduct and rules of ethics to govern our actions and inactions and create a value based society. The assumption is that morality in Ghana has declined. The proposal appeals to the moral integrity of public officials and the general public, and it survives on personal conscience about right and wrong. The approach hints at the cultural interpretations of what constitutes corruption. It calls for civic education, since ‘[i]t is difficult if not impossible to sustain anti- corruption reforms without … effective public/civic education’ (Gyimah-Boadi, 2004, p.3). It means educating the people about what is right and wrong and what constitutes corruption, which ties in with the sensitisation and education of the public about the dangers of corruption. This will entail normative measures based on Ghanaian cultural values, for ‘[l]aw is a culturally variable phenomenon, and each society, culture, and subculture makes of it something that belongs to that grouping’ (Silbey, 1989, p. 17). As noted by Sparling (2016, p. 172) ‘[b]oth mores and laws are essential for the minimization of corruption’. Considering that the failed laws included embarking on moralising crusades with the help of religious and community leaders, one wonders how the new ethical approach contained in the NACAP can help fight corruption comprehensively. Perhaps, the multilateral approach comes to the rescue. The multi-stakeholder/multilateral approach sees the fight against corruption as a collective responsibility, as one MP puts it, ‘we [are] insisting on ownership of this process [of fighting corruption] by Ghanaians this time’ (Hansard 22 July 2015/ Col.3347). It advocates the involvement of all stakeholders through sensitisation of the general public on corruption; empowering the citizenry to report corruption; building public capacity to condemn, rather than celebrating and admiring ‘those who flaunt their ill-gotten wealth’ (Banaseh, 2017, p. 58). It also involves engaging the media and organisations; nationwide awareness creation and sensitization on corruption; and public education to improve understanding of issues relating to corruption. The multilateral approach is significant because, as the Committee notes, [t]he fight against corruption will bear little fruit without raising the awareness of public officials and the general public to the dangers of corruption and the duty of every citizen to combat corruption. Involving the public in the fight against corruption is important, for corruption renders the people less public-spirited (Sparling, 2016, p. 160), which in turn erodes public trust and threatens democracy. Engaging the citizenry is publicity, for ‘as secrecy is a defining component of corrupt practices, publicity is a defining component of anti-corruption campaigns’

Page 54 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana (Bukovansky, 2006, p. 186). Publicity is a means by which civil society groups ‘can exert pressure on government and the private sector for greater transparency and accountability’ (Eigen, 1998, n.p). A key tool for such publicity is the enactment and implementation of the right to information law2, which is more likely to empower citizens to partake in the fight against corruption. Encouraging publicity is to engage the media and the general public in the democratic process, which strengthens the democratic credentials of the country and the corruption fight. Media freedom and democracy are considered to be key in the fight against corruption (Bhattacharyya and Hodler, 2015). Democracy involves social empowerment, which is essential in the battle against corruption. As noted by Johnston (1998, p. 85): While institutional reforms enhancing transparency and accountability in the state and economic institutions are indispensable parts of any anti-corruption strategy, they also need a long-term social foundation, particularly where corruption is systemic. Social empowerment – expanding and protecting the range of political and economic resources, and alternatives, open to ordinary citizens – is one way to address the task. Social empowerment entails strengthening civil society in order to enhance its political and economic vitality, providing more orderly paths of access and rules of interaction between state and society, and balancing economic and political opportunities. While the publicity method appears plausible, we daresay that if the legal approach does not play its role of arresting and prosecuting corrupt officials, the exposure of corruption alone will not be enough, and vice-versa. For instance, what is the point and to what end, if exposed corrupt official go scot-free? As Manasseh Azure Awuni3 argues, what is important for journalists is knowing that the issues of corruption they expose are dealt with and that Ghana will not get anywhere in the fight against corruption when those exposed go without punishment. This is the more reason the legal approach is crucial in fighting corruption. The legal approach appears to be the most common (and important) among the means to combat corruption, as suggested by the MPs. It encompasses strengthening the country’s Anti-Corruption Legislative Framework through the passage of several Anti-Corruption Laws, conduct[ing] effective investigations and prosecution of corrupt conduct, the right of access to information law, resourcing and making use of anti-corruption bodies such as Commission on Human Rights and Administrative Justice (CHRAJ), the Ghana Police Service, Economic and 2 The Right to Information Bill, which was passed into law in March 2019, gives substance to Article 21 (1) (f) of the Constitution which states that ‘All persons shall have the right to information subject to such qualifications and laws as are necessary in a democratic society’. The preparation of the bill started in 1999, but it had not been passed since 2010 when it was presented to Parliament. The bill seeks to enable the citizenry to ensure accountable governance by holding governments accountable. 3 Manasseh Azure Awuni is a Ghanaian investigative journalist with the Multimedia Group Ltd. He is an anticorruption crusader. He won the 2018 West African Journalist of the year award. He made this statement when he delivered a speech after winning the award on October 14, 2018, Accra, Ghana. See the Ghana News Agency’s report at: http://www.ghananewsagency.org/social/ manasseh-azure-wins-2018-west-african-journalist-of-the-year-140040

Legon Journal of the Humanities Vol. 30.2 (2019) Page 55 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) Organised Crime Office, Audit Service. Even though MPs recognise the need to resource these anti-corruption institutions to enable them to fight against corruption, the bodies lack funds and resources to perform their functions. For instance, a former commissioner of CHRAG, Justice Emile Short, has bemoaned the lack of funds and resources to enable anti-corruption institutions to operate effectively (Banaseh, 2017). There is evidence to show that governments and MPs have often stated the need to resource these institutions. In the 2014 budget statement read to Parliament, the then finance minister stated that, as part of the government’s transparency and anti-corruption initiatives: additional budget provisions have been made to strengthen anti-corruption agencies, notably, CHRAJ. Also, as directed by H. E. the President, a percentage of proceeds of corruption when recovered, will be paid to anti-corruption agencies instrumental in exposing such acts, and this will further strengthen their resource base. Government’s Financial Policy, 2014 (para. 1026) The expression of commitment to resource anti-corruption bodies to fight corruption abounds in the data. Thus, lack of resources for the anti-corruption institutions indicates a disconnect between political rhetoric and practical action. As one MP notes, ‘[w]e can only succeed in the fight against corruption if we move from rhetoric to action’ (Hansard 26 November 2013/Col.1403). The former CHRAJ boss, Emile Short, partly attributes Ghana’s inability to fight corruption to the kind of multiparty democracy practised in Ghana, where political parties and politicians are funded by corrupt people who use all manner of means to recoup what they have lost. This means that the state’s inability to sponsor political parties is a recipe for corruption. However, the MPs appear to be silent on the nature of our multiparty democracy as a contributory factor for our failure to fight corruption. In the discussion of the NACAP on the floor of Parliament, MPs unanimously agree that the NACAP is an important step towards the fight against corruption; MPs seem to hold a common sense of purpose in the corruption fight. However, when it comes to debates on the issue of corruption and the fight against it, they appear to lack the single-mindedness needed to tackle corruption, as they engage in a blame game that leads to othering. Thus, we argue that the difficulty in fighting corruption partly emanates from what is called othering, a means by which MPs relegate the fight against corruption to the background and engage in a blame game, as discussed in the next section.

Page 56 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana Othering and the fight against corruption Othering is one of the commonest ways of constructing corruption in parliamentary debates. Often in the parliamentary debates, government and opposition MPs label each other as corrupt. Corruption is largely dependent upon the control of state resources and it is the ruling party/government which controls the resources. Thus, corruption seems to turn when the tables turn, that is, when governments lose power and new ones take over. For example, in a reaction to an NPP MP’s allegations of corruption against the NDC government, an NDC MP cautions ‘political office holders to be measured in terms of the words they use against one another. This is because the table only just goes round, and we find yourself [ourselves] on the other side’ (Hansard 26 November 2013/Col.1385). The extracts in Table 4 below, which come from debates on the floor of Parliament, illustrate othering by MPs. Instances of othering abound in the debates, but for want of space, we are considering only these. Extracts 1A and 2A come from opposition MPs, while 1B and 2B are from position/government MPs. Extract 1B is a response to 1A, whereas 2B is to 2A. In Example 1A, Mr Iddrisu accuses the President of paying lip service to the fight against corruption, and indirectly indicts the President for superintending over corruption, as in, the office of accountability operates in darkness. Mr Iddrisu achieves his purpose through the use of evaluative language, as in: i. I think that the President’s declaration… ii. But in my view it is not to be achieved with the self-serving Office of Accountability… iii. I think that if we are serious and the President means business… he should come with a legislation. The expressions I think and in my view indicate Mr Iddrisu’s subjective evaluation of the President’s commitment to the fight against corruption. He sees the President’s establishment of the Office of Accountability as self-serving, a negative evaluative adjective suggesting a ‘negation of identity’ of the President (Staszak, 2008, p. 44). That the Office operates in darkness demonstrates Mr Iddrisu’s impression that the Office is corrupt and that the President

Legon Journal of the Humanities Vol. 30.2 (2019) Page 57 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) Table 4: Extract samples illustrating othering by MPs

Extract 1A Extract 1B Mr. Haruna Iddrisu (NDC, opposition) 8 Mr. Maxwell Kofi Jumah (NPP, position) 9 February 2006/Col.566 February 2006/Col.665 Mr. Speaker, I think that the President’s … We have seen Ministers in other governments, declaration of ‘zero tolerance’ was in other regimes in this country. We have seen significant for our national fight against them. I counted the number of cars in a former the canker of corruption. But in my view Ministry of Education and there were six official it is not to be achieved with the self- cars–– six –– [Interruption] –– do not even talk serving Office of Accountability when about it. Mr. Speaker, and these people have the we do not know the cases they deal with, nerve to talk about corruption. which persons are affected, what issues are involved. Parliament is not informed … Mr. Speaker, the Auditor-General’s Report and, the Public is not informed. Mr. that is right here in this House, let us take one Speaker, what corruption hates is exposure; year, any year, let us probably toss a coin and therefore it loves darkness. The Office pick any year and look at it, and we would not of Accountability operates in darkness. need to talk about corruption again because Nobody seems to know what that office compared to them we are all angels in the New does. I think that if we are serious and Patriotic Party ( NPP). All of us, everybody the President means business in fighting in the NPP is an angel, compared to them. Of corruption, he should come with legislation course, we need to do more because this is a so that –– [Interruption.] Government that aspires to do even better, even though it is doing good. Extract 2A Extract 2B Mr I. K. Asiamah (NPP, opposition) 26 Mr Murtala M. Ibrahim (NDC, position) 26 November 2013/Col.1385 November 2013/Col.1405 Mr Speaker, the report further states that … Now, if there is any commitment that any there has been a financial loss to the State Government has ever made towards fighting to the tune of about GH¢150 million, that is corruption, between what the Government about ¢9.5 trillion [old cedis]. Mr Speaker, has done and a Government that says that it it goes on to state that there is a further would not prosecute corrupt people because ration debt to the tune of GH¢259 million. I risk collapsing my Government. Between a So, in total, we have about GH¢1.2 billion, Government which has taken the steps and said that is about GH¢12 trillion [old cedis] lost investigate, prosecute and the President who to the State. Mr Speaker, this is the highest says he will not investigate alleged corruption level of corruption that we should never in his Government because if he did that he entertain. Mr Speaker, they create, loot and risked collapsing his Government, she stands share. there questioning the commitment of that Government, which is fighting corruption. Is that the kind of nation we would want to build? I think that if there is any commitment that has ever been made, it is about this Government.

Page 58 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana has no commitment towards the fight against corruption. In this sense, Mr Iddrisu’s subjective focus is more on the President than on the issue of the fight against corruption. Such a position usually elicits counter-attacks. Thus, in response to the accusation of corruption against the President and his government, Mr Jumah (1B) counter-accuses the former NDC administration of massive corruption exemplified in the number of cars that were in the former Ministry of Education, questioning the moral authority of the members of the previous government to talk about corruption. He claims the NPP are comparatively better, saying compared to them we are all angels. This is a comparison of the NPP’s and NDC’s positions on the fight against corruption, leading to a value evaluation of the two (Thompson & Hunston, 1999). Again, consider Mr Jumah’s subjective evaluation of the performance of the previous NDC administration, as seen in such forms as we, I, these people and all of us in i-vi. i. We have seen Ministers in other governments, in other regimes in this country. ii. We have seen them. iii. I counted the number of cars in a former Ministry of Education… iv. … these people have the nerve to talk about corruption. v. … we are all angels in the New Patriotic Party ( NPP). vi. All of us, everybody in the NPP is an angel, compared to them. There is a construction of us and them, an in-group (we) and an out-group (these people) (Staszak, 2008), a creation of otherness (Barter-Godfrey & Taket, 2009). Similarly, in Extract 2A, Mr Asiamah accuses the NDC administration of being corrupt, that is, they create, loot and share, a value-laden evaluation (Thompson & Hunston, 1999) of the governance style of the NDC. In 2B, Mr Ibrahim defends his government as having shown a better commitment to fighting corruption and condemns the former President (NPP administration) for refusing to investigate and prosecute corrupt officials in order not to collapse his government. Consider Mr Ibrahim’s indirect comparison between the previous NPP government and the current NDC government: i. … a Government which has taken the steps and said investigate, prosecute (i.e. the Government, which is fighting corruption); and ii. … the President who says he will not investigate alleged corruption

Legon Journal of the Humanities Vol. 30.2 (2019) Page 59 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) in his Government because if he did that he risked collapsing his Government… Mr Ibrahim evaluatively portrays the NDC and its government as desirable, with a better commitment towards the fight against corruption, and the NPP and its previous government as undesirable, lacking the needed commitment (Fairclough, 2003; van Leeuwen, 2008). The extracts are, therefore, a ‘judgement’ (Martin & White, 2005, pp. 42-43) of the attitudes of the two political groups towards the fight against corruption in Ghana. Each group sees the other group as not having the requisite commitment to fighting corruption. These extracts can be said to be instances of othering, ways of distinguishing ‘us’, the honest; from ‘them’, the corrupt; ‘the normal from the deviant’ (Barter- Godfrey and Taket, 2009, p. 166), attributing attitudes, opinions and views to opponents (Gabriel, 2012). Corruption is both a moral and a legal issue. As Sparling (2016, pp. 157, 161) notes, ‘[c]orruption discourse is political morality’. Thus, to call someone corrupt is to question and attack one’s moral integrity and love for one’s country, a diminution of one’s virtue. It is a construction of a virtuous self as against a bad other, which is an identity formation through positive and negative othering. It is not surprising, therefore, that there are accusations and counter- accusations between position and opposition MPs. If corruption ‘entails an alteration of people’s primary desires such that they no longer serve to unite disparate individuals’ (Sparling, 2016, p. 166), the discourse of corruption in parliament divides parliamentarians in a more starkly uncompromising way. This is because ‘a primary concern of MPs [during debates] is to reinforce their own credibility (Ilie, 2003, p. 81), while they raise doubts about the trustworthiness of their opponents. As portrayed in the extracts above, the discussion of corruption in parliament suggests a trial of the political values and norms of political opponents. For example, they create, loot and share damages the reputation of the NDC government. MPs create a sense of ‘saints’ and ‘demons’ through ethical appeals that result in emotional outbursts from MPs whose honesty is questioned. For instance, Mr Osei-Prempeh (NPP – Nsuta-Kwamang/Beposo) questions the credibility of the President in fighting corruption when he says: The President comes and he talks against corruption, yet he appoints people who have been tainted with corruption into his Government, into top Cabinet positions … And you say that you are fighting corruption? (Hansard: 2 Mar 2010/Col.1447- 1448). The implication is that there is no unity of purpose required from MPs in fighting corruption. The stark division in the face of discussions of the issues of corruption

Page 60 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana is a relegation of the fight against corruption. By framing corruption through othering, rather than the substantive problem of corruption, brings into question parliamentarians’ commitment towards the fight against corruption. As an MP admonishes his fellow MPs: ‘when the issue of corruption crops up, we [MPs] should all work together as a unit to ensure that boldly and decisively, we tackle [it]’ (Hansard 2 March 2011/Col. 1928). Othering reduces the fight against corruption to a political blame game, for ‘[p]ublicly calling a person or government corrupt is a political act’ (Bukovansky, 2006, p. 185). It portrays the MPs as lacking the ‘political will’, that is, ‘the demonstrated credible intent of political actors … to attack perceived causes or effects of corruption at a systemic level’ (Kpundeh, 1998, n.p). It must be noted, however, that personality attacks are not necessarily bad in politics and political campaigns. The attacks and counter-attacks are a feature of negative campaign. According to Mayer (1996), negative campaigning and negative portrayals of political opponents are important for measuring the quality, trustworthiness and credibility of candidates who put themselves up for elections. All parliamentary (and political) debates can be justifiably considered as political campaigns, for MPs always have in mind the next election, when they make statements about the weaknesses and strengths of incumbent (or past) governments to either discredit or credit them. Every political statement has the potential to influence future elections. As stated by one NPP MP: ... in 2008, all of us, we are going to present ourselves to the electorate who voted us in here. … This is the time to separate the weed from the chaff; this is the time to ask ourselves, are we doing this on principle? Even if I am going to be voted against, I am going to vote for this because it is the right thing and not politics. Hansard 9 February 2006/Col.664 Calling opponent MPs corrupt and lacking the ability to fight corruption is to expose their weakness and dishonesty and their inability to govern the country, which implies that the electorate must consider voting them out of government. The ability to expose the weaknesses of political opponents offers the criticiser the opportunity to propose an alternative, for without knowing the weaknesses of the incumbent and the deficiencies of the current ways of fighting corruption will make it difficult for the opposition to offer alternative proposals. This is necessary because, from serious domains such as academia/academic research writings, through popular and everyday debates to politics, arguments and campaigns are organised in ways that first examine the weakness, defects and shortcomings of present research findings, arguments or policies before proposing new ones or solutions (Mayer, 1996). And ‘a candidate who is challenging an incumbent should

Legon Journal of the Humanities Vol. 30.2 (2019) Page 61 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) be required to show the weaknesses and shortcomings of his opponent, and then to indicate how and why he should do better’ (Mayer, 1996, p. 454). This accounts for the opposition MPs’ incessant portrayal of incumbent governments as corrupt. On the other hand, ‘[a]n incumbent should defend his own record, and then (since that record is unlikely to be entirely without blemish) should be able to point out the ways in which the challenger would be deficient’ (Mayer, 1996, p. 454). It is not surprising, therefore, that there are accusations and counter-accusations between government and opposition MPs. The argument we are making here is that even though the personality attacks and counter-attacks appear to drive attention away from the substantive issue of corruption, they are not necessarily detrimental. What is important is for us to consider whether the accusations and counter-accusations are based on truth and are relevant and provide a guide to understanding the personality and character of those accused of corruption. Conclusion The purpose of this paper has been to analyse, through a content analytical approach, the ways of fighting corruption in Ghana as proposed by the National Anti-Corruption Action Plan (NACAP), including MPs’ debates and Parliamentary Committee reports on the NACAP. Through the principle of evaluation, the paper also explored otherness in parliamentary debates on corruption in order to observe how otherness affected MPs’ attitudes and commitment towards the fight against corruption in Ghana. The aim was to assess the intentions, the psychological, attitudinal and behavioural responses of the MPs towards the fight against corruption. The paper notes that the NACAP proposes a three-pronged approach to the fight against corruption, namely: (i) ethical/educational approach, that is, the creation of a value-based society with strong moral principles that promote high ethics and integrity, and the sensitisation of the people on the dangers of corruption; (ii) multi-stakeholder approach, involving stakeholders such as the media, civil society and the general public to expose corruption; and (iii) legal approach, which involves the strengthening of anticorruption legislation, resourcing anticorruption agencies and the effective implementation and enforcement of anticorruption laws. The paper argues that the failure of these approaches in the fight against

Page 62 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana corruption results from Ghanaian cultural and social systems, where familial relationships, social ties, class, status and organisational positions have huge negative influences on the rule of law. In other words, Ghana has failed to build a culture of the rule of law. In the view of this paper, the main approach that can successfully combat corruption is commitment to the implementation and enforcement of, particularly, anti-corruption laws and, generally, the rule of law. The study has shown that one arena where othering (and otherness) manifests itself is parliament, where ideological tensions revolve around the social representations of the majority (government MPs) and the minority (opposition) MPs, who ‘are positioned in the place of a dichotomy that polarises [their] collective identity in relation to what is perceived as good or bad’ (Linares, 2016, p. 131). Othering allows for the objectification, differentiation and stereotyping of government and opposition MPs, ideologically constructing themselves as a binary us and them (van Dijk, 2011, p. 396). Through the concept of othering, Ghanaian MPs label each other as corrupt, making the discussion of the fight against corruption merely incidental to their debates and discussions. In other words, Ghanaian MPs’ debating strategies hinder the fight against corruption in Ghana. Instead of collectively agreeing that they have to fight corruption together and work towards that, the MPs are preoccupied with indicating which of them (the Majority and the Minority) are more corrupt than the other. Thus, the MPs do not necessarily deny that they are corrupt; they disagree on who is more or less corrupt, making them lose focus on the substantive issue of the fight against corruption. This brings into question Ghanaian parliamentarians’ commitment towards the fight against corruption. To this end, we do agree with the current Majority Leader and Minister for Parliamentary Affairs, Mr Osei Kyei-Mensah-Bonsu, that the fight against corruption must be taken away from the corridors of Parliament4. While it is well known that legislation and the law (Lawal, 2007; Cleveland, Favo, Frecka and Owens, 2009), strengthening state institutions (Börzela, Stahn and Pamuk, 2010), among others, are useful for fighting corruption, this paper has demonstrated that the attitudes of those who implement anti-corruption laws are essential for the fight against corruption. We argue that one of the challenges facing the enforcement of Ghana’s strong anti-corruption legal framework (GAN Integrity, 2016) emanates from lack of commitment on the part of Ghanaian MPs as a result of their othering orientations. The implication is that studies on corruption and the fight against it must turn attention to the attitudes and the commitment of those who implement the anticorruption laws. 4 Mr Kyei-Mensah-Bonsu made this statement in an interview with Nana Aba Anamoah’s State of Affairs, current affairs show on GHOne TV. The programme is held at 7pm every weekday.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 63 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) References Agbele, F. (2011). Political economy analysis of corruption in Ghana. European Research Centre for Anti-Corruption and State-Building (ERCAS) Working Paper No. 28. Amakali, J.M.L. (2018). Speech acts and their rhetorical purposes in the Namibian parliamentary discourse, 2015-2016. (PhD Thesis). The University of Namibia. Armah, E.N.A., Sarfo-Kantankah, K.S., Bauer, G. and Mensah, G. (2018). Report on the first session of seventh parliament 2017. A report conducted by Odekro, a non-profit parliamentary monitoring organisation, and sponsored by STAR-Ghana with funding from UKaid, DANIDA and the European Union. Baker, J.T.E. (2015). Defending the indefensible? the use of argumentation, legitimation, and othering in debates on refugees in the Canadian House of Commons, 2010-2012. (PhD Thesis). Memorial University of Newfoundland. Banaseh, M.A. (2017). Ghana losing fight against corruption – Emile Short. Daily Graphic, November 29, p. 58. Barter-Godfrey, S. & Taket, A. (2009). Othering, marginalisation and pathways to exclusion in health. In A. Taket, B.R. Crisp, A. Nevill, G. Lamaro, M. Graham & S. Barter- Godfrey (Eds.), Theorising social exclusion (pp.166- 172), Routledge, Abingdon, England. Bengtsson, M. 2016. How to plan and perform a qualitative study using content analysis. NursingPlus Open, 2, 8-14. Retrieved from http://dx.doi. org/10.1016/j.npls.2016.01.001 Bhattacharyya, S. & Hodler, R. (2015). Media freedom and democracy in the fight against corruption. European Journal of Political Economy, 39, 13-24. Börzel, T.A., Stahn, A & Pamuk, Y. (2010). The European Union and the fight against corruption in its near abroad: can it make a difference?, Global Crime, 11(2), 122-144. DOI: 10.1080/17440571003669142 Breen, M.J., Haynes, A. & Devereux, E. (2006). Fear, framing and foreigners: the othering of immigrants in the Irish print media. International Journal of Critical Psychology, 16, 100-121. Bukovansky, M. (2006). The hollowness of anti-corruption discourse. Review of International Political Economy, 13(2): 181-209.

Page 64 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana Casas-Zamora, K, Walecki, M., & Carlson, J. (no date). Political Integrity and Corruption – an international perspective. Retrieved from https://ifes.org/ sites/default/files/political_integrity_introduction_english_0.pdf Charteris-Black, J. (2006). Britain as a container: immigration metaphors in the 2005 election campaign. Discourse & Society, 17(5), 563-581. Cleveland, M., Favo, C., Frecka, T., & Owens, C. (2009). Trends in the international fight against bribery and corruption. Journal of Business Ethics, 90, 199-244. Retrieved from http://www.jstor.org/stable/40665294 Cooey, P.M. (1990). Emptiness, otherness, and identity: a feminist perspective. Journal of Feminist Studies in Religion, 6(2), 7-23. Downe-Wambolt, B. (1992). Content analysis: method, applications and issues. Health Care for Women International, 13, 313-321. Duck, J. M., Hogg, M. A., & Terry, D. J. (1995). Me, us and them: political identification and the third‐person effect in the 1993 Australian federal election. European Journal of Social Psychology, 25(2), 195-215. Eigen, P. (1998). Message from Transparency International. In Corruption and Integrity Improvement Initiatives in Developing Countries. New York: United Nations Development Programme. Fairclough, N. (2003). Analysing discourse: Textual analysis for social research. London:Routledge. Fraenkel, J.R., Wallen, N.E. and Hyun, H.H. (2012). How to design and evaluate research in education (8th ed.). New York: McGraw-Hill. Gadowska, K. (2010) National and international anti-corruption efforts: the case of Poland. Global Crime, 11(2), 178-209. DOI:10.1080/17440571003669191 GAN Integrity. (2016). Ghana Corruption Report. Accessed 26 July 2017, at: http:// www.business-anti-corruption.com/country-profiles/ghana Ghana Integrity Initiative. 2017. Press Release: The Corruption Perception Index 2016. Retrieved from https://www.tighana.org/assets/Uploads/DOC- 20170125-WA0000.pdf Göl, A. (2005). Imagining the Turkish nation through ‘othering’ Armenians. Nations and Nationalism, 11(1), 121-139. Gyimah-Boadi E. (2004). New anti-corruption governments: the challenge of delivery – Ghana: a case study. Prepared for the New Anti-Corruption Governments Meeting, Organised by the Government of Kenya, TI-Kenya and the TI Secretariat, Nairobi, October. Retrieved from http://www.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 65 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) worldbank.org/wbi/governance/ghana/pdf/gha_thechallengeofdelivery_ac asestudy.pdf. Hunston, S. (2011). Corpus approaches to evaluation: Phraseology and evaluative language. London: Routledge. Husserl, E. (1969). Formal and transcendental logic. The Hague, NL: Martinus Nijhoff. Ilie, C. (2003). Discourse and metadiscourse in parliamentary debates. Journal of Language and Politics, 2(1): 71-92. Jakaza, E. (2013). Appraisal and evaluation in Zimbabwean parliamentary discourse and representation in newspaper article. (PhD Thesis). Stellenbosch University. Jensen, S.Q. (2011). Othering, identity formation and agency. Qualitative Studies, 2(2), 63-78. Jibril, H.S. & Gyasi, Y.O. (2017). English expressions in Ghana’s parliament. International Journal of English Language and Linguistics Research, 5(3), 49-63. Johnson, J.L., Bottorff, J.L., Browne, A.J., Grewal, S., Hilton, B.A., & Clarke, H. (2004). Othering and being othered in the context of health care services. Health Communication, 16(2), 253-271. Johnston, M. (1998). Fighting systemic corruption: social foundations for institutional reform. The Journal of Development Research, 10(1), 85-104. https://doi.org/10.1080/09578819808426703 Kamenova, D. (2014). Media and othering: how media discourse on migrants reflects and affects society’s tolerance. Plitické Vedy, 17(2), 170-184. Kan-Dapaah, A. (2015). Parliament’s role in the fight against corruption. Accra: Institute of Economic Affairs (IEA). Kaplan, J. T., Freedman, J., & Iacoboni, M. (2007). Us versus them: political attitudes and party affiliation influence neural response to faces of presidential candidates. Neuropsychologia, 45(1), 55-64. Kinder, D. R. & Kam, C. D. (2010). Us against them: ethnocentric foundations of American opinion. University of Chicago Press. Kirkwood, S. (2017). The humanisation of refugees: a discourse analysis of UK parliamentary debates on the European refugee ‘crisis’. Journal of Community & Applied Social Psychology, 27(2), 115-125.

Page 66 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana Kpundeh, S.J. (1998). Political will in fighting corruption. In Corruption and Integrity Improvement Initiatives in Developing Countries. New York: United Nations Development Programme. Krippendorff, K. (2004). Content analysis: An introduction to its methodology (2nd ed.) Thousand Oaks, California: Sage Publications. Kyei-Mensah-Bonsu, O. (2013). A strong parliament is key to fighting corruption – Minority Leader. Retrieved from https://www.myjoyonline.com/ politics/2013/November- 13th/a-strong-parliament-is-key-to-fighting- corruption-minority-leader.php Lawal, G. (2007). Corruption and development in Africa: challenges for political and economic challenge. Humanity & Social Journal, 2(1): 01-07. Lazaridis, G. & Tsagkroni, V. (2015). Securitisation of migration and far right populist othering in Scandinavian countries. In G. Lazaridis & K. Wadia (Eds.), The Securitisation of Migration in the EU (207-236). London: Palgrave Macmillan. Leudar, I., Marsland, V. & Nekvapil, J. (2004). On membership categorization: ‘us’, ‘them’ and ‘doing violence’ in political discourse. Discourse & Society, 15(2-3), 243-266. Linares, S.M. (2016). Othering: Towards a critical cultural awareness in the language classroom. HOW, 23(1), 129–146. Lirola, M. M. (Ed.). (2014). Discourses on immigration in times of economic crisis: a critical perspective. Cambridge Scholars Publishing. Marquette, H. (2007). Civic education for combating corruption: lessons from Hong Kong and the US for donor-funded programmes in poor countries. Public Administration and Development, 27, 239-249. Martin, J.R. & White, P.R.R. (2005). The language of evaluation. Hampshire: Palgrave. Mayer, W.G. (1996). In defence of negative campaigning. Political Science Quarterly, 111(3): 437-455. Mchakulu, J.E.J. (2011). Framing political communication in an African context: a comparative analysis of post-election newspaper editorials and parliamentary speeches in Malawi. (PhD Thesis). University of Leicester. Montoya, C. & Agustin, L.R. (2013). The othering of domestic violence: the EU and cultural framings of violence against women. Social Politics, 20 (4), 534-557.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 67 Sarfo-Kantankah, K.S. & Arko, J./Legon Journal of the Humanities Vol. 30.2 (2019) Mwiinga, C. (2015). A critical discourse analysis of the parliamentary debate on the lifting of the immunity of the fourth republican president of Zambia in 2013. (MA Thesis). University of Zambia. Muller, J. Z. (2008). Us and them: the enduring power of ethnic nationalism. Foreign Affairs, 18-35. Parliamentary Centre. (2009). The Public Accounts Committee of the Parliament of Ghana. Canada/Ghana: Parliamentary Centre. Parliament of Ghana. (2018). Public Accounts Committee. Retrieved from https:// www.parliament.gh/committees?com=27 Riggins, S. H. (Ed.). (1997). The language and politics of exclusion: Others in discourse. Communication and Human Values, 24, Said, E. W. (1978). Orientalism: western representations of the orient. London, UK: Routledge. Sarfo, E. (2016). Questioning and debating in UK and Ghanaian parliamentary discourse. (PhD Thesis). University of Leeds. Sarfo-Kantankah, K.S. (2018). Corruption is a big issue: A corpus-assisted study of the discursive construction of corruption in Ghanaian parliamentary discourse. Legon Journal of the Humanities, 29(1), 226-258. Silbey, S.S. (1989). A sociological interpretation of the relationship between law and society. In Richard J. Neuhaus, (Ed.), Law and the ordering of our life together. (1-27) Grand Rapids, Michigan: William B. Eerdmans Publishing Company. Sosoni, V. (2017). The rhetoric of othering in the Greek parliament: representation of the Troika and the self/other dichotomy. Lodz Papers in Pragmatics, 11(2), 271-286. DOI: https://doi.org/10.1515/lpp-2015-0014 Sparling, R. (2016). Motesquieu on corruption: Civic purity in a post-republican world. In Geoffrey C. Kellow and Neven Leddy, (Eds.), On Civic Republicanism (157-184) Toronto: University of Toronto Press,. Retrieved from: http://www.jstor.org/ stable/10.3138/j.ctt1kk65xt.12 Spivak, G. C. (1985). Can the subaltern speak? Wedge, 7(8), 120-130. Staszak, J.-F. (2008). Other/otherness. In R. Kitchin & N. Thrift (Eds.), International encyclopedia of human geography, 43-47.

Page 68 Legon Journal of the Humanities Vol. 30.2 (2019) Sarfo-Kantankah, K.S. & Arko, J./Compared to them, we are angels: Parliament, othering and the fight against corruption in Ghana Stemler, S. (2001). An overview of content analysis. Practical Assessment, Research & Evaluation, 7(17). Retrieved from http://PAREonline.net/getvn. asp?v=7&n=17 Thompson, G. and Hunston S. (1999). Evaluation: An introduction. In: Hunston, S and Thompson, G. (Eds.) Evaluation in text: Authorial stance and the construction of discourse. Oxford: Oxford University Press, pp.1-27. UNODC (2003). The United Nations convention against corruption. Retrieved from http://www.unodc.org/documents/treaties/UNCAC/ Publications/ Convention/08- 50026_E.pdf UNODC (2018a). Strengthening global integrity and ethics: UNODC finalizes series of new university modules under the education for justice initiative. Retrieved from http://www.unodc.org/dohadeclaration/en/news/2018/05/ strengthening-global- integrity-and-ethics_-unodc-finalizes-series-of-new- university-modules-under-the- education-for-justice-initiative.html UNODC (2018b). UNODC’s action against corruption and economic crime. Retrieved from http://www.unodc.org/unodc/en/corruption/thematic-areas. html Van Der Valk, I. (2003). Right-Wing parliamentary discourse on immigration in France. Discourse Society, 14, 309-348. Van Dijk, T.A. 1997. Political discourse and racism: describing others in Western parliaments. In S.H. Riggins (Ed.), The language and politics of exclusion (pp. 31-64). Thousand Oaks, CA: Sage Publications, 1997. Van Dijk, T.A. (2000a). On the analysis of parliamentary debates on immigration. In M. Reisigl & R. Wodak (Eds.), The semiotics of racism: approaches to critical discourse analysis (pp. 85-103). Vienna: Passagen Verlag. Van Dijk, T. A. (2000b). Ideologies, racism, discourse: Debates on immigration and ethnic issues. Comparative Perspectives on Racism, 91-116. Van Dijk, T.A. (2011). Discourse and ideology. In T.A. van Dijk (Ed.), Discourse studies: a multidisciplinary introduction (pp. 379-407). London: Sage. Van Houtum, H. & van Naerssen, T. (2001). Bordering, ordering and othering. Tijdschrift voor Economische en Sociale Georgafie, 93(2), 125-136. Van Leeuwen, T. (2008). Discourse and practice: New tools for critical discourse analysis. Oxford: Oxford University Press. Weber, R.P. 1990. Basic content analysis (2nd ed.). London: Sage Publications.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 69

Bamgbose, G.A. & Ajayi, T.M./Legon Journal of the Humanities Vol. 30.2 (2019) DOI: https://dx.doi.org/10.4314/ljh.v30i2.4

Pub Interactions in Southwestern Nigeria Ganiu Abisoye Bamgbose Lecturer Department of Mass Media and Writing, Pan-Atlantic University, Lagos, Nigeria University of Ibadan, Ibadan, Nigeria Email: [email protected] Temitope Michael Ajayi Lecturer Department of Linguistics and African Languages University of Ibadan, Ibadan, Nigeria Email: [email protected]

Submitted: December 18, 2018/ Accepted: July 14, 2019/ Published: December 30, 2019

Abstract Studies in the area of discourse analysis in Nigeria have explored so many aspects of human life. However, existing studies have glossed over the nature of discussions in pubs. Sourcing data from selected pubs in three southwestern states of Nigeria, this study investigates the nature of interactions in pubs, using register and common ground theoretical framework. Analysing a total of nine (9) excerpts, the study shows that women/ladies, football and politics are three common topics in the discourse in southwestern pubs. The means of conversing is always spoken and this makes for the spontaneous flow of the discussions. Discussions are usually casual despite the fact that affective involvement is usually high. The affective involvement is usually due to the passion people have for the topics identified. Power relation is usually equal and, in few cases where it is not, common ground’s coordinating devices like convention, precedence and manifest are utilised to sustain cordiality and avoid domination or intimidation. The study concludes that language use in pubs has its peculiarity and uniqueness and therefore deserves linguistic enquiry.

Keywords: pubs, register, common ground, coordinating devices, southwestern Nigeria

Introduction Pubs in Nigeria are called beer parlours. Like in other parts of the world, they are relaxation centres where people visit to drink and meet with their friends. Most pubs serve mainly alcoholic drinks. In Nigeria, one can safely say there are two major dispositions towards pubs: those who see them as a good place to hang

Legon Journal of the Humanities Vol. 30.2 (2019) Page 71 Bamgbose, G.A. & Ajayi, T.M./Legon Journal of the Humanities Vol. 30.2 (2019) out and for relaxation and refreshments, and those who on the grounds of religion or morality frown on these places. To this latter group, these places are not meant for people who have dignity. In fact, this notion was once experienced by one of the authors of this paper. As a ‘born again’ Christian, he1 visited two friends that he had not seen in a long time. In the course of their interaction, one of them offered to entertain them (the trio) and then suggested that they visited a nearby pub for drinks and pepper soup. He was a bit reluctant because he felt such a place was not meant for decent fellows let alone a born again Christian. On the contrary, a personal encounter with people in this former group has shown that, beyond relaxation and refreshments, testimonies of great experiences, help and connections are formed from the interactions that take place in these public places. While it is not the focus of this paper to take a position as to whether it is decent to visit pubs or not, it is considered expedient to investigate language use in this domain. This is with the aim of understanding the nature of the discourse in this domain of language use, a phenomenon that has been overtly glossed over in scholarship, especially in linguistic studies. Even the few studies on pubs have left the discourse of pubs unexamined. Situating the current study The very little attention paid to pub-related scholarship, especially in Nigeria, explains why there is scanty literature on the phenomenon. Some of the extant studies in this regard include Dumbili (2013), Ojedokun and Ademuson (2014), Bada (2014), Dumbili (2015), Lamidi (2016), Dumbili and Henderson (2017), Dumbili and Williams (2017), and Ayodele et al (2018). Dumbili (2013) attempts an exploration of the factors responsible for the changing pattern of alcohol consumption in Nigeria, particularly as it relates to the ‘new trend of excessive alcohol consumption habit’ noticed among Nigerians. He observes that there is a fundamental shift in the drinking orientation of many Nigerians, as the complex socio-cultural role alcohol played in the primordial African (and Nigerian) traditional society has taken a new dimension. He notes that disintegration of traditional values, non regulation, advertising, among others are key factors responsible for the changing patterns of alcohol consumption among Nigerians. Ojedokun and Ademuson (2014) is a sociological exploration of the motivation for engaging in beer parlour business by married women as well as possible challenges confronting them. These scholars observe that, while the need 1 One of the researchers

Page 72 Legon Journal of the Humanities Vol. 30.2 (2019) Bamgbose, G.A. & Ajayi, T.M./Pub Interactions in Southwestern Nigeria for self actualisation and empowerment is a major reason for married women’s engagement in beer parlour business in Nigeria, negative perception by some members of the public is a serious challenge faced by these women. Similarly, studies such as Dumbili and Henderson (2017), Dumbili and Williams (2017), Dumbili (2015) and Bada and Adebiyi (2014) have largely focused on the gendered alcohol consumption behaviour of Nigerians, with particular emphasis on the drinking behaviour of women in Nigeria, and young adults’ drinking habit in Nigeria. Largely, these studies emphasise the fact that gender plays a significant role in the alcohol consumption behaviour of Nigerians. Lamidi (2016) examines how humour is pragmatically deployed in the ‘beer parlour politics’ section of the Nairaland2 forum with particular focus on the linguistic devices employed by participants to instantiate humour on the platform. Ayodele et al. (2018), from a socio-psychological perspective, opine that operation of pubs and beer parlours is a strong factor that accounts for high rate of crime and criminality in Lagos State, Nigeria, for instance. However, as pointed out earlier, none of these aforementioned studies has attempted an exploration of the nature of pub discourse. This paper, thus, is an essential intervention in this regard. Theoretical Orientation This study adopts, as its theoretical framework, register and common ground theories. While register caters for the semantic aspect of the data following the submission of Mathiessen (1993) that register is primarily a semantic concept, common ground theory caters for the contextual clues that are found in the data. Odebunmi (2007) sees register as “a broad concept that covers all forms and kinds of communication in terms of the linguistic forms used, the activities performed, the participants and their roles in the communication, the medium of communication, and the interconnectivity between one linguistic form and another”. The term register can be discussed from two perspectives: broad and narrow labels. From the broad perspective, register means the words that are used in a particular field. In this sense, register will mean jargon. From the narrow perspective, we are looking at the use of register from a linguistic angle; especially among functional linguists. Register in this specific regard is “a sort of social genre of linguistic usage” (Stockwell, 2002, p. 7). Register here is sometimes described as a sociolect, for example, the language of a newspaper article. It is in this narrow sense of register as a sociolect (language use in the society) that this study aims at exploring the language of pubs in southwestern Nigeria.

2 A popular online socio-political platform on which Nigerians, particularly youths discuss socio-political issues.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 73 Bamgbose, G.A. & Ajayi, T.M./Legon Journal of the Humanities Vol. 30.2 (2019) Halliday (1978) deals extensively with the three dimensions of register: field, mode and tenor. Field, according to him, refers to the “on-going activity and the particular purpose that the use of language is serving within the context of that activity” (Halliday, 1978, p. 62). The words employed depict the activity that is being carried out. It is concerned with the choice of words employed; whether it is technical or not. The field of discourse includes the topic and the interactants. Eggins (1994, p. 52) considers mode as “the role language is playing in the interaction”. Mode concerns “what it is that the participants (of a transaction) are expecting language to do for them in that situation” (Hasan and Halliday,1985, p. 12). It is also concerned with the medium of communication, whether spoken or written. Halliday (1978) views mode as covering Dell Hyme’s channel, key and genre (Odebunmi 2006). Channel has to do with the medium through which the message intended to be disseminated is passed across. It can be through signal, smoke, writing, speech, etc. The key is concerned with the tone and manner with which something is uttered or written, that is, whether the message is persuasive or dissuasive. Genre refers to the type of event that is taking place, such as lecture, poem, letter, sermon, etc. Tenor has to do with who are taking part in the transaction as well as the nature of the participants, their status and roles (Hasan and Halliday, 1985, p. 12). It deals with the role and status relationships between interlocutors in a situation. The tenor of discourse can be examined from the perspectives of three continua as identified by Polyton (1985) which are: power, affective involvement, and contact (Odebunmi 2006). Power deals with the distribution of influence between the interactants. It can be symmetrical or asymmetrical. Affective involvement is concerned with whether the interactants have high or low emotional involvement and committed in a situation. Contact refers to whether the role being played by the interactants causes them to have frequent contacts or not (Eggins 1994). These three aspects of register are used to establish how meaning is generated and how relationships are sustained in pub discourse. Enfield (2008) defines “common ground” as a resource that speakers exploit in inviting and deriving pragmatic inference, as a way to cut costs of speech production by leaving much to be inferred by the listener. The common ground theory of communication is a concept proposed by Clark and Brennan (1991). The theory refers to “mutual knowledge, mutual beliefs, and mutual assumptions” that are believed to be essential for successful communication between people. Central to the theory of common ground is its emphasis on the context of language. Clark finds, for example, that all people take as common ground aspects of human nature such as physical senses, communal lexicons (i.e., sets of word conventions

Page 74 Legon Journal of the Humanities Vol. 30.2 (2019) Bamgbose, G.A. & Ajayi, T.M./Pub Interactions in Southwestern Nigeria in individual communities), and cultural facts, norms, and procedures. Clark (1996) argues that face-to-face communication, rather than written language, should be the basis of a theory of language. Common ground, like all other theories, has its tenets, ranging from grounding, to cost of grounding, constraints on grounding, coordinating devices, etc. Relevant to this present study is coordinating devices. Coordinating devices are rationale for mutual expectations that make partners believe that they will converge on the same joint action (Clark 1996). The relationship between coordination and the common ground is two-fold. On the one hand, the process of coordination can be described as updating this common ground through the addition of new information (in the form of coordination devices). On the other hand, coordination devices are nothing more than a shared basis meant to enable the coordination of participatory action that would eventually contribute to joint actions. The coordination devices are explicit agreement, the conventions, precedent and the manifest. They are briefly discussed below: • Explicit agreements are occurrences of dialogues in which parties explicitly communicate their own intentions. If two persons, for instance, want to meet at a certain place, they often solve that problem by talking to each other and agreeing on the place and the time. • Conventions range from rules and regulations to less formal codes of appropriate conduct. Conventions can also be more “local” and less bound to the community; for instance, three persons who want to meet for lunch every Tuesday (a joint action) can agree on meeting at a certain place at 12:15. They then each go there at that time every Tuesday (their participatory action). In this case, going to the meeting place is not a habit but a convention that has been set by participants of the joint activity. According to Clark (1996), this is because “that is what they mutually expect each other to do, based on the regularity in their recent behaviour.” As we see in this example, an explicit agreement can evolve into a convention if it is established as the agreed solution to a recurrent problem. For the purpose of this study, we shall be accounting for elements of Yoruba cultural convention in the data since the study is situated in southwestern Nigeria, a region occupied mainly by the Yoruba. • Precedent applies to norms and expectations developed within the on-going experience of the joint activity. If we remember that our friend Kunle was in a place on Friday afternoon of a previous

Legon Journal of the Humanities Vol. 30.2 (2019) Page 75 Bamgbose, G.A. & Ajayi, T.M./Legon Journal of the Humanities Vol. 30.2 (2019) week, we may anticipate that he will be there this week. Unlike the convention of meeting every Tuesday at 12:15, precedents do not rest on mutual agreement about an action, they are just anticipations based on the previous experiences of participants. • Manifest is also called “perceptual salience”: it captures situations in which the environment (or the available information) makes the next move obvious within the many actions that could clearly be chosen. During a surgery operation, for example, pointing at a certain element of anatomy can make it obvious to all those involved what to do next. Methodology and Data Discussion Following the arguments of Bernard (1994) and deMunck and Sobo (1998), this study employed an ethnographic method of data collection, particularly participant and non-participant observation techniques. This method has been described as the most reliable means of collecting natural sociolinguistic data that reflect the true language behaviour/practice of research participants. The data for this study comprised ten interactions among pub customers in ten randomly sampled pubs in three southwestern Nigerian states: Lagos, Ogun and Oyo between 2015 and 2017, three of which involved the researchers as pub customers. The interactions were observed and detailed in a field note. During data elicitation, particular attention was paid to the linguistic and paralinguistic clues deployed by the participants. Participants in the data are labelled as speakers: speaker 1, speaker 2 and so on. The analysis followed a pattern of discussing the three components of register as found in each excerpt (interaction) by carefully identifying and engaging the linguistic and nonlinguistic elements in each excerpt and then further discussed the contextual factors in the excerpts in line with the analytical tools of common ground. The discursive interactions were transcribed and presented in the forms they were uttered with the glosses presented in parentheses after each excerpt in order to avoid the loss of the relevant contexts that are embedded in the participants’ mode of communication. Data Analysis This section focuses on the analysis of the observed interactions. The dominant language in the conversations/interactions is Pidgin English. This is because Pidgin English is the lingua franca that unifies the Nigerian heterogeneous community especially in informal contexts such as pubs. A background is given for each of the excerpts for a clearer understanding of the analysis that follows.

Page 76 Legon Journal of the Humanities Vol. 30.2 (2019) Bamgbose, G.A. & Ajayi, T.M./Pub Interactions in Southwestern Nigeria Excerpt 1 The excerpt below centres on the controversy that ensued between two football enthusiasts in a pub over the performance of a player, Benzema.

Speaker 1: This Benzema na useless player. I no know why dem never sell am troway (Benzema is a useless player. I don’t know why he hasn’t been sold out) Speaker 2: Sell who? Sell Benzema? Guy you know ball at all? Nawa for you o (Sell who? Benzema? Are you sure you know about football?) Speaker 1: Guy abeg I no follow you talk o. I no want insult because we dey look ball together o, abeg (I wasn’t talking to you pls. I wouldn’t tolerate insult because we’re watching match together) Speaker 2: oya no vex (ok I’m sorry) The field of discourse in this excerpt is football. This subject matter is made clear through certain vocabulary and lexical collocation as deployed by the participants. Common nouns like player and football and the proper noun Benzema (a French footballer who plies his trade in Real Madrid) make the topic of discourse clear. Collocations like useless players, sell out, know football make the argument of the discourse clear. At the level of mode, the excerpt is spoken and is therefore marked by spontaneity and immediacy. This makes the exchange succinct as the speakers are both physically present at the setting. For the tenor, the discourse is a fan-to-fan relationship. Power relation is both symmetrical and asymmetrical, as each fan can freely express his or her opinion about a fan or a team but also resorts to cultural expectations when confrontations arise, especially between persons with clear age differences. Affective involvement can be said to be high as each fan tends to be keen on his view about the player being discussed. Convention, as one of the coordination devices of common ground, seems to resolve this equal power of all fans watching football together as one of the fans quickly resorts to the African or precisely Yoruba cultural convention where age differences must be considered, irrespective of closeness or intimacy. Speaker 2, sharing the same (Yoruba) socio- ideological notion of showing deference to superior interlocutors in interpersonal interactions, immediately adheres to this cultural convention by tendering apology, knowing he had earlier flouted the norm.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 77 Bamgbose, G.A. & Ajayi, T.M./Legon Journal of the Humanities Vol. 30.2 (2019) Excerpt 2 The interaction in the excerpt below involves people who do not have previous relationship or intimacy and the topic centres on the Nigerian politics immediately after the 2015 presidential election which brought Muhammodu Buhari into power as the president of Nigeria. Speaker 1: Buhari, Buhari, Buhari, the Buhari has come now. There is no money to eat. Even in Jonathan’s time, things were not this hard Speaker 2: My brother let me tell you something. If a country will develop, one thing we must do is to control the economy. See I was not so young during IBB’s era, probably you were (voices were heard laughing) but seriously that’s not the point. The point is there was money in circulation during IBB’s era but the country was almost ruined economically. So I’ll advise we just give Buhari sometime. Speaker 1: Ok sir o Speaker 3: Hunger should sha not kill us (All laughed out) The field of this excerpt is politics. The lexical choices show thatthe discussion centres on the Nigerian political situation. Proper and common nouns like Buhari, Jonathan, IBB, country, economy all point to this field of discourse. Other phrases, syntagmatic relations and collocations such as: control the economy, money in circulation, ruined economically, point to a more specific aspect of the polity being discussed, which is the economic state of the country. The mode is spoken and it makes for a free flow discourse. The tenor is one between formal and informal. This is because although the setting is informal, the topic of discourse is one that is serious as it affects the life of all the discourse participants. Power in this discussion can be said to be unequal as there is a deliberate attempt by speaker 2 to show this age difference. This is achieved through the coordinating device of precedence with which speaker 2 makes reference to a past administration humorously with the belief that speaker 1 knows little or nothing about this past administration. However, this unequal power relationship is meted out through the use of ‘my brother’ by speaker 1 which serves as a cultural attempt to create familiarity with speaker 2. It is most likely the case that speaker 2 gains the attention of speaker 1 through this cultural affiliation of siblinghood. Excerpt 3 The excerpt is about a casual discussion on the way women dress and its psychological effect on men. The topic ensued at the instance of a lady who passed

Page 78 Legon Journal of the Humanities Vol. 30.2 (2019) Bamgbose, G.A. & Ajayi, T.M./Pub Interactions in Southwestern Nigeria in front of three friends who were having a chat. The three participants in the discourse jokingly contributed their turns to the topic. Speaker 1: Chimooooo, these babes no go kill person with these leggings wey dem dey wear (Oh God, these ladies won’t kill me with these leggings they wear) Speaker 2: My brother, I think say na only me the thing dey disturb o. (My brother, I thought I alone get affected by the sight of it) Speaker 3: My own be say the moment I see any woman yansh inside da tin, chai, even if I dey write exam, I must fail da paper ( For me, once I see a lady’s buttocks inside it, Oh, even if I were writing an exam, I sure will fail that paper) The field of this excerpt is chitchat about ladies. A lady who was wearing a pair of leggings passed by three young men who were involved in a conversation and she caught their attention. The first speaker exclaimed at the sight of the lady’s buttocks and others cued into the gist. Lexical items like babes, leggings, women point at the focus of the discourse. The mode is spoken and interpersonal as the participants come into the discussion spontaneously. The tenor is casual and power relation was equal among the participants because they were friends and contemporaries and the three contributed freely and spontaneously to the discussion. The coordinating devices of manifest and convention were employed in the excerpt as seen in the deployment of certain words. The exclamatory discourse marker Chimooo which literally means God in Ibo (a language spoken by Eastern Nigerians) and typically used to express surprise is used by speaker 1 to make manifest a topic which immediately caught the attention of his other friends and an interaction ensued immediately. Convention surfaced in the use of brother by speaker 2 as a discursive strategy for creating rapport or chitchat out of speaker 1’s comment. Excerpt 4 The excerpt centres on known political commentators in Nigeria, Femi Fani- Kayode, who was also a minister in the Jonathan led administration which preceded the Buhari led administration and Nnamdi Kanu, leader of a group of agitators for the secession of the Ibo people of Nigeria. Speakers 1 and 2 in the excerpt know each other while speaker 3 has separately come to the pub. Speaker 1: Kunle, have you seen today’s paper? Speaker 2: No. Any catchy news? Speaker 1: This Fani Kayode of a guy will not shut his mouth until he lands in jail.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 79 Bamgbose, G.A. & Ajayi, T.M./Legon Journal of the Humanities Vol. 30.2 (2019) Speaker 3: Waoh! What a coincidence. Exactly what I was saying to my wife this morning when I was listening to news. If not that he’s even a bastard Yoruba child, should someone of his age even talk to an elderly man like Buhari that way? Speakers 1&2: Abi o Speaker 1: Anyway, he will join his noisy friend Kanu soon. The names mentioned in the discourse show it is a discussion on Nigerian politics. Femi Fani-Kayode, Buhari and Nnamdi Kanu are all contemporary Nigerian political actors. Apparently, the discussion centres on Femi Fani-Kayode who was a key player in Nigerian politics during President Goodluck Jonathan’s administration. The mode is spoken and it is marked by spatial/interpersonal distance, especially when one understands that speakers 1 and 2 have a relationship before the discourse and speaker 3 is a different person who is not known to speakers 1 and 2 prior to the interaction. The tenor can also be said to be casual as the participants do not seem to be so connected to the topic of the discourse. On this ground, affective involvement can be said to be low. Convention helps to make sense of speaker 3’s reference to age as the Yoruba culture frowns at a younger person confronting an elderly person as done by Fani Kayode, in the news being discussed by these men. It is a precedent among the discussants that Fani Kayode was at the time of this discourse a chieftain in the then ruling political party in Nigeria, Peoples Democratic Party (PDP), and he had on several occasions castigated Muhammodu Buhari, who was at that time the presidential flag bearer of the opposition party as a form of propaganda and loyalty to his party, All Progressives Congress (APC). Fani Kayode was, however, seen to have flouted the Yoruba cultural convention which demands giving respect to elders irrespective of political allegiance. Precedence also helps to understand the reference to Kanu who, at the time, had been jailed by the Federal Government of Nigeria for his anti-government activities which border on the secession of the eastern part of the country under the umbrella of Republic of Biafra. The reference to Kanu clearly shows that all of the speakers know about his imprisonment. Excerpt 5 The interaction in this excerpt takes place between two football fans during a match. Evidently one of the speakers is a supporter of Chelsea Football Club while the other is not. Speaker 1: For Heaven’s sake, what’s the problem with Mourinho, shee e no fit comot Terry ni? Abi which kind wahala be this now

Page 80 Legon Journal of the Humanities Vol. 30.2 (2019) Bamgbose, G.A. & Ajayi, T.M./Pub Interactions in Southwestern Nigeria (For Heaven’s sake what’s the problem with Mourinho, can’t he substitute Terry? Or what kind of problem is this?) Speaker 2: Una never see anything. Make una do as una dey do na. Una kuku get money (You guys haven’t seen anything yet. You should better do as usual. Good you have money Speaker 1: Before nko, if he manage finish this season, make e just pass field dey go Portugal after last match (Definitely, if he survives as a coach for the rest of this season, he should just travel to Portugal from Stadium after the last match.) The field of discourse is football as indicated by words like Mourinho (a former coach of Chelsea Football Club who now coaches Manchester United), Terry, match and stadium. The interaction shows that a football match is going on and perhaps a particular player is not playing so well as expected. The mode is oral and it aids the spontaneity of the discourse. The tenor is a fan versus fan gist with equal power and a high affective involvement as seen in the utterance of speaker 1 who passionately condemns Mourinho’s decision not to substitute one of the players, Terry. The high affective involvement is seen in the use of the word problem and its Yoruba equivalent, wahala which indicates the writer’s interest in the team and his fear that the team might lose the match if the coach does not make a substitution The common ground shared by the two speakers makes it easy for speaker 1 to know that the usual thing speaker 2 is referring to is Chelsea’s regular habit of sacking coaches. This common ground also helps the interpretation of Portugal as speaker 1 knows and also knows that the other speaker knows that Mourinho’s hometown is Portugal. Excerpt 6 The excerpt is an exchange between two lovers of beers who try to portray their choices as the best kind of beer in town. Speaker 1: Oh boy nothing like Shine Shine bobo (Oh boy, there’s nothing like shine shine 3bobo) Speaker 2: Dey deceive yourself, na watin go let you dey piss up and down you like abi? Drink stout make you know say you be man. (You’re deceiving yourself. So you like what will make you urinate frequently? Drink stout and feel like a man) Speaker1: Oya no vex na. Make everybody answer him papa surname (both men laughed) Please don’t be angry, let everybody bear their surnames. 3 The expression is often used to refer to Star Lager Beer in the Nigerian context.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 81 Bamgbose, G.A. & Ajayi, T.M./Legon Journal of the Humanities Vol. 30.2 (2019) The discussion reveals the field as being one that centres on alcohol; with names like shine shine bobo (Star Beer), stout, beer and drink. The argument revolves around beer preference. This is seen in the speakers’ turns as they try to justify their brand with the former asserting that there is no beer like stout and the latter nullifying the claim on the ground that the former’s choice causes constant urination. The mode is oral and highly interpersonal; this makes the discussion lively. The tenor is that of a beer lover versus another beer lover who in a jovial and highly spirited manner show support for their preferred brands of beer. There is an equal power relation as both end up happily sticking to their preferred brands of beer. Common ground makes for a successful communication as Star beer is referred to as shine shine bobo, a conventional name given to the beer in the Nigerian context. Excerpt 7 The excerpt is a gossip between two friends on ladies and their love for fun. Speaker 1: But God know say all these girls sabi like flexing. See babe wey been tell me say she dey go church this afternoon don hear music now tell me say she dey come meet me now now (God knows girls so like outings. A babe that had told me she was going to church heard music and immediately said she is coming over here to meet me.) Speaker 2: Naso their life be o. Dem no dey ever carry am (That’s how their life is. They don’t ever miss out of fun.) Speaker 1: Carry watin? Na 25 unit course, ahahahahahahahahah (Carry what? It’s a 25 unit course) (Both men laughed) Speaker 2: To carry am na extra year. Again the discourse field here is women or ladies. That is evident with the use of words like girls, babe and she. The talk is about the disposition of ladies towards outings. The interactants delightfully shared their belief that ladies do not, for any reason, miss out of any opportunity of a convivial gathering. The mode is spoken and has high interpersonal flow. The tenor shows a jovial mood and power relation is equal as both parties express their opinions about women with a level of affective involvement. The common ground of the university grading system helped speaker 1 to make sense of speaker 2’s expression: they don’t ever carry it. This common ground was built on when speaker 1 made an amusing reaction to it and both men laughed. Speaker 2 builds on the common ground of the university grading system by comparing a lady’s absence from a social conviviality to having to repeat a session. This unpalatable experience in the life of any student is likened to a lady’s emotional state when she misses the opportunity of a social outing.

Page 82 Legon Journal of the Humanities Vol. 30.2 (2019) Bamgbose, G.A. & Ajayi, T.M./Pub Interactions in Southwestern Nigeria Excerpt 8 The excerpt below is an exchange between three young undergraduate students whose institutions are on nationwide strike. They are seen talking about the strike action during their drinking leisure. Speaker 1: Me I’m tired o. what is this gist about another ASUU strike now. When person wan graduate for this life? (I am tired. When will one get to graduate?) Speaker 2: Oh boy me no even dey reason that one again. We sha go graduate one day (I have stopped thinking about it. I know we’ll graduate someday) Speaker 3: My children, so it was in our time too. I spent six years on a four-year course back then in Ife. Let’s just pray God help this country. The field of discourse here is education as evident in lexical items likeASUU (Academic Staff Union of University), strike, graduate and course. The topic centres on strike in Nigerian universities. The mode is spoken and can be said to be spatial/interpersonal due to the involvement of speaker 3 who is actually distant from speakers 1 and 2. The tenor is somewhat formal because of the topic being discussed. Power relation differs as speaker 3 is evidently older and more experienced than speakers 1 and 2. Affective involvement can be said to be high as all participants in the discourse have personal involvement with the experience being discussed. Speaker 3 adopts a cultural convention as a discursive strategy for contributing to the ongoing discourse. He achieves this by referring to speakers 1 and 2 as his children which is an extended kinship term peculiar to Nigerians. Nigerians conventionally assume that anyone who is old enough to give birth to a person can assume the role of a parent both in their linguistic choices and societal expectations. This justifies Speaker 3’s contribution to the discourse which was presented as a succor, in form of a prayer, to the two other speakers as a way of playing a culturally expected fatherly role. Excerpt 9 The excerpt is an exchange of pleasantries between relatively distant friends who only get to meet at the pub. Speaker 1: Bro, how na ( How are you, brother?) Speaker 2: Heey my brother. Didn’t see you sitting o. How is work na? Speaker 1: We are managing o. How huzzle and your people na? (How is work and family)

Legon Journal of the Humanities Vol. 30.2 (2019) Page 83 Bamgbose, G.A. & Ajayi, T.M./Legon Journal of the Humanities Vol. 30.2 (2019) Speaker 2: We thank God o and my children? Speaker 1: Dem dey eat eba (They are eating well) The field of this excerpt is a casual exchange of pleasantries. It is the case that when people meet regularly at social gatherings like pubs, they soon become friends and begin to extend family tiers and love to one another. Lexical items like work, your people, children are indications of the casualness of the discourse. The mode of discourse is, of course, spoken and spontaneous. The tenor is familial as seen in the exchange of pleasantries by the participants. The conventional use of words like brother and my children outside their standard usage makes for a warm rapport between the discourse participants and foregrounds the cultural convention of extended family ties among Nigerians. Conclusion This study has examined language use in pubs in three southwestern states with a view to investigating the patterns of discourse in the interactions which ensue among the convivial participants in the discourse. The study finds that language fulfils a high interactional function in these places as people relax and chat about many topics. Using the instrument of register as a theoretical guide, the study shows that women/ladies, politics and football are three common topics/ fields among discussants in pubs in southwestern Nigeria. The mode of discourse is at all times spoken. The oral nature of this kind of discourse often makes for spontaneous and free flow discussions. Tenor reveals that most discussions tend to be casual even when there is high affective involvement. Power relation is mostly equal and in very few instances where there is power inequality, some elements of coordinating devices in common ground were used to maintain cordiality in the discussions. Findings of the paper attest to the fact that language use in pubs can be topical and reveal discourse patterns. Further research can investigate the patterns of discourse in pubs in other climes. In addition, further studies can be conducted on the pragmatics of interpersonal relations in the conversations which ensue during conviviality in pubs.

Page 84 Legon Journal of the Humanities Vol. 30.2 (2019) Bamgbose, G.A. & Ajayi, T.M./Pub Interactions in Southwestern Nigeria References Ayodele, J. O., Adeleke, K. H., and Gandonu M. B. (2018). Crime and adolescent drug use in Lagos, Nigeria. Sociology International Journal, 2(2):64–73 Bada, F.O. and Adebiyi, D. R. (2014). Alcohol consumption behaviour among secondary school students in Nigeria. Journal of Educational and Social Research, 4 (3), 507-510. Bernard, H. S. (1994). Research methods in anthropology: qualitative and quantitative approaches (2nd ed.). Walnut Creek, CA: Altamira Press Clark, H. (1996). Using Language. Cambridge University Press. Clark, H. and Brennan, S. (1991). Grounding in communication. In L.B. Resnick, J.M. Levine and S.D. Teasley (Eds.) Perspectives on socially shared cognition (pp. 127-149). American Psychological Association: Washington DC. deMunck, V. C. and Sobo, Ellisa J. (Eds) (1998). Using methods in the field: a practical introduction and casebook. Walnut Creek, CA: Altamira Press. Dumbili, E.W. (2013) Changing patterns of alcohol consumption in Nigeria: an exploration of responsible factors and consequences. Medical Sociology online, 7 (1), 20-33. Dumbili E. W. (2015). ‘What a man can do, a woman can do better’: gendered alcohol consumption and (de) construction of social identity among young Nigerians. BMC Public Health, 15, 167. Dumbili, E. W and Henderson, L (2017), Mediating alcohol use in Eastern Nigeria: a qualitative study exploring the role of popular media in young people’s recreational drinking. Health Education Research, 32, 279–291. Dumbili, E. W. and Williams, C (2017). Awareness of alcohol advertisements and perceived influence on alcohol consumption: a qualitative study of Nigerian university students. Addiction Research & Theory, 25(1), 74-82. Eggins, S. (1994). An introduction to systemic functional linguistics. London: Pinter. Halliday, M.A.K. (1978). Language as social semiotic. London. Edward Arnold. Halliday, M.A.K and Hassan, R. (1991). Language, context and text: Aspects of language in a social semiotic perspective. Oxford: Oxford University Press.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 85 Bamgbose, G.A. & Ajayi, T.M./Legon Journal of the Humanities Vol. 30.2 (2019) Lamidi, I. M. (2016). Humour markers and their interpretations in the Nairaland virtual community . https://semanticsarchive.net/archive/mm4ytiwm/ lamidihumor.pdf (Accessed 30 December, 2018). Matthiessan, C. (1993). Register in the Round: Diversity in a Unified Theory of R Register Analysis. In M. Ghadessy (Ed.), Register Analysis (pp. 221-292). London and New York: Pinter publishers. Odebunmi, A. (2006). A Pragmatic reading of Ahmed Yerima’s proverbs in Yemoja, Attahiru, and Dry leaves on Ukan trees. Intercultural Pragmatics. 3(2), 153- 169. Odebunmi, A. (2007). Meaning expressions of some English register. The International Journal of Language, Society and Culture. Issue 21. Ojedokun, U. A. and Ademuson, A. O. (2017). Gender role and vocational experiences of married women beer parlour operators in Ibadan, Nigeria. Ibadan Journal of Sociology, 6 (1), 67-84. Poyton, C. (1985). Language and gender: Making the difference. Geelong.vic: Deakin Stockwell, P. (2002). Sociolinguistics. London and New York: Routledge.

Page 86 Legon Journal of the Humanities Vol. 30.2 (2019) Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) DOI: https://dx.doi.org/10.4314/ljh.v30i2.5

A Reconsideration of Grammatical Categorization in English: The Clause

John Franklin Wiredu Associate Professor Department of English University of Ghana, Legon, Ghana Email: [email protected] or [email protected]

Submitted: November 2, 2018/ Accepted: July 14, 2019/ Published: December 30, 2019

Abstract Various models for categorizing grammatical units exist in the literature. Particularly nebulous, however, are terminologies for conceptualizing categories at the level of the clause. For example, the term complement has been used, confusingly sometimes, as both a functional term and, on some occasions, as a structural term. In this study, a model of categorization is proposed, which keeps functional classification separate from structural classification, in order that concepts used in grammatical analysis can be clearly understood and be applied more easily in text analysis. Accordingly, the specific aim of this study is to examine some already ‘accepted’ subcategories of the dependent clause and, in the process, propose a new approach for the analysis of this clause type. This approach, it is hoped, will remove (a) semantic considerations from the analysis of these grammatical units and (b) the confusion between the notional and syntactic description of the units. In the end, this work involves a reconsideration of some traditional ways of doing grammatical categorization with the hope that it will ease identification of grammatical categories. The overall goal is to offer a more consistent and structured approach to the grammatical analysis of the clause in English.

Keywords: clause, dependent clause, grammatical role, grammatical function

Introduction The human tendency to categorize is assumed to be a necessary process that enables humans to deal with reality. Not unexpectedly, therefore, all aspects of language analysis involve categorization in one form or the other. Indeed, as Dienhart (1999, p. 98) explains it: We spend our time comparing one thing to another, and behaving according to the categorization we make … Growing up in any society involves, in a large measure, discovering what categories are relevant in the particular culture in which we find ourselves.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 87 Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) Categorization, therefore, is a fundamental process in the human investigation of reality. In other words, categorization is a process which enables humans to organize, differentiate and classify our experiences into categories in order to understand them. Categorization is said to give “structure to the enormous amount of sensory imprint” (Rauh, 2010, p. 1). Indeed, in the opinion of this linguist, the most important value of categorization is that it enables us to formulate generalization so that our statements do not refer to individual issues. This means that categorization enables us to organize human phenomena into insightful groups. This view of categorization is given further support by Givon (2010, p. 27), when he says: At whatever level of complexity, organisms depend for their survival on sorting out the myriad individual tokens of incoming experience into much fewer adaptively- significant types, and then tailoring their reaction to the adaptive value of those categories. It is for such reasons that categories are considered indispensable in language study. For, once we recognize categories, we are able to describe the relations that exist between units and, in the process, are able to provide structures within which these units operate. Therefore, for purposes of language analysis, the underlying assumption is that linguistic behavior and cognition both involve recognizing categories and establishing category boundaries. Unfortunately, however, when it comes to what linguistic studies say about categories, a fuzzy picture emerges. There is, for example, a variety of confusing labels to describe grammatical units of analysis. An associated problem, therefore, is not knowing whether this plethora of labels will refer to different categories or will be considered as synonymous. For instance, according to Haspelmath (2001), there is some confusion about labels such as word class, parts of speech, lexical category and syntactic category (c.f. Chomsky 1986; Radford, 1997; Haegeman and Gueron, 1999). It is in this connection that Rauh (2010, p. 4) remarks that: … the terms employed for linguistic categories do not give a clear picture. On the one hand, different category labels are used for the same set of items and, on the other, the same category labels refer to different sets of items…it is the exception rather than the rule to find a clear explanation of the basis on which categories are established, the kind of generalization this categorization expresses, and the insights into language which are thereby gained. For instance, the term complement is used, sometimes, as a functional label and, at other times, as a structural label. Similarly, the traditional classification of the relative clause can be problematic. The conceptual distinction made between restrictive (defining) andnon-restrictive (non-defining) clause types is semantically

Page 88 Legon Journal of the Humanities Vol. 30.2 (2019) Wiredu, J.F./A Reconsideration of Grammatical Categorization in English: The Clause motivated and, therefore, has often created some confusion in grammatical analysis. The result of this state of affairs is that there are many confusing labels being used to describe clausal elements so that a student analyzing a sentence cannot be sure of what to make of all these labels and what actually they stand for. Motivation for the Study This work has arisen precisely because of the problem we often encounter in the analysis of the dependent clause in English. As we have just remarked, there is some fuzziness in classifying the dependent clause types. For instance, one recurring question that may be legitimately asked is: what makes the dependent clause in 1(a) below a complement clause but the dependent clause in 1(b) a relative clause? 1. a) The information that Henry is a pastor is shocking. b) The information that Henry wrote down is shocking. Why can’t we simply identify them both as belonging to the same subtype of dependent clause? Similarly, what makes the relative clause in 2(a) restrictive but the clause in 2 (b) non-restrictive? 2. a) The doctor whom I visited yesterday is my brother. b) The doctor, whom I visited yesterday, is my brother. Thirdly, Quirk et al. (1985) have identified the italicized structures in the sentences below as verbless clauses: 3. a) When ripe, oranges are delicious. b) Though doubtful at first, the students agreed with the plan. c) While in the city, the man visited the zoo. d) Whether successful or unsuccessful, he puts in his best. e) He stumbled downwards, conscious of heavy footsteps behind him. The reason offered for classifying them as verbless clauses is that they are elliptical versions of finite dependent clauses. They have no overt subject or predicate, but their missing constituents are recoverable from the context of information supplied by the independent clause in each sentence (Popa, 2008). In other words, the following are all ‘verbless’ clauses: 4. a) when ripe b) though doubtful at first c) while in the city d) whether successful or unsuccessful

Legon Journal of the Humanities Vol. 30.2 (2019) Page 89 Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) e) conscious of heavy footsteps behind him because they are considered to be reduced versions of the following longer structures: 5. a) when (they are) ripe b) though (the students were) doubtful at first c) while (the man was) in the city d) whether (he is) successful or unsuccessful e) (and he was) conscious of heavy footsteps behind him As expressed by Downing and Locke (2006, p. 209), these structures are not dependent on any particular constituent within the independent clause but they expand the meaning of the main clause by providing a circumstantial feature as time, condition, concession - thus enhancing the message. For a similar reason, Aarts (2008, pp. 54 - 55) refers to the bracketed structures in (6) below as small clauses, because “they lack an overt verb, but can be said to contain an implicit verb be….” 6. a) Martin considers [Tim a creep] b) Phil deems [Henry foolish] That is, these sentences are considered to be ellipted versions of the following examples: 7. a) Martin considers [Tim to be a creep] b) Phil deems [Henry to be foolish] The issues that arise from this state of affairs, as illustrated in (1) – (7), involve problems of definition and classification of the dependent clause. It is these issues relating to confusing definitions and unclear designations of labels which this work intends to tackle; for this is what has prompted this study. Objectives Accordingly, in this study, the aim is to examine some already accepted categories of the dependent clause and, in the process, propose a new approach for the analysis of this clause type. The model of categorization that is proposed keeps functional classification separate from structural classification in order that concepts are understood and can be applied more easily in text analysis. This proposed approach, it is hoped, will remove:

8. a) semantic considerations from the analysis of grammatical units, and

Page 90 Legon Journal of the Humanities Vol. 30.2 (2019) Wiredu, J.F./A Reconsideration of Grammatical Categorization in English: The Clause b) the confusion between notional and syntactic descriptions of units of grammar. Underpinning these objectives are the following assumptions: 9. a) the subject-matter of this study is the dependent clause, its identification and classification b) the study is focused on the syntactic (not the semantic) description of this clause type and is, therefore, concerned with the structure of the clause; c) by structure, we are referring to the position of the clause type in a sentence and its relationship with other elements in the sentence. Analysis and Discussion The prime importance of the linguistic unit called clause in grammatical analysis is captured by Fawcett (2000, p. 200) in the following statement: The unit of the clause is almost certainly central in every human language. It is the syntactic unit that corresponds to the semantic unit of situation ... A clause has the syntax potential that it does because of the particularly rich and complex array of meanings that we wish to express to each other about events. For instance, in the Systemic Functional theory, the clause is considered to be the pivotal unit in the lexico-grammatical analysis of language (Halliday and Matthiessen, 2004, p. 25). And the reason is that The clause is the central processing unit in the lexico-grammar – in the sense that it is in the clause that meanings of different kinds are mapped into an integrated structure. Similarly, in Cognitive theory, clauses are pivotal because they serve as our vehicle for talking about the world and relating occurrences to our own circumstances. Usually, a discourse consists primarily of a series of clauses; only rarely does it comprise a series of nominals. (Langacker, 2008, p. 354) However, two associated problems arise. First, there is a definitional hurdle – which is summed up in the statement below: For grammatical analysis, most linguists agree that there is a unit of language called the clause, which corresponds to the unit of sentence in written language… But our main problem is that it is nearly impossible to define. To the best of my knowledge, there is no existing (therefore, satisfactory) definition. (Fontaine, 2013, p. 20)

Legon Journal of the Humanities Vol. 30.2 (2019) Page 91 Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) To begin with, some theories do not recognize this category at all. For a start, it is observed that theories in the Chomskyan tradition do not use the term clause in their theoretical approach to language analysis (cf. Radford, 2004). In Chomsky’s Minimalist Program, for instance, the term phrase is used to describe such structures as: 10. a) We will visit the man at home Tense Phrase b) ……that we are visiting the man at home Complementizer Phrase c) …….to visit that man at home Infinitival Phrase d) The pupils will visit the man at home Determiner Phrase Also, even those theories which recognize the clause as a linguistic unit (e.g. Cognitive, Functional-Lexical, Construction, Role and Reference, etc.) do not define it. All these theories take it for granted that it is ‘easy’ to identifya clause. (Evans and Green, 2006; Givon, 2010; Quirk et al, 1995; Van Valin, 2010). Moreover, among those theories which attempt a definition, the information is couched in general and inapplicable terms; thus, it can be daunting trying to apply their definitions. For instance, the following ‘definitions’ are found in the literature: 11. a) the clause can be thought of as a structural grammatical unit that expresses a given situation … In functional terms, it describes or relates a particular process and particular participants (Fontaine, 2013, p. 20) b) … clauses have a Subject-Predicate structure with the Predicate being finite or non-finite (Kortmann, 2005, p. 129) c) clauses are the principal structures of which sentences are composed (Leech and Svartvik, 1975, p. 211) According to Matthews (1981), an important distinctive feature of the clause is the concept of predication. This view of the clause ties in with the Aristotelian predicate condition in grammatical analysis: 12. a) every clause has a Subject and b) every clause is an instance of predication This concept of predication, accordingly, may be used to describe the relationship (or tie) between the Subject and the Predicate in a clause (or a sentence). The implication, therefore, is that, since predication is essential, every clause must have a predicative construction.

Page 92 Legon Journal of the Humanities Vol. 30.2 (2019) Wiredu, J.F./A Reconsideration of Grammatical Categorization in English: The Clause For this reason, traditional classical grammar considers that a clause has two main parts – the Subject and the Predicate. In the view of Carnie (2007, p. 51), the Predicate must necessarily contain a verb which may or may not permit other elements to complete its structure. The following examples are all clauses by virtue of the presence of the verbal elements - walks and paid: 13. a) Bentum walks to school b) Our teacher paid our fees In effect, if there is a clause, then, by implication, there is a verb present. Without a verb, there is no clause (Van Valin, 2007). The verb, accordingly, may be seen as the governor which controls the other elements in a clause. And, as expressed by Matthews (2007, p. 58) in his definition of a clause: any syntactic unit whose structure is (or seen as) a reducedform of a sentence. Thus, in particular, one which includes a verb and other elements that accompany it. Therefore, there are evidential grounds, as a result, to define a clause as 14. a linguistic unit which must necessarily contain a verb It does not matter whether the verb is finite or nonfinite. This view of the centrality of the verb in a clause is given further support in some linguistic theories. For instance, in Dependency Grammar and Valency theory, the verb is realized as the regent and, therefore, serves as the governor in the clause. The reason for placing the verb on this pedestal is that a clause is “primarily about a specific act… not about a specific quality or person or object” (Agel and Fischer, 2010, p. 225). Furthermore, Systemic Functional Grammar recognizes the clause as a a linguistic (grammatical) resource for expressing a situation which describes who or what is involved and what kind of relation or activity is involved. (Fontaine, 2013, p. 21) And, in the analytical framework of the Role and Reference Grammar, the CORE element is the clause and, within the clause, there is the NUCLEUS which contains the Predicate. Together, the nucleus and the arguments constitute the core elements which form the clause in this theory (Van Valin, 2010). From these arguments, then, one can make the following deductions about the clause:

15. a) any structure that contains a verb is a clause b) the number of verbs in a stretch of language will correspond to the number of clauses in the text

Legon Journal of the Humanities Vol. 30.2 (2019) Page 93 Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) It can be noted then that the following are all examples of clauses: 16. a) that girl is quite hungry b) whose fish swam fast c) (in order) to pay for his release from prison d) what these boys told us e) (for) buying oranges in town Irrespective of whether the verbs are finite or non-finite, their presence in the structures signals that they are operating within clauses. It must be remarked that the conclusion stated in 15(a) rejects the idea that only structures with finite verbs can be clauses. As has been discussed above, finiteness as a consideration is irrelevant in identifying a clause (Kortmann, 2005, p. 129). For this reason, this study does not accept a definition like the following: 17. (a phrase) is a group of words without a finite verb and which functions as a noun, an adjective, a verb or an adverb (Dadzie and Grant, 2008, p. 8) For implied in this definition of a phrase is the notion that a group of words which has a non-finite verb is a phrase and that only when there is a finite verb can the structure be identified as a clause. Thus, based on this definition above, the italicized sections in the following structures are phrases not clauses: 18. a) The defendant admitted his guilt by nodding his head. b) Disturbed by my letter, she wept. It is contended in this work, however, that, as per the definitions in (14) and 15(a), they qualify to be clauses. As regards 15(b), the number of verbs in each structure below corresponds to the number of clauses therein: 19. a) You are our hope (= 1 clause) b) As the man climbed the tree, we all clapped for him (= 2 clauses) c) In order to see my carpenter, who was ill, I had to make a detour. (= 3 clauses) d) We have realized that those baskets sold in the market belonged to the woman waiting outside (= 4 clauses) The next issue of interest in this study involves what we consider to be the problems associated with the subcategorization of certain clausal structures. This results from the often-confusing categorization of the dependent clause.

Page 94 Legon Journal of the Humanities Vol. 30.2 (2019) Wiredu, J.F./A Reconsideration of Grammatical Categorization in English: The Clause To begin with, there is noticeable absence of a recognizable and clearly defined basis for the classifications that have been established for the dependent clause. For example, there are different classificatory labels assigned to a dependent clause such as 20. waiting for you in the room when it is used as follows: 21. a) Waiting for you in the room was your wife b) I met your wife (who was) waiting for you in the room c) Waiting for you in the room, your wife slept off. Though it has the same structure in each sentence, each of its use is tagged differently; the grammatical context of its occurrence determines its classificatory label. On the other hand, the following clauses have different constituent structures but are not tagged differently because they occur in a similar linguistic environment: 22. a) They know when Hassan arrived in town b) They know whom Hassan wants to marry. Accordingly, it should be possible to assign clear linguistic principles which can explain these phenomena. In addition, dependent clause types are given different labels in grammar texts. And, quite often, there is a mix-up of semantic, syntactic and discoursal criteria when such classifications are done (cf. Biber et al, 1999; Huddleston and Pullum, 2002; Sinclair, 1990). This results in indistinct, confusing dependent clause types such as those identified below: 23. a) contact clause The girls were afraid of the stories she told them b) proportional clause As she grew desperate, (so) her work deteriorated c) free (nominal) relative clause He only eats what he orders from the shop d) comment clause As you know, this concept is centuries old e) what-clause What I want you to do is to go to the Turkish Embassy f) wh-clause I thought I knew what was on his mind

Legon Journal of the Humanities Vol. 30.2 (2019) Page 95 Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) Indeed, one can go on endlessly listing such subtypes and their accompanying misleading definitions and examples. The argument, therefore, is that there isa need to reconsider the parameters that have been used in current literature and to define, in very clear terms, the basis for the categorization of the dependent clause. Accordingly, in this study, a two-tier approach is proposed when sub- classifying the dependent clause. This approach involves the consideration of the dependent clause according to a) its formal structure b) its grammatical function Formal Structure The term structure is being used in the sense that Peregrin (1995, pp. 86-87) has defined it: …let us notice that to speak about a structure is possible only where it is possible to speak about parts and wholes. Indeed, structure is the way of organizing parts into a whole… to speak about a structure is necessary only where two different wholes may consist of the same parts. Indeed, structure then iswhat makes the difference. Using this argument as a basis, we can see that the difference in the dependent clauses found in the structures below arises because they differ in their internal structural components: 24. a) Because Ralph used to ride a bicycle, he has grown strong. b) Riding a bicycle can be a healthy exercise. c) To ride a bicycle, one needs to be brave. The argument is that these dependent clauses should be treated as different because their formal structures are different. For instance, 24(a) is a clause with a subordinator because. 24 (b) is different in structure because the verb is participial, while 24 (c) has an infinitival verb form. As a result, these clauses will be treated as belonging to different sub-types. Basing our subcategorization of dependent clauses on their structural differences, the following sub-types can be identified: 25. a) subordinative b) interrogative c) participial d) infinitival e) complementizer f) relative

Page 96 Legon Journal of the Humanities Vol. 30.2 (2019) Wiredu, J.F./A Reconsideration of Grammatical Categorization in English: The Clause The Subordinative Clause This dependent clause type is introduced by subordinators: after, although, because, before, since, until, etc. Examples of such clauses are: 26. a) After you left the town, the riots broke out. b) They watched the game, while we cooked the meal. c) Immediately we sighted the moon, we ended the fast. In all these clauses, there is a subordinator present at clause-initial position.

The Interrogative Clause This is a clause type introduced by interrogators such as who, whose, what, where, when, how, etc. Examples of their use are found in the sentences below: 27. a) The witness did not know what to say in court. b) Can you tell us why you couldn’t win the race? c) How they managed to enter the locked room remained a mystery. It is seen that the clauses have the interrogators – what, why and how – at clause-initial position. The Participial Clause In such a clause, the verb has the continuous (-ing) or perfective (-en/ed) form:

28. a) The man sitting under the tree is waiting for his son. b) Challenged by the pupils, the teacher became angry. c) Written in 1760 by Cicero, this book is the oldest in the library. The Infinitival Clause This is a clause whose verb form is in the infinitive: 29. a) The need to train more teachers has become critical. b) To win them to your side, you must listen to their views. c) Aren’t you happy to receive this prize? The Complementizer Clause This is a clause introduced by the complementizer that. Examples of this subtype are found in these sentences below: 30. a) The news that her team lost the game has shaken her confidence. b) I have discovered (that) he sings very well in church. c) She is aware that she is brilliant. The complementizer that occurs at clause-initial position in each clause.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 97 Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) The Relative Clause These are clauses which contain the relative pronouns – who, whom, which, that. Examples of their occurrence are found in the sentences below:

31. a) That driver (whom) we all disliked has been arrested. b) The lion chased the cow which ran into the bush. c) That prefect who spoke to us was brilliant. In each case, there is a relative pronoun which occurs at clause-initial position. The point to stress here is that, using the structural information about the clause, it is possible to account for the dependent clause types in English. We have not found any dependent clause type that cannot be accounted for by this approach. It is important to make one observation here. It is noted that two of these clause types – the infinitival and the participial – are identified based on the form of the verbal element. But, the other four types – subordinative, interrogative, complementizer, relative – are identified not by their verbal forms but by the class of the word which occurs at clause-initial position. Thus, for instance, if the initial word is a subordinator, then the clause is identified as subordinative. In addition, it is observed that, in some cases, a clause can have more than one of these criteria used here. For example, the following clauses are both subordinative and infinitival: 32. a) In order to meet the admission requirements, he rewrote the paper. b) Sharon eats early so as to avoid indigestion at night. Similarly, it is possible to have clauses which are subordinative as well as participial: 33. a) Before travelling to his village, Thomas had a meeting with his staff. b) He slept off while waiting for the teacher. And, in some cases, we can have a clause which is both interrogative and infinitival:

34. a) What to wear at the wedding is my greatest worry. b) Do you know who to contact in the office? The conclusion one can draw from the discussion is that it is necessary to be clear what criterion one is using as a basis for the classification of the clause.

Page 98 Legon Journal of the Humanities Vol. 30.2 (2019) Wiredu, J.F./A Reconsideration of Grammatical Categorization in English: The Clause Grammatical Function It is important to explain what the term function means in this study. To begin with, linguistic theory can be grouped into two major strains – formalist and functionalist. The formalist approach focuses on the form and structure of the language. It concerns itself with the deductive properties of a language system. The formalist theories view grammar as an autonomous system devoid of meaning. Thus, emphasis is placed on principles which are solely defined in terms of structure (cf. Nichols, 1984; Radford, 1982). In contrast, the functionalist approach works on the idea that linguistic study should focus on the communicative function of grammatical structures and that linguistic elements must be studied in terms of how they contribute to the communicative functions of language (Hengeveld and Mackenzie, 2008). In discussing the functionalist tradition, Downing and Locke (2006) discuss three ways in which a grammatical approach can be said to be functional. In the first case, a functional analysis works on the assumption that we organize language in such a way as to describe our experiences and to enable us interact with others. Accordingly, linguistic analysis (from this angle) should focus on identifying the systems which we use to achieve these goals. This is the view of the term functional in Systemic Functional Grammar (Fawcett, 2000; Chapman and Routledge, 2009). In this theory, grammatical structures are seen as the results of choices made from a set of options (i.e. systems). However, this is not the view we hold in this work. The second sense in which functional is used refers to the regular patterns which are said to reflect the uses to which we put language. For example, the declarative sentence pattern is used for specific pragmatic purposes – statement of facts, narration and other speech acts. Thus, such analysis is aimed at explaining how we interpret these linguistic forms to convey meaning (Leech and Svatvik, 2013). This is not the sense the term is used in this study. There is a third sense in which the term function is used. In this case, the term is used to describe a linguistic approach which focuses on the relations between constituents in a clause. In other words, in this method of analysis, a linguistic element is not considered independently of other elements but is seen in relation to other elements – the idea being that a constituent can exist only in relation to another constituent with which it occurs in an environment. What this means is that different functions can be assigned to a constituent based on its relationship with other constituents. In English, for instance, the function of a constituent is realized vis-à-vis its position to another constituent. Thus, a subject function is realized only because of its position in relation to the verb. Accordingly, we can say that

Legon Journal of the Humanities Vol. 30.2 (2019) Page 99 Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) a constituent like the Nominal Group has a potential to realize different functions (Matthews, 2007; Miller 2011). This is the sense in which this label is used in this work. In this sense, then, when we talk of sub-classifying dependent clauses on the basis of their ‘grammatical function’, we mean that sub-types of the clause are set up based on their grammatical roles. For, in English, grammatical role is determined by the position a constituent assumes in relation to a ‘parent’ constituent. Thus, word order is crucial in determining a specific grammatical function in the language. Therefore, considering their relative position within a sentence, dependent clauses can be identified as performing the following functions: 35. a) nominal b) adjectival c) adjunctive The key assumption is that, irrespective of its formal structure, a dependent clause can be classified as nominal, or adjectival, or adjunctive. In addition, it is important to stress that, because these functional labels of the dependent clause are determined by the position of the clause in relation to other elements in a sentence, a given clause may be identified with different functional labels. For instance, a clause such

36. who won the competition is categorized in different terms in the sentences below: 37. a) We can point out who won the competition b) We can point out the lady who won the competition Even though the clause has the same structural constituents, its functional identity is different in each sentence. Accordingly, instead of referring to a clause as a nominal clause, for example, it is more proper to identify the clause as performing a nominal function. In view of this observation, then, we can now discuss below the three grammatical functions that a dependent clause can perform in the language. The Nominal Clause This refers to a clause which, within a sentence, occupies a position that a Nominal Group is expected to occupy. In English, a Nominal Group can potentially occur as Subject, Object, and Complement within the clause and as the completive element in a Prepositional Group. This is exemplified by the various positions the Nominal Group, my friend, assumes below:

Page 100 Legon Journal of the Humanities Vol. 30.2 (2019) Wiredu, J.F./A Reconsideration of Grammatical Categorization in English: The Clause 38. a) as Subject: My friend is a good player. b) as Object: The cat likes my friend. c) as Complement: Appiah is my friend. d) within PG: They have spoken to my friend. Clauses which can behave similarly as my friend in these cited examples will be identified as performing nominal functions. Accordingly, the under-listed dependent clauses are identified as nominal clauses: 39. a) That the school lacks equipment is no news. b) Please tell us how we can overcome this obstacle c) To remain opposed to our stand has not helped you. d) Relying on outsiders can be very risky. e) His response was what we expected to hear. f) We praised him for speaking the truth. The Adjectival Clause A clause which occurs at rankshifted qualifier position within a Nominal Group or Adjectival Group is referred to as an adjectival clause. If we consider the following sentence: S P C

40. This table which we bought yesterday | is | quite expensive the Subject is 41. the table which we bought yesterday We note, however, that this Subject has an embedded clause 42. which we bought yesterday. The structure of the Subject will have the constituents analyzed as follows: 43. NG m H q | this | table [[ which we bought yesterday ]] Since the clause is operating as an element in the Nominal Group, it is said to be rankshifted to operate as qualifier to the Head – table. Clauses which are embedded in such a position within a Nominal Group or an Adjectival Group are identified as performing adjectival functions. Examples of such clauses are found in the sentences below:

Legon Journal of the Humanities Vol. 30.2 (2019) Page 101 Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) 44. a) The evidence which Peter produced in court was rejected. b) The news that Peter stole the car was proved beyond doubt. c) This is the school where Peter served as the Head Prefect. d) They were happy to receive the award yesterday e) Is it true that you have been promoted now? f) The duty to protect citizens in a country rests on soldiers. It must be noted, however, that the term adjectival can be misleading, because it erroneously connotes the notion that we are dealing with structures which behave like adjectives. Indeed, this is the same sense which we find in many grammar books which use the terms adjectival or relative clause interchangeably to refer to clauses such as: 45. a) The lawyer who defended the accused is our friend. b) Here is the filewhich you were looking for. In many such treatments, these dependent clauses are referred to as ‘adjectival’ or ‘relative’ clauses. But, it is clear that, from our examples in (44) above, it is not only relative clauses which can occupy such positions (as has been so assumed), but it is possible to have infinitival, complementizer, participial and interrogative clauses occur in such positions. Some other examples can be found in the following: 46. a) Her ambition to become the first female MP can prove disastrous. b) The man lying on that bed was once our teacher. c) Are you sure that we will win the cup? d) This is the time when you need your friends around you. e) The award that was conferred on her was well deserved. The Adjunctive Clause Clauses of this nature behave as Adjuncts in a sentence. Usually, Prepositional Groups, Adverbial Groups and a limited number of Nominal groups can function as Adjuncts in a clause. According to Aarts (2008), an important characteristic of Adjuncts is that they are extremely mobile. Indeed, the Adjunct is the only clausal element which is most mobile in relation to the Predicator. Usually, Adjuncts can occur at clause-initial, clause-medial and clause-final positions as in the following examples: A S P O 47. a) Occasionally Ruth praises her children b) Last week the school organized a public lecture c) At the event our students sang the national anthem

Page 102 Legon Journal of the Humanities Vol. 30.2 (2019) Wiredu, J.F./A Reconsideration of Grammatical Categorization in English: The Clause S A P O 48. a) Ruth occasionally praises her children b) The school last week organized a public lecture c) Our students at the event sang the national anthem S P O A 49. a) Ruth praises her children occasionally b) The school organized a public lecture last week c) Out students sang the national at the event anthem When we transfer this positional process to the dependent clause, we realize that some of these clauses can function as Adjuncts. In other words, they can be quite as mobile: 50. a) Simon could not join the choir, because he was too old. b) Simon, because he was too old, could not join the choir. c) Because he was too old, Simon could not join the choir. 51. a) Having lost earlier, Shehu prepared well for the next game. b) Shehu, having lost earlier, prepared well for the next game. c) Shehu prepared well for the next game, having lost earlier. Based on this premise, then, the following examples show dependent clauses which are adjunctive in their function: 52. a) People adore her, wherever she goes b) I left home early to avoid missing the class. c) Thinking of his rejection, the teacher became angry. d) Left alone in the house, I realized my mistake. From the examples above (i.e. 39, 44, 46, 52), we can see that these functional types encompass many of the structural types. In other words, in theory, it is possible for the different structural clause types to be categorized as any of the functional types identified above. That is, based on this proposed analytical approach, we can have the following configurations: 53. a subordinative clause performing a nominal function a) Because you failed the examination is no good reason to kill yourself. b)A fter she lost the election was when she became depressed.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 103 Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) 54. a subordinative clause performing an adjunctive function a) Since my daughter got her new position, I have not seen her b) He’ll refuse the gift if you don’t tell him the source 55. subordinative clause performing an adjectival function a) The period before we went on strike was extremely tense for staff b) Two years after we had returned from the UK, we had no job offers. 56. an interrogative clause performing a nominal function a) I don’t understand what Harriet has been saying. b) Let’s show you who lives in this house. 57. an interrogative clause performing an adjectival function a) This is the place where the police found the car. b) Can you tell us the time when the meeting will hold? 58. an interrogative clause performing an adjunctive function a) Whatever her earlier opinions were, she still voted for me. b) The students hate Robert, however much he tries to please them. 59. a participial clause functioning at a nominal position a) Preparing for this race can be quite daunting. b) The children are fond of playing tennis. 60. a participial clause at an adjectival position a) Those boys walking around the pool are foreign tourists. b) Teams defeated in the first round were not rewarded. 61. a participial clause at an adjunctive position a) Peeping through the hole, I could see everybody in the room. b) Completely drenched by the rainfall, he ran off to the bathroom. 62. an infinitival clause at a nominal position a) To remain opposed to our stand has not helped you. b) Her only option was to resign from that post. 63. an infinitival clause functioning as adjectival a) The first person to cross this bridge wins a scholarship. b) It is your responsibility to persuade Louis.

Page 104 Legon Journal of the Humanities Vol. 30.2 (2019) Wiredu, J.F./A Reconsideration of Grammatical Categorization in English: The Clause 64. an infinitival clause used at an adjunctive position a) To qualify for the competition, she had to train each day. b) You must apologize to Phillip to cool him down. 65. a complementizer clause functioning as a nominal a) That the citizens voted this way was quite unexpected. b) I have just found out that Stephen lied to us. 66. a complementizer clause functioning as an adjectival a) The news that Peter scored the penalty spread to the town. b) The news that was broadcast to us was false. In ending this section, we are reminded of the words of Radford and others (2009:247): It is traditionally said that sentences are structured out of words, phrases and clauses, each of which belongs to a specific grammatical category and serves a specific grammatical function within the sentence containing it. It is in recognition of this distinction between form and function that we have proposed this method so that, in identifying a dependent clause, we will consider each clause on the basis of these two parameters. Conclusion This work, we must repeat, aims to offer another method for categorizing the dependent clause in English. The implied acknowledgement, therefore, is that there are several approaches to categorizing this clause type. Because the clause has generated a lot of interest in linguistic study, linguists have introduced a wide variety of analytical tools. Interestingly, in spite of the diverse range of analytical tools, there is still a constant search for new ways of analysis. Accordingly, we are often faced with a plethora of divergent approaches to analyzing English grammar. It is a testimony to how complex language is that no single approach has been found to receive universal endorsement. As a result, there are disagreements among language analysts. All the same, we must not give up but continue with the search for the ‘best’ approach. It is with this goal in mind – to contribute towards finding the ‘best’ method – that this study has been undertaken. For we consider the method we have introduced in the study much clearer and more easily comprehensible, because we have a clearly defined set of concepts to use in analyzing the dependent clause in the language.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 105 Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) References Aarts, Bas (2008). English syntax and argumentation (3rd ed) New York: Palgrave Macmillan. Agel, Vilmos and Fischer, Klaus (2010). Dependency grammar and valency theory. In B. Heine and H. Narrog (Eds). The Oxford handbook of linguistic analysis. Oxford: Oxford University Press. pp. 223 – 255. Biber, D., Johansson, S., Leech, G., Conrad, S. and Finegan, E. (1999). Longman grammar of spoken and written English. London: Longman. Chapman, S. and Routledge, C. (2009). Key ideas in linguistics and the philosophy of language. Edinburgh: Edinburgh University Press. Carnie, A. (2007). Syntax: A generative introduction. Maiden, MA: Blackwell. Chomsky, Noam (1986). Knowledge of language, its nature, origin and use. New York: Praeger. Dadzie, Anthony and Grant, Judith (2008). New gateway to English for senior high schools. Students’ Book 3. Accra: Sedco. Dienhart, John (1999). A linguistic look at riddles. Journal of Pragmatics. 31: 95 – 125. Downing, Angela and Locke, Philip (2006). English grammar. London: Routledge. Evans, Vyvyan and Green, Melanie (2006). Cognitive linguistics. Edinburgh: Edinburgh University Press. Fawcett, Robin (2000). A theory of syntax for systemic functional linguistics. Amsterdam: John Benjamins. Fontaine, Lise (2013). Analysing English grammar. Cambridge: Cambridge University Press. Givon, Thomas (2010). The adaptive approach to grammar. In B. Heine and H. Narrog (Eds). The Oxford handbook of linguistic analysis. Oxford: Oxford University Press. pp. 27-49. Haegeman, Liliane and Gueron, Jacqueline (1999). English grammar: A generative approach. Oxford: Blackwell. Halliday, Michael and Matthiessen, Christian (2004). An introduction to functional grammar (3rd ed.) London: Hodder. Haspelmath, Martin (2001). Word classes and parts of speech. In P. Baltes and N. Smelser (eds) International Encyclopedia of the Social and Behavioral Sciences. Amsterdam: Pergamon. pp 16538–16545.

Page 106 Legon Journal of the Humanities Vol. 30.2 (2019) Wiredu, J.F./A Reconsideration of Grammatical Categorization in English: The Clause Hengeveld, K. and Mackenzie, J. (2010). Functional discourse grammar. In B. Heine and H. Narrog (eds). The Oxford handbook of linguistic analysis. Oxford: Oxford University Press. pp 367 – 400. Huddleston, Rodney and Pullum, Geoffrey (2002). The Cambridge grammar of the English Language. Cambridge: Cambridge University Press. Kortmann, Bernd (2005). English linguistics: Essentials. Berlin: Cornelsen. Langacker, Ronald (2008). Cognitive grammar. Oxford: Oxford University Press. Leech, Geoffrey and Svartvik, Jan (2013). A communicative grammar of English. (3rd ed). London; Routledge. Matthews, P (2007). Oxford course dictionary of linguistics. Oxford: Oxford University Press. ------(2007). Syntactic relations: A critical survey. Cambridge: Cambridge University Press. Miller, Jim (2011). A critical introduction to syntax. London: Continuum. Nichols, Johana (1984). Functional theories of grammar. Annual Review of Anthropology. 13: 97 – 117. Peregrin, Jaroslav (1995). Structural linguistics and formal semantics. Travaux du Cercle Linguistique de Prague 1 Amsterdam: John Benjamins. Popa, Ecaterina (2008). Non-finite and verbless clauses: Textual values. Rapid Research Letters. 3:329 -339. Quirk, Randolph; Greenbaum, Sydney; Leech, Geoffrey and Svartvik, Jan (1985). A comprehensive grammar of English. London: Longman. Radford, Andrew (1982). Transformational syntax: A student’s guide to Chomsky’s extended standard theory. Cambridge: Cambridge University Press. ------(1997). Syntactic theory and the structure of English. Cambridge: Cambridge University Press. ------(2004) English syntax. Cambridge: Cambridge University Press. ------Atkinson, Martin; Britain, David; Clahsen, Harald; Spencer, Andrew (2009). Linguistics. Cambridge: Cambridge University Press. Rauh, Gisa (2010). Syntactic categories. Oxford: Oxford University Press. Sinclair, John (1990). Collins cobuild English grammar. London: Collins. Van Valin, R (2007). An introduction to syntax. Cambridge: Cambridge University Press.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 107 Wiredu, J.F./Legon Journal of the Humanities Vol. 30.2 (2019) ------(2010). Role and reference grammar and framework for linguistic Analysis. In B. Heine and H. Narrog (Eds). The Oxford handbook of linguistic analysis. Oxford: Oxford University Press. pp. 707 – 738.

Page 108 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) DOI: https://dx.doi.org/10.4314/ljh.v30i2.6

The syntax of Dagbani personal pronouns: an analysis

Abukari Kwame PhD Student in Interdisciplinary Studies College of Graduate and Postdoctoral Studies University of Saskatchewan Saskatoon, Canada Email: [email protected]

Submitted: November 16, 2018/ Accepted: June 17, 2019/ Published: December 30, 2019

Abstract Cross-linguistically, personal pronouns are noted as being deficient in relation to some morphosyntactic and phonological properties. Some striking asymmetries have been identified between strong and weak personal pronouns in relation to modification, coordination/conjunction, whether they have a semantic referent, and can encode focus. This study explores the personal pronominal system of Dagbani along Cardinaletti and Starke’s (1994) typology and observed asymmetries. Using insights from published literature on Dagbani pronouns as well as my understanding as a native speaker, I argue that, unlike personal pronouns in Romance/Germanic languages, Dagbani personal pronouns can be modified by quantifiers, can be coordinated, and can occur in conjunction constructions, as well as encode topic and focus as salient semantic discourse properties. Furthermore, the pre/post verbal distinctions among nonemphatic pronominal forms in Dagbani still hold, even as these occur in coordinated and modified constructions, due to structural constraints imposed on them by coordinating conjunctions and quantifiers.

Keywords: Dagbani, personal pronouns, pronoun coordination, structural deficiency, focus contexts

Introduction Dagbani is a Gur language spoken in Northern Ghana and has three dialects with slight variations in tone and some lexical words (Hudu, 2010). Bodomo (1993) proposed that due to the sibling relationship that exists between Dagbani, Mampruli, Dagaare, and Kusaal, the term Mabia would better serve as a group label for this family of languages. Dagbani is an SVO language with very poor morphological markings. Grammatical roles (including case among nonemphatic pronouns) of arguments are influenced by syntax. Personal pronouns in Dagbani

Legon Journal of the Humanities Vol. 30.2 (2019) Page 109 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) are often grouped into preverbal/postverbal, emphatic/nonemphatic, or weak/ strong pronouns (Purvis, 2007; Olawsky, 1999, 2002). Over the years, the pronominal system of the language has been the focus of some research, however, some disagreements in findings exist with regard to the syntax of personal pronouns. Whereas some studies on Dagbani personal pronouns use the pre/post verbal criterion (Olawsky, 2002) to examine the distribution of personal pronouns (particularly the nonemphatic/weak forms), others have used the grammatical case based on the functions pronouns play within a sentence or a DP (Issah, 2011, 2013). For instance, Olawsky (2002) claims that Dagbani pronouns offer a good example of what he calls “clitics”. He has argued that, since the nonemphatic forms cannot bear stress or stand isolated as full prosodic units unless they are attached to a host phonological unit, these nonemphatic forms of personal pronouns qualify as proclitics. Olawsky has also proposed that the preverbal forms of nonemphatic pronouns are the unmarked forms. Purvis (2007), on the other hand, is of the view that the pre/post verbal notion is problematic. Instead, he offered a primarily phonologically motivated categorization of Dagbani personal pronouns based on processes of cliticization. Additionally, Purvis questioned the notion of emphatic/ nonemphatic labels, stressing that the terms “strong/weak” are better for the emphatic/nonemphatic pronouns, respectively. Furthermore, in studying reflexive pronouns and the structure of a simple noun phrase in Dagbani, Issah (2011, 2013) categorized nonemphatic pronouns according to grammatical case, putting them into nominative, accusative, and genitive pronouns. Cross-linguistically, Cardinaletti and Starke (1994) have studied the pronominal classes of Romance and Germanic languages and offered some interesting descriptive generalizations based on the relationships between the classes of pronouns in these language families. Using morphological, semantic, distributional/syntactic, and phonological (prosodic) asymmetries, Cardinaletti and Starke demonstrated how strong and deficient pronouns in these languages differed. Striking asymmetries are noted in terms of c-modification and coordination, conjunction, having a semantic referent, and the ability to encode contrastive focus (detailed discussion of these will be provided in the following section) between deficient and strong personal pronouns. My aim in this paper is to use Cardinaletti and Starke’s (1994) typology/ model to examine how far Dagbani personal pronouns pattern with the properties and asymmetries identified in their study. Using my native speaker intuition as well as published papers on Dagbani, particularly that of Purvis (2007), I intend to provide a broader descriptive picture of Dagbani personal pronouns in terms of the morphosyntactic differences between emphatic and nonemphatic pronouns.

Page 110 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis Furthermore, I aim to contribute to knowledge on the pre/post verbal categorization of Dagbani personal pronouns, especially in relation to conjunction and quantifier modification. This paper is structured as follows: The next section covers a summary of the main issues discussed in Cardinaletti and Starke (1994) to lay the foundation for the discussion of Dagbani personal pronouns. In this segment, I present the salient distributional differences between strong and deficient pronouns in Romance and Germanic languages as observed by these authors. This is followed an account of the personal pronouns of Dagbani. Both emphatic and nonemphatic pronouns will be stated and a brief description of them provided in terms of their structural differences as well as previous distributional notions about them. After that, I discuss Dagbani personal pronouns according to some of the criteria used in Cardinaletti and Starke (1994). In this part, the Dagbani personal pronouns are discussed in line with their ability to exhibit some of the asymmetries noted in Cardinaletti and Starke (1994) such as: conjunction and modification, containment (the structural relationship between the two sets of pronouns), pre/post verbal occurrences, and their ability to serve as topic or focus in sentences. I then provide an attempted syntactic account on pronoun coordination and modification. Finally, I summarize the main issues raised in this paper and draw my conclusions. Before I proceed to the next section, I would like to make a few terminological observations. When I refer to the sets of Dagbani personal pronouns, I maintain the emphatic and nonemphatic labels so that references to the personal pronouns in Romance/Germanic languages still retain the labels used in Cardinaletti and Starke (1994). Hence, when the labels “weak”, “strong”, and “clitics” are used, they refer to the personal pronouns as used in Cardinaletti and Starke’s study. Also, where a cited source makes use of these terms in relation to Dagbani personal pronouns, a notification is made. Before I proceed to the next section, I would like to make a few terminological observations. When I refer to the sets of Dagbani personal pronouns, I maintain the emphatic and nonemphatic labels so that references to the personal pronouns in Romance/Germanic languages still retain the labels used in Cardinaletti and Starke (1994). Hence, when the labels “weak”, “strong”, and “clitics” are used, they refer to the personal pronouns as used in Cardinaletti and Starke’s study. Also, where a cited source makes use of these terms in relation to Dagbani personal pronouns, a notification is made.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 111 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) Asymmetries in personal pronouns in Cardinaletti and Starke’s (1994) study Several asymmetries have been noted in the phonology, morphology, syntax, and semantics of personal pronouns in Romance and Germanic languages in Cardinaletti and Starke’s study, The Typology of Structural Deficiency (1994). These asymmetries are outlined in this section and then later referred to when Dagbani personal pronouns are discussed. Morphologically, Cardinaletti and Starke (1994) observed that class 2 personal pronouns (deficient forms) are systematically reduced with respect to class 1 (strong) personal pronouns in both Romance and Germanic languages (1). (1) Deficient pronouns have reduced forms relative to strong pronouns

Deficient form Strong form (a) loro a-loro (to them) Italian (b) ho je-ho mu je-mu (him) Slovak The data in (1) suggests that the deficient forms may be contained within the strong forms or are morphologically reduced in form in relation to the strong pronouns. Additionally, Cardinaletti and Starke maintained that class 2 (deficient) pronouns cannot be coordinated, but that they can refer to both human and non- human referents, whereas class 1 pronouns (the strong forms) may be coordinated. However, their interpretation will be limited to only human referents, as in (2). (2) Coordination in personal pronouns (Cardinaletti and Starke, 1994, p. 43) <+human> <-human> German (ꞓ Germanic) (a) Sie sind groẞ √ √ (b) Sie und die daneben sind groẞ √ * they and those besides are tall/big Gun (ꞓ Kwa) (c) Yélè yon wankpè √ √ (d) Yélè kpo yélè kpo yon wankpè √ * She and she and know beauty

Page 112 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis In (2a), the pronoun sie, when used as a deficient pronoun in German, cannot occur in a coordinated structure. However, in such a context, it can refer to both human and non-human referents. On the other hand, in (2b), where sie is used as a strong pronoun, it can occur in a coordinate structure. However, the non- human interpretation is lost even though, morpho-phonologically the shape of the pronoun does not change. The same pattern applies in the Gun data in (2c, d). In the languages in (2), as well as in many other Romance languages, Cardinaletti and Starke indicated that only strong pronouns can occur in coordinated syntactic structures, and that in such structures they can refer only to human entities (2b, d). Based on syntax, Cardinaletti and Starke opined that deficient pronouns have a restricted distribution relative to strong pronouns. Firstly, they maintained that deficient pronouns cannot occur in a base-generated or theta (θ) position, but strong pronouns can occur in these positions, as shown in (3). Secondly, deficient pronouns cannot occur in peripheral positions, such as in clefts, left/right dislocation, or in isolation, as shown in the Italian data (4). They further observed that this pattern cuts across several languages, such as the behaviour of Dutch het (‘it’), Slovak mi (‘to me’), and English it (p. 47). (3) Positional restriction of deficient and strong pronouns (Cardinaletti and Starke, 1994, p. 46) essa- {essa-D ; leiS; Maria} forse l’ha fatto {* D; leiS; Maria} da sola

It-D ; sheS; Mary may be it-has done DA alone (4) Deficient pronouns and peripheral positions (Cardinaletti and Starke, 1994, p. 47) essa- (a) E’ {* D; leiS; Maria} che è bella (cleft)

It is {*3Sg.fm-D; 3Sg.fm-S; Mary} that is pretty. It is Mary that is pretty. essa- (b) {* D; leiS; Maria}, lei è bella. (left dislocation)

{*3Sg.fm-D; 3Sg.fmS; Mary}, she/it is pretty She/it is pretty. essa- (c) pro arriverà presto, {* D; lei-S; Maria}. (right dislocation)

pro will arrive soon, {*3Sg.fm-D; 3Sg.fm-S; Mary}. She/it will arrive soon. essa- (d) Chi è bella? {* D; lei-S; Maria}. (isolation)

Who is pretty? {*3Sg.fm-D; 3Sg.fm-S; Mary}. Cardinaletti and Starke explained that the deficient pronoun essa in Italian cannot occur in a base position, as signalled by the [*] marking. Hence, deficient forms can only occur in a specially-derived position. In (4a-c), it is shown that the

Legon Journal of the Humanities Vol. 30.2 (2019) Page 113 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) Italian deficient pronounessa cannot occur in cleft, left or right dislocation, or stand alone as an answer to a wh- question, as in (4d). Note that [-D] stands for a deficient pronoun whereas [-S] indicates that the pronoun is a strong one. Furthermore, Cardinaletti and Starke (1994) proposed that, semantically, deficient pronouns must have an antecedent which is already prominent inthe discourse, as illustrated by the data in (5). Note that ostension (accompanying gesture) can be used to establish discourse saliency, as shown in the data by the arrow, suggesting that deficient pronouns must refer to discourse prominent antecedents. (5) Deficient pronouns and discourse prominence (Cardinaletti and Starke, 1994, p. 49) (a) *J’ai vu Marie puis je → l’ ai vu √ J’ai vu Marie puis j’ ai vu → elle. I have seen Mary then I her have seen her (b) √ Mets- toi iҫi et regardes cette maison. To → la vois bien maintenant? Come here and look at this house. You it see well now? Come here and look at this house. Can you see it well now? (c) √Mais, tu ne vois donc pas ce livre? Bien sûr que je → le vois But, you don’t see this book? Of course that I it see But don’t you see this book? Of course I see it. Based on (5), Cardinaletti and Starke claimed that “the deficient elements are permissible with contrastive stress and ostension, only if they refer to an entity which is ‘already prominent in the discourse’” (emphasis in original) (p. 49). Accordingly, they argued that (5a) is impossible, to the extent that both contrastive focus and ostension refer largely to a non-prominent entity in the discourse. However, in (5b, c), the prominent topic of the discourse serves as the referent of the deficient pronounsla and le. In the same vein, Cardinaletti and Starke added that in both expletive and impersonal (non-referential) contexts/constructions, personal pronoun subjects must be deficient, since strong pronouns are uninterpretable in such constructions. Prosodically, it is argued that strong pronouns and lexical nouns are able to form a single prosodic unit, whereas for deficient pronouns, they must be adjacent to a lexical element. This means that they cannot function as a prosodic unit in isolation. Furthermore, Cardinaletti and Starke observed that only deficient pronouns may undergo reduction phenomena.

Page 114 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis In summary, the major asymmetries noted by Cardinaletti and Starke (1994) and further highlighted in Grohmann (2000) in relation to the personal pronouns in Romance and Germanic languages are that deficient pronouns are reduced with reference to strong ones, and can only occur in surface structures in a specially- derived position. This means that deficient pronouns cannot occur in a base- generated position under the VP, dislocated, or in cleft structures. Also, deficient pronouns cannot be coordinated and c-modified and that, where possible, deficient pronouns are preferred over strong ones. Furthermore, deficient pronouns are the only forms that can prosodically restructure, for example, can undergo contraction/ reduction or liaison. Finally, deficient pronouns can refer to both human and non- human referents in a discourse, and can also be used in expletive and impersonal constructions, whereas strong pronouns make reference to only human entities, and can bear their own range-restriction by serving as discourse prominent antecedents (such as acting as contrastive focus and/or ostension – signal gesture). Cardinaletti and Starke, however, argued strongly against some historical notions about weak pronouns. They postulated that, due to what appears to be a historical accident, the inaccurate generalization that deficient pronouns cannot be stressed has come to be seen as a fundamental property of deficient elements (1994, p. 56). Consequently, the fact that deficient pronouns do not occur coordinated, modified, or with ostension has been linked to the notion that they mostly occur unstressed. To contest these notions, Cardinaletti and Starke (1994, p. 57) provided several examples of counter-evidence to show that deficient elements can be stressed (e.g. in essi vanno in chiesa (‘they go to church’), with essi being stressed) in Italian, and can also bear contrastive focus when discourse conditions are satisfied. They also noted that in Gun, a Kwa language in Africa, some strong pronouns may also occur unfocused, which supports their argument that deficient pronouns are not always prosodically inert. To account for the distributional variations between strong and deficient pronouns, Cardinaletti and Starke (1994) proposed that there is a structural deficiency among weak pronouns. Accordingly, a three-class system is proposed where deficient pronouns are further grouped into weak (mildly deficient pronouns) and clitic (severely deficient pronouns) forms. They indicated that some structural containment relationship is obtained between the weak and strong pronouns, for example, as in (6), and between the weak and clitic pronouns, as demonstrated in (7).

Legon Journal of the Humanities Vol. 30.2 (2019) Page 115 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) (6) Containment relationship between weak and strong pronouns (Cardinaletti and Starke, 1994:70) (a) Strong: je-ho je-mu a-loro (b) Deficient: ho mu loro him. Slovak to.him. Slovak to.them. Italian (7) Containment relationship among all the three pronouns (Cardinaletti and Starke, 1994:67). clitic weak strong (a) s < es (Olang Tirolean) il = il (French) (b) ho < jeho (Slovak) loro < a loro (Italian) il < lui (French) sie = sie (German) In both (6) and (7) the structural relationship among all pronominal forms is shown, where clitics are contained in the weak forms, while the weak forms are also contained in the strong forms. Yet, it is supposed that some morphemes are missing in these patterns, hence the proposal that, in terms of morphology, a clitic can be smaller or the same as a weak form, and a weak form can also be smaller or the same as a strong form, as in (8). (8) clitic ≤ weak ≤ strong To explain the difference in structure and function, Cardinaletti and Starke argued that both weak and clitic pronouns lack the CP functional head in their projections. As a result, to compensate for this, and in order to ensure specifier- head agreement, derivation must apply for these deficient forms to function well at the surface level. The detailed syntax involved in the specificier-head agreement derivation, as demonstrated by Cardinaletti and Starke (1994), will not be covered in this paper. Having given a summary of the main issues that Cardinaletti and Starke discussed concerning the nature and distribution of personal pronouns in Romance and Germanic languages, the next section of this paper will present the Dagbani personal pronouns and review them alongside the proposals as outlined by Cardinaletti and Starke (1994).

Page 116 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis Dagbani personal pronouns Dagbani personal pronouns have been grouped into different sets based on either grammatical functions or syntactic distribution in sentences in relation to the verb (Issah, 2011, 2013; Olawsky, 2002). As Purvis (2007) observes, the set of emphatic pronouns in Dagbani has generally been accepted as one distinct set with little dispute or confusion about their description and possible distribution. However, nonemphatic pronouns have not only proved to be difficult to describe, but they also constitute a source of debate among researchers. The description and distribution of this set of pronouns have often been done in relation to the verb, as either pre/post verbal or as subject/object in function. The emphatic pronouns and the nonemphatic ones are presented below Emphatic personal pronouns in Dagbani Personal pronouns in Dagbani mark only person, number, and animacy. These pronouns do not mark gender, unlike third person forms in other languages with very rich morphology, or as in English. Purvis (2007) claimed that Dagbani emphatic personal pronouns are not marked for a grammatical role or for any other function, which means that they can assume any grammatical role in sentences. Further, he observed that emphatic pronouns can be represented in their non-elided forms, suggesting that Dagbani emphatic pronouns have their full forms and elided/ truncated forms1 (for example; mani vs. man’ and ŋuna vs. ŋun’), both of which can occur in sentences. Another interesting observation that Purvis made about Dagbani emphatic pronouns is that they can be coordinated or adjoined freely with many other lexical nouns or pronouns. The set of Dagbani emphatic pronouns are presented in Table 1 in (9). (9) Table 1: The set of emphatic pronouns in Dagbani (Purvis, 2007:240)

Person/Number Singular Plural 1st person mani (man’) tinima (tinim’) 2nd person nyini (nyin’) yinima (yinim’) 3rd person [+animate] ŋuna (ŋun’) bana, banima bannima (bannim’) 3rd person [–animate] dina (din’) ŋana (ŋan’) [dinnima] Table 1 in (9) shows the truncated forms of each pronoun in parenthesis beside the untruncated ones. As Purvis argued, the plural form of the 3rd person inanimate pronoun in recent times is [dinnima]2. The forms [ŋana, ŋan’ or ŋannima] were 1 A reviewer pointed out that, this is not surprising since final vowel deletion in emphatic personal pronouns is a common phonological phenomenon in Mabia languages. 2 A reviewer observed that pointed the form [din-nima] seem more uniform and transparent since it contains [-ma] or the default plural marker [-nima], which seems a general plural suffix of emphatic pronouns. Nonetheless, whether speakers prefer this form

Legon Journal of the Humanities Vol. 30.2 (2019) Page 117 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) all attested to in the Western dialect (Tomosili) of Dagbani spoken in communities around Tamale. However, I am not sure if these forms are still very actively used in the Tomosili dialect. Also, in the plural forms of the 3rd person animate pronoun, the form [banima] is an interrogative pronoun (i.e. ‘which people’) rather than an emphatic pronoun. The emphatic form of that pronoun is [bannima], which is also noted by Issah (2013) in his analysis of Dagbani simple noun phrases. Again, the form [bɛna] is also attested to as a 3rd person animate plural pronoun, in the Nayahili dialect of Dagbani. Nonemphatic personal pronouns of Dagbani Like the emphatic pronouns, the nonemphatic personal pronouns also mark person, number, and animacy features, with the last feature restricted to only the 3rd person forms. As pointed out earlier, different arguments have been made about this set of personal pronouns in terms of their sentential distributive properties. Issah (2011) grouped these pronouns into nominative, accusative, and genitive forms, while Olawsky (1999, 2002) categorized them into preverbal and postverbal forms. However, Purvis (2007) claims that, a closer look at these pronouns reveals that the variation in opinion among scholars appeared to be based on structural and grammatical considerations influenced by the subject/object grammatical case distinctions made in English or other similar languages. Accordingly, Purvis believed that this categorization of Dagbani nonemphatic pronouns by educated Dagbani speakers is due to the influence of English grammar. A list of nonemphatic pronouns is presented in Table 2 in (10), as taken from Olawsky (1999, p. 22), but with some modifications. (10) Nonemphatic pronouns

(a) Table 2: Dagbani nonemphatic personal pronouns

Person Singular Plural preverbal unmarked postverbal preverbal unmarked postverbal 1st person n/ŋ/m ma [tɨ] ti [tɨ] 2nd person a yi [yɨ] ya 3rd person o bɛ [bɨ] ba [+animate] 3rd person di [dɨ] li di [dɨ] (ŋa) li [-animate] over others is an empirical question.

Page 118 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis (b) bɛ chaŋ-Ø puu-ni. 3Pl.Prev go-Perf farm-Loc They have gone to the farm. (c) Adam ti-Ø ba pini. Adam give-Perf 3Pl.Postv gift. Adam gave them a gift. For illustration, (10b,c) shows the pre/post verbal occurrence of some of the pronouns in Table 2. This issue will be taken up later in relation to the argument that the pre/post verbal distinction is not strict, since Purvis (2007) argued that postverbal pronouns can occur in preverbal positions and vice versa in sentences. Furthermore, in Table 2 (10a), the preverbal form of the 1st person singular [n] is homorganic and gets assimilated to the place of articulation of the following stop sounds and may thus surface as [n, m, ŋ]. Phonologically, Olawsky (1999, 2002) observed that the structure for most of these pronouns is that they contain a consonant and a schwa-like vowel which appears in the orthography as . Accordingly, he proposed that if those pronouns are seen as containing an epenthetic schwa, then their forms will be /d/, /l/, /t/, and /b/ for [dɨ], [lɨ], [tɨ], and [bɨ] respectively. This explains why the other forms of these pronouns are in square brackets in Table 2 with the muted – i [ɨ]. Another interesting observation about the nonemphatic pronouns is that the singular form of the 2nd person pronoun is the same in both preverbal and postverbal positions, as is the 3rd person animate form. Also, the plural form for the 1st person pronoun is the same in the pre/post verbal forms, while there is number syncretism for the 3rd person inanimate forms. These pronouns with identical morphemes either in their pre/post verbal forms or in their singular and plural forms will have some implications concerning their distribution in sentences as well as in coordinated structures, as I will show subsequently. For a clearer picture, the set of nonemphatic pronouns in (10) are repeated in (11), below, with the identical ones noted in boldface.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 119 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) (11) Non-emphatic pronouns which are identical in form Singular forms Plural forms Prev. Postv. Prev. Postv. 1st n ma ti ti 2nd a a yi ya 3rd o o bɛ ba di li di li Given that the nonemphatic pronouns noted in (11) are the same, it is difficult to tell which form of the 1st plural, the 2nd singular, and 3rd person singular pronouns are preverbal, and which are postverbal, as well as which of them can occur in subject or in object positions. Further discussion of this will be presented in the following section. Note also that the 3rd person animate pronouns in the final row have no distinction between their singular and plural forms. Having outlined the two sets of personal pronouns in Dagbani, I shall explore these pronouns along the lines proposed by Cardinaletti and Starke (1994). I will examine the relationship between the two sets of pronoun types, concerning their ability to conjoin, to be modified, to be focused, to be used as topics, and whether there is any containment relationship between them. Preverbal forms in post verbal positions and vice versa It has been observed that the preverbal forms of the nonemphatic pronouns can occur in postverbal positions while the postverbal forms can also occur in preverbal positions. However, the data in (12) seem to suggest otherwise. (12) Personal pronouns and their positional restrictions in Dagbani (a) Yakubu ti-Ø ma pini. Yakubu give-Perf 1Sg.Postv.Nonemph gift.Sg Yakubu gave me a gift. (b) Abu tu-Ø ba pam. Abu insult-Perf 3Pl.Postv.Nonemph much Abu insulted them so much. (c) *ti ti-Ø bɛ nimdi. 2Pl.Prev.Nonemph give-Perf 3Pl.Prev.Nonemph meat We gave them meat. (d) *ma chaŋ-Ø puu maa ni zuŋɔ. 1Sg.Postv.Nonemph go-Perf farm Def Loc today I went to the farm today.

Page 120 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis In (12a), the postverbal form of the 1st singular pronoun is used while in (12b) the postverbal form of the 3rd person plural form is used. Both forms occur in postverbal positions, which is why the sentences are grammatically correct. However, (12c, d) are ungrammatical because the preverbal form [bɛ], the 3rd person plural, is used in a postverbal position, rather than the postverbal form [ba], and in (12d) the postverbal form of the 1st person singular, [ma], is used in a preverbal position instead of the preverbal form [n]. These sentences show that the preverbal forms cannot be used in a postverbal position except in conjunction or quantified phrases. Exploring Dagbani pronouns based on Cardinaletti and Starke’s (1994) typology Based on the summary regarding the nature of personal pronouns in Romance and Germanic languages presented earily in the paper, this section will explore Dagbani personal pronouns along those asymmetries to determine how well they pattern with the generalisations noted by Cardinaletti and Starke. Basic similarities and differences will be noted as I explore the personal pronouns of Dagbani. Containment relationships between Dagbani personal pronouns Using insights from Cardinaletti and Starke (1994), I will say that the relationship between nonemphatic and emphatic personal pronouns involves some form of containment. Table 3 in (13) shows both the nonemphatic and emphatic personal pronouns of Dagbani. In order to explore the containment proposal of Cardinaletti and Starke, I divide each emphatic pronoun into two morphemes (base forms + suffixes) to determine whether they pattern with observations made by Cardinaletti and Starke regarding deficient/strong pronouns in their study. (13) Table 3: The combined emphatic and nonemphatic personal pronouns Person Nonemphatic pronouns Emphatic pronouns Singular forms Plural forms Singular Plural Prev Postv Prev Post 1st person n ma ti ti ma-ni ti-nima/tinim’ 2nd person a a yi ya nyi-ni yi-nima/yinim’ 3rd person o o bɛ ba ŋu-na ba-na,

[animate] ban-nima 3rd person di li di li di-na din-nima/ [-animate] din-nim’

Legon Journal of the Humanities Vol. 30.2 (2019) Page 121 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) A closer look at the forms between nonemphatic and emphatic pronouns seems to indicate that there is some amount of containment relationship. In the 1st person pronoun forms, we could have [ma] contained in [ma-ni] and [ti] contained in [ti-nima], and same for the 3rd person inanimate forms [di] in [di-na] and [di- nima]. The 2nd and 3rd person singular animate forms (marked in grey in Table 3) are monosyllabic in structure. However, their plural forms also seem to show some kind of containment relationship with their emphatic counterparts. The 2nd person form [yi] is contained in [yi-nima], the 3rd person forms [bɛ/ba] are contained in [ba-na] or [ban-nima], and the 3rd person inanimate form [di] is also contained in the emphatic form [di-na] or [din-nima]. Based on the above, it would appear that the singular forms of the emphatic pronouns are realised as [ma-] and [nyi-] plus some suffix [-ni] for the 1st person and 2nd person pronouns, while the 3rd person forms have [ŋu-] and [di-] plus a suffix [-na] for the animate/inanimate forms. Nonetheless, this segmentation does not produce any meaningful forms in Dagbani. Also, it is difficult to provide an explanation as to how the singular forms of the 2nd [a] and 3rd [o] nonemphatic pronouns change into the emphatic forms [nyi-ni] and [ŋu-na], respectively. The plural emphatic forms, on the other hand, take the plural forms of the nonemphatic pronouns and add the suffix [-nima] or its shortened form [-nim’] to form the plural emphatic pronouns. Given that [-nima] is a default plural marker for some nouns, including loans (Olawsky, 1999, p.93), it makes sense to segment it as a number morpheme in these pronominal forms. Hence, the nonemphatic forms [ti, yi, bɛ, di] are contained in the emphatic forms [tinima, yinima, bannima, dinnima], respectively. Two patterns are observed in Table 3. First, there are inconsistencies in the morphological shape of the singular nonemphatic pronouns and their singular emphatic counterparts; therefore, they defy the establishment of any containment relationship. Second, the plural forms of the nonemphatic pronouns show a containment relationship with their plural emphatic counterparts. Therefore, comparing the relationship between the emphatic/nonemphatic pronouns in Dagbani to those of the strong/weak pronouns in Cardinaletti and Starke’s study, I cannot conclusively argue that Dagbani personal pronouns pattern exactly with the structural relationships that exist between strong/weak pronouns in Cardinaletti and Starke (1994), although some form of containment can be seen between the plural forms of the emphatic/nonemphatic pronouns in Dagbani.

Page 122 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis Dagbani personal pronouns and conjunction Purvis (2007, p. 241) observed that emphatic pronouns can be coordinated with other nouns and pronouns, examples of which are presented in (14). (14) Coordinated emphatic personal pronouns (a) Yisa ni nyini n yɛn chaŋ. Yisa Conj. 2Sg.Emph Foc will go Yisa and you will go. (b) nyini ni mani ka yɛliɡu. 2Sg.Emph Conj. 1Sg.Emph NEG say You and I have no say. (c) ŋuna ni yinima, ninvuʔ-yo-ya n nyen ya. 3Sg.Emph Conj. 2Pl.Emph, person-bad-Pl Foc be 3Pl. Nonemph he/she and you, you are bad people. In these sentences, (14a) indicates that we can have a noun conjoined with an emphatic pronoun. In (14b, c), we have emphatic pronouns in a coordinate structure. However, these forms seem to be marked and if the pronoun is not in a focused context the nonemphatic forms may be preferred. For instance, (14a, b, and c) could be uttered as (15a, b, and c), respectively, with nonemphatic pronouns. For naturalness in speech, the forms in (15) are preferred over those in (14), which seems to suggest that the nonemphatic forms in Dagbani pattern with the deficient pronouns in a way described by Cardinaletti and Starke (1994, p. 46) that, where possible, deficient pronouns are preferred over strong ones. Thus, even though both emphatic and nonemphatic pronouns in Dagbani can be coordinated, the nonemphatic pronouns are preferred. (15) Nonemphatic personal pronouns in coordinate structures. (a) a mini Yisa n yɛn chaŋ. 2Sg.Nonemph Conj Yisa Foc will go You and Yisa will go. (b) m mini a ka yɛligu. 1Sg.Nonemph Conj. 2Sg.Nonemph NEG say I and you have no say. (c) a mini o, ninvuɣ-yo-ya n nyen ya. 2Sg.Nonemph Conj. 3Sg.Nonemph person-bad-Pl Foc be 2Pl. Nonemph You and he/she, you are bad people.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 123 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) (d) Yisa mini a n yɛn chaŋ. Yisa Conj. 2Sg.Nonemph Foc will go You and Yisa will go. From (15) above, it is obvious that the nonemphatic pronouns can also be coordinated with lexical nouns or with other nonemphatic pronouns. However, the occurrence of the 2nd and 3rd person singular forms [a] and [o] do not settle the pre/post verbal debate, since they are the same in either position, as already noted in Table 3 in (13). Also, two emphatic pronouns can be conjoined with [bee] (‘or’) in cases of yes/no questions, where falling intonation is used, as in (16). (16) Two emphatic pronouns in a coordinate structure nyini bee bɛna n da-Ø palo maa? 2Sg.Emph Conj. 3Pl.Emph Foc buy-Perf plot Def Is it you or they who bought the plot? Another important factor in examples (14) – (15) is the presence of the focus head in addition to the conjunction markers, as I will discuss later. Furthermore, nonemphatic personal pronouns in Dagbani can also occur in coordinate structures, as shown in (17). (17) Conjoined nonemphatic personal pronouns (a) [Adam mini ba] chaŋ-Ø puu-ni Adam and 3Pl.Nonemph go-Perf farm-Loc Adam and them have gone to the farm. (Adam has gone to the farm with them) (b) [N mini ya] ni chaŋ-Ø Tamale dali. 1Sg.Nonemph Conj. 2Pl.Nonemph Fut go-Perf Tamale a day after tomorrow I and you will go to Tamale a day after tomorrow. (I will go to Tamale with you…) (c) Doo maa ti-Ø [yi mini ba] lɨʔiri. Man Def give-Perf 2Pl.Prev.Nonemph Conj. 3Pl.Postv.Nonemph money The man has given you and them some money. In (17a), a lexical noun and a 3rd plural postverbal pronoun are conjoined in the preverbal position, while in (17b) a 1st singular preverbal pronoun and a 2nd plural postverbal pronouns are conjoined in a preverbal position. In (17c) a 2nd plural preverbal pronoun and a 3rd plural postverbal pronoun are conjoined in a postverbal position. Purvis (2007) argued that the correct forms of the pronouns, given the pre/ post verbal predictions about Dagbani personal pronouns, should be *[Adam mini bɛ] in (17a), *[N mini yi] in (17b), and *[ya mini ba] in (17c). Nonetheless, these forms are unattested in the language. Based on this, Purvis (2007) maintained that the pre/post verbal distinction, as far as the distribution of the nonemphatic

Page 124 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis pronouns is concerned, is problematic. However, what Purvis failed to notice is the fact that, even in such coordinate structures, the postverbal form is always the second element. What this means is that the postverbal nonemphatic forms can never occur as the first conjunct of a conjunction structure, as illustrated in (17c). This data shows that there is a structural constraint imposed on the pronouns even in a conjunction structure, such that the postverbal forms can only occur after the conjunction marker, as shown in (18). (18) More coordinate constructions with nonemphatic pronouns [n mini ba] ‘I and them’ [bɛ mini ya] ‘they and you (Pl.)’ [n mini li] ‘I and it’ [yi nimi ba] ‘you and them’ *[ba mini ya] ‘them and you’ *[yi mini di] ‘you (Pl.) and it’ *[n mini di] ‘I and it’ My argument is that two postverbal forms cannot be coordinated, for example *[ba mini ya], ‘them and you’, and *[ya mini li], ‘you and it’. Also, in a coordinate structure with two personal pronouns, the first pronoun must be a preverbal form, while the second is a postverbal form, as in [bɛ mini ya], ‘they and you’, [ yi mini ba], ‘you and them’, and [n mini li] ‘I and it’. Furthermore, two preverbal forms cannot also be coordinated in the same structure, as in *[m mini bɛ], or *[yi mini di]. Based on these patterns, I propose that the conjunction construction has a structure, as shown in (19), which may impose some constraints on how the pre/ post verbal nonemphatic pronouns should be combined in a coordinate structure. (19) Conjunction phrase structures

Legon Journal of the Humanities Vol. 30.2 (2019) Page 125 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) With respect to the nonemphatic pronouns, I propose tentatively that the Spec of ConjP is occupied by preverbal pronouns, and the Compl of ConjP is occupied by the postverbal pronouns. Then, Spec-head agreement restricts the preverbal forms to the first position of the conjunction structure. The assumption is that it is this co-occurrence restriction that makes it possible for a preverbal pronoun in a coordinated DP (either a lexical noun and pronoun, or two pronouns) to occur in both pre and postverbal positions in sentences, especially with the so- called “preverbal pronouns in postverbal positions”, as observed by Purvis (2007). Comparing the behaviour of both emphatic and nonemphatic pronouns in Dagbani with respect to conjunction, I will say that Dagbani pronouns differ from the personal pronouns of Romance and Germanic languages, in that the deficient (weak and clitic) pronouns in the latter languages cannot occur in coordinated structures. In addition to this, in Romance and Germanic languages, when strong pronouns occur in coordinate structures, they cannot refer to human entities which, again, make them different from Dagbani personal pronouns since, in Dagbani, coordinated emphatic pronouns can refer to human entities. Modifying Dagbani personal pronouns In terms of modification, both emphatic and nonemphatic personal pronouns can be modified with quantifiers. However, the kinds of modifiers they take can differ slightly. In the case of emphatic pronouns, their truncated forms are what surface during modification, a phenomenon which is common among lexical nouns in the language (see Hudu, 2005, for details). The quantifiers that can modify personal pronouns are presented in (20). However, note the difference in form between [-kam] and the rest of the modifiers/quantifiers. kam[- ], ‘every’, appears as a bound form, while [zaa], ‘all’, [gba], ‘also’, and [ko/kɔkko], ‘alone/only’, are all free forms which can modify both lexical and pronominal forms. By way of illustration, (21)-(22) demonstrate how emphatic and nonemphatic pronouns are modified with these quantifiers/modifiers in Dagbani. (20) Some quantifiers in Dagbani (a) zaa ‘all’ kɔnko/ko ‘only, alone’ ɡba ‘also’ shɛba/ shɛŋa ‘some’ pam ‘many/much’

Page 126 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis (b) -kam ‘every’. sokam, ‘everybody’, shɛlikam ‘everything’ yinokam ‘everyone’ (21) Emphatic pronouns and modification (a) ŋun ko/konko, gba ‘he/she only/alone, also’ man ko/konko, gba ‘I only/alone, also’ nyin ko/konko, gba ‘you only/alone, also’ din ko/konko, gba, kam ‘it only/alone, also’, everything’ (b) bɛn zaa, ko/konko, gba, shɛba ‘they all, only, also, some of them’ tinim zaa, ko/konko, gba, shɛba ‘we all, only, also, some of us, yinim zaa, ko/konko, gba, shɛba ‘you all, only, also, some of you’ dinnim zaa, ko/konko, gba, shɛŋa ‘it all, only, also, some of it’ From (21a) we can see that the singular forms of emphatic pronouns can be quantified with konko],[ ‘alone’, and [gba], ‘also’, whereas the plural forms of emphatic pronouns can take [zaa,], ‘all’, [ko/konko], ‘only/alone’, [gba], ‘also’, and [shɛba/shɛŋa], ‘some’. When a singular emphatic pronoun is quantified with [zaa], as in [nyin zaa m-bala], ‘you all that’ which means [that is all of you], it indicates limited ability or possession. For instance, during a conversation, if Azima says to Dokurugu, [nyin zaa m-bala], then what Azima meant is that Dokurugu has no further ability or has nothing more to offer/show in relation to whatever they were talking about. In addition, plural nonemphatic pronouns can take [yino], ‘only/alone’, as a quantifier, but emphatic pronouns cannot co-occur with yino],[ ‘only/alone’, as shown in (22). [yino] has a singular number interpretation, which might suggest that there is a co-occurrence restriction between the plural emphatic pronouns and the quantifier, as shown in (22b)), but this restriction does not seem to apply between plural nonemphatic pronouns and [yino], as illustrated in (22a). (22) Non-emphatic pronouns and quantifiers (a) ti yino (1Pl.Nonemph only) ‘one of us’ yi yino (2Pl.Nonemph only) ‘one of you’ bɛ yino (3Pl.Anim.Nonemph only) ‘one of them’ di yini (3Pl.Inanim.Nonemph only) ‘one of it’

Legon Journal of the Humanities Vol. 30.2 (2019) Page 127 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) (b) *tinim’ yino (1Pl.Emph only) ‘one of us’ *yinim’ yino (2Pl.Emph only) ‘one of you’ *bannim’ yino (3Pl.Emph only) ‘one of them’ *dinnim’ yino (3Pl.Inanim.Emph only) ‘one of it’ Again, when it comes to preference, using nonemphatic pronouns with quantifiers and modifiers seem natural and is likely preferred over emphatic pronouns in some discourse contexts, which could suggest that there may be less restriction on nonemphatic pronouns and quantifiers. However, this is not actually the case, because postverbal nonemphatic pronouns cannot occur with quantifiers at all, except when they occur in coordinate structures. Nevertheless, the data shown in (20)-(22) indicates that, unlike the deficient pronouns in Cardinaletti and Starke’s study, both emphatic and preverbal nonemphatic Dagbani pronouns can take quantifiers as modifying elements. Cardinaletti and Starke’s typology showed that only strong pronouns and lexical nouns can be modified in Romance and Germanic languages. Also, Purvis (2007) observed that when nonemphatic pronouns are quantified, the pre/post verbal distinctions disappear. In relation to that, I propose that the quantifier phrase (QP), as shown in (23), has a structural imposition which makes Purvis’ observation possible. (23) The structure of a quantified DP (a)

yi (you) zaa (all) ŋ (I) konko (alone) bɛ (they) pam (many) *ma (I) gba (also) (b) A quantified conjunctive DP

Page 128 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis

(c) Conjunction has scope over the quantifier

Given the above structures (23b and c), and like the conjunction structure, the quantifier phrase also has a structural restriction where only the preverbal pronouns occur with the quantifier in both pre and postverbal positions in sentences. In (23b), the quantifier has scope over the coordinate structure and will be read as[[yi mini ba] zaa], whereas in (23c) the quantifier has scope over only the 3rd person preverbal pronoun and will be read as [m mini [bɛ zaa]], as detailed in (24). (24) Combining conjoined structures with quantifiers in Dagbani (a) [[m mini ba] zaa] ni chaŋ yiŋa 1Sg.Nonemph Conj 3Pl.Nonemph.Postv all Fut go home I and they all will go home. (I together with them [all of us] will go home)

Legon Journal of the Humanities Vol. 30.2 (2019) Page 129 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) (b) [m mini [bɛ zaa]] ni chaŋ yiŋa. 1Sg.NonEmph Conj. 3Pl.Non-Emph.Prev all Fut go home I and they all will go home (I and all of them will go home).

In (24a) the quantifier has scope over the entire conjunction structure, as shown in the tree diagram below the sentence, hence, the 3rd person plural nonemphatic pronoun being inclusive in the conjunction structure. However, in (24b), the quantifier has scope over only the 3rd person plural pronoun, as also shown in the tree diagram. This means that the quantifier, all’ ‘ , only relates to [bɛ], ‘they’. This suggests that, like the conjunction markers, a quantifier phrase must have a certain structure that makes it possible for the preverbal nonemphatic pronouns to occur in both pre and postverbal positions, when they occur in a quantifier phrase. Otherwise, without any quantifier, the pre/post verbal distinction will remain, as shown in the discussion on “Preverbal forms in post-verbal position and vice versa”, as well as in (25). (25) Dagbani nonemphatic pronouns and quantifiers (a) Akonsi puhi-Ø [*ba/bɛ zaa] viɛnyela. Akonsi greet-Perf 3Pl.Postv/Prev.Nonemph all properly Akonsi greeted all of them properly/very well. (b) Adam puhi-Ø *bɛ/ba viɛnyela. Adam greet-Perf 3Pl.Prev/Postv.Nonemph well Adam greeted them properly/very well. (c) [di zaa] ka o da-Ø. 3Pl.Inanim.Nonemph all Foc 3Sg.Nonemph buy-Perf

Page 130 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis It was all of it (that) he bought. (di is in focus here) In (25a) we have a preverbal form of the pronoun occurring in a postverbal position with the quantifier to form [bɛ zaa]. The occurrence of [ba] in such a context is illicit, since it cannot take a quantifier. However, in (25b), the postverbal form of the pronoun [ba] is used instead of [bɛ] due to the absence of the quantifier. Further, in (25c) a preverbal form occurs, quantified in the postverbal position, and is moved to the preverbal position, as signalled by the focus maker [ka] in the construction. The underlying form of (25c) is [o da-Ø di zaa], ‘he/she bought all of it’. The constituent, [di zaa], is then moved to the initial position in the sentence and becomes focused, as shown in (25c). My argument is that all the postverbal forms, [ba, ya, ma, li], of the nonemphatic pronouns cannot occur with quantifiers, therefore they will always be postverbal in sentences. Lastly, it is also important to note that it is common for native Dagbani speakers to combine both singular and plural nonemphatic pronouns, as well as lexical nouns, with [gba zaa], as in [n gba zaa], ‘I also’, [ti gba zaa], ‘we also’, etc. In such structures, especially where the pronoun/lexical noun is singular, [zaa] may pragmatically function as an emphasizer rather than a quantifier. Dagbani personal pronouns and topic/focus encoding The concepts of topic and focus are important discourse notions which add to the semantic enlightenment and interpretation of conversational exchanges (Kaiser, 2005; Krifka, 2008; Zimmermann & Onea, 2011). The topic in a discourse relates to what is being talked about in an information structure, which can be the same as the grammatical subject, or not. Kaiser (2005) observed that subjecthood, givenness, and pronominalization can all increase the topicality of a referent in a discourse. Topics in sentences can be located at the left periphery, under the CP layer, as well as sometimes at the right edge of a sentence, constituting the entity in a clause or sentence to which the rest of the clause or sentence is about (Rizzi, 1997). Focus, on the other hand, is a salient discourse feature which contributes new, contrastive, or non-derivable information in a discourse.3 Pragmatically, it is the most salient element in a sentence. Focus is also conceived of as indicating the presence of alternatives in discourse, which are important for the interpretation of given linguistic expressions (Zimmermann & Onea, 2011; Krifka, 2008). Focus can be expressed morpho-syntactically, prosodically, or both, and can be signalled in discourse through intonation, pitch accent, or stress (Krifka, 2008; Zimmermann & Onea, 2011). Kiss (1998, pp. 245-246) observed that contrastive 3 Krifka (2008, p. 257), however, suggested that conceiving focus as that which signals importance, newness, or presupposition of existence need not figure in the definition of focus. To Krifka, the basic notion of focus should be about indicating the presence of alternatives for interpretation in the common ground management space.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 131 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) focus (identificational focus) expresses exhaustive identification, and involves syntactic reordering, while presentational focus (information focus) merely conveys new nonpresupposed information without expressing exhaustive identification. Contrastive focus is an operator with features [+exhaustive, +contrastive] and has binding effects, whereas presentational focus is present in every sentence and has no scope (Kiss, 1998). Focus markers and focus constructions in Dagbani have been studied extensively, where [la], [ka], and [n] are identified as morphological focus makers in Dagbani (Hudu, 2012; Issah, 2008, 2012; Schwarz, 2005; Issah & Smith, 2018). However, there are some disagreements among researchers concerning the staus of [la]4, the postverbal focus marker in Dagbani. Scholars have agreed that [la] is a focus marker, but they disagree on its ability to establish contrastive focus. Turning to the behaviour of Dagbani personal pronouns in relation to topic and focus, the following are my observations based on previous studies, as well as my intuition as a native speaker. In Dagbani, emphatic personal pronouns can function as topics in sentences or in discourse, and in some contexts, nonemphatic pronouns may also occur as topics. In (26), the pronouns function as topics. (26) Dagbani personal pronouns and topic/focus encoding (a) ŋuni n tu- Ø bi-hi maa? who Foc insult-Perf child-Pl Def who insulted the children? (b) *n/mani/Adam 1Sg.Nonemph/1Sg.Emph/Lexical noun I/Adam (I insulted the children/ADAM insulted the children (c) mani, man’ n- mɛ dimbɔŋɔ zaa.(Purvis, 2007, p. 241) 1Sg/Emph 1Sg/Emph Foc build.Perf Dem all ‘Me, [It is] I [who] built all these.’ (Disjoint, Emphasis. Subject.) The question in (26a) is in the context of focus and thus requires an answer that is in focus. This answer, provided in (26b), can be an emphatic pronoun or a lexical noun. Hence, the 1st singular emphatic pronoun [mani], or Adam can answer the question, but not the 1st person nonemphatic pronoun [n]. Also, in (26c), the emphatic pronoun [mani] is the topic, and the truncated form [man’] is the focus, as it is immediately followed by the focus marker [n], which Purvis glossed as an infinitive marker used for emphasis. In both (26b) and (26c), the emphatic pronoun [mani], ‘I’, is exhaustively identified from any other alternatives within the privileged possible worlds (PPWs), to use Zimmermann and Onea’s (2011)

4 Hudu (2012) claims that [la] is a postverbal contrastive focus marker which focuses a noun, pronoun, or adjunct in the postverbal position. However, Issah (2008, 2012) thinks that [la] is just a presentational focus marker and not a contrastive focus particle. For details, readers can consult these references.

Page 132 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis conception of focus, that could undertake the activity of insulting or building. In relation to nonemphatic pronouns and focus marking, I argue that nonemphatic personal pronouns can only mark a presentational focus of giving new information in a sentence, but not a contrastive focus, as shown in (27). Studies have even shown that weak pronouns are banned in a focus position and cannot be topicalized either (Agbedor, 1996; Grohmann, 2000). (27) Dagbani pronouns in topic/focus positions (a) bɔ ka Adam da-Ø? what Foc Adam buy-Perf What did Adam buy? (b) loori pal-li ka Adam da-Ø. Lorry new-Sg Foc Adam buy. Adam bought a NEW LORRY. (c) *li ka Adam da-Ø zaɣ’palli. 3Sg.Inanim.Nonemph.Postv Foc Adam buy-Perf one-new Adam bought IT as new (It being focused here). (d) dina ka Adam da-Ø zaɣ’palli. 3Sg.Inanim.Emph Foc Adam buy-Perf one-new It is it that Adam bought as new. In the question in (27a), the object of the sentence is in a focused context. Therefore, the answer must also be focused. This explains why the answer, [loori palli],‘a new lorry’, is focused by preposing it to the sentence initial position where focus is signalled morphologically by the contrastive focus maker [ka], which mean that ‘a new lorry’ is exhaustively identified from other options (such as an old lorry or a slightly used lorry). When the DP [loori palli] is replaced with a pronoun, we get the 3rd person inanimate nonemphatic form [li], which cannot stand alone as an answer to the question. This implies that nonemphatic personal pronouns cannot bear contrastive focus, as doing so results in ungrammaticality in (27c). However, when [li] is replaced with an emphatic form, the ungrammaticality is resolved in (27d). The person uttering (27d) must also be pointing at the ‘new lorry’ (by ostension). Strangely, however, it is possible to focus conjoined nonemphatic pronouns, as shown in (28), even though a single nonemphatic pronoun cannot be focused, as already demonstrated in (27c). Note that the coordinated nonemphatic pronouns which form a constituent are in square brackets.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 133 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) (28) Conjoined nonemphatic pronouns bear contrastive focus a. Munira ti-Ø [n mini ba] liʔiri. Munira give-Perf 1Sg.Nonemph.Prev Conj. 3Pl.Nonemph.Postv money Munira gave I and them money. b. [n mini ba] ka Munira ti-Ø liʔiri. 1Sg.Nonemph.Prev Conj. 3Pl.Nonemph Foc Munira give-Perf money It was I and them that Munira gave money. As can be seen in (28b), the coordinated nonemphatic pronoun is moved to the sentence initial position where it bears focus which is signalled by the focus marker [ka]. Without being in a conjunction structure, nonemphatic personal pronouns in Dagbani cannot bear focus. Another observation is that quantified nonemphatic personal pronouns can also be focused. That is, when a nonemphatic personal pronoun is modified with a universal quantifier, such as zaa],[ ‘all’, and [ko/konko], ‘alone/only’, then the pronoun can be focused in the sentence initial position, as shown in (29b), as well as become topicalized, (29c). (29) Quantified nonemphatic personal pronouns and focus (a) Adamu tu-Ø [bɛ shɛba]. Adamu insult-Perf 3Pl.Nonemph.Prev some of them Adamu insulted some of them. (b) [bɛ shɛba] ka Adamu tu-Ø. 3Pl.Prev some of them Foc Adamu insult-Perf Adamu insulted SOME OF THEM (as opposed to all of them). (c) [yi zaa], ninvuʔ-yo-ya n nya ya.

[2Pl.Prev.Nonemphatic all]TOP person-bad-Pl Foc be 2Pl.Postv You, you are ALL CRIMINALS (All of you are criminals) In summary, emphatic pronouns in Dagbani can serve as topics as well as occur in focused contexts. This observation about Dagbani emphatic pronouns concerning their ability to encode focus or topic makes them pattern with the behaviour of strong pronouns in Cardinaletti and Starke’s (1994) study. Nonemphatic pronouns can only occur in focus or topic positions when they are modified by quantifiers or coordinated with other pronouns or lexical nouns. Towards a syntactic account of pronoun coordination/modification In this section, I attempt to account for some of the observations made earlier in the paper. I draw insights from the studies of Agbedor (1996), Grano (2006), and Grohmann (2000).

Page 134 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis In a study on personal pronouns in English using both corpus and experimental data, Grano (2006) noted interesting findings on pronoun coordination in English. Although frequency of use, the impact of prescriptive grammar, and other pragmatic factors (e.g., forms considered to be prestigious) were found to be relevant, Grano’s (2006) Pronoun Linear Precedence (PLP) Constraints, as in (30), is of much importance in this paper, especially those that are bolded. (30) Pronoun Linear Precedence (PLP) Constraints (Grano, 2006, p. 40) 1. “X and I”: [anything] < I 2. Lexical Status: pronoun < full-NP 3. Case: nominative < accusative 4. Person: 2 < 1 < 3 Where 1 is first and 3 is third person singular pronouns in English. Grano (2006) explained that under the PLP, the constraint “X and I” ranks highest, and means that anytime the pronoun “I” is involved in a coordinate structure, it will be ordered second. Between a pronoun and a full-NP, a pronoun will tend to precede a full-NP whenever there is no “I”. And when neither “I” nor a full-NP is involved, a nominative pronoun will tend to precede an accusative one. Finally, if none of these constraints apply, then a 2nd person pronoun will tend to precede a 1st person pronoun, and a 1st person pronoun will tend to precede a 3rd person pronoun. In exploring these constraints among coordinated Dagbani nonemphatic pronouns, it will be argued that the “X and I“ constraint is independently ruled out in Dagbani based on the data presented in (31). (31) Coordinating Dagbani nonemphatic pronouns and Grano’s (2006) PLP Constraints (a) [X and 1] (b) [pronoun < full-NP] *[di mini n] ‘3Sg.Inanim and 1Sg’ [n mini Azima] ‘1Sg and Azima’ [n mini ya] ‘1Sg and 2Pl’ [yi mini Azima] ‘2Pl and Azima’ [n mini ba] ‘1Sg and 3Pl’ [bɛ mini Azima] ‘3Pl.Anim and Azima’ *[yi nimi ma] ‘2Pl and 1Sg’ [di mini Azima] ‘3Sg.Inanim and Azima’ (c) [Nom. < Acc.] [n mini li] ‘1Sg.Prev and 3Sg.Inanim.Postv’ [m mini ya] ‘1Sg.Prev and 2Pl.Postv’ [bɛ mini li] ‘3Pl.Anim.Prev and 3Sg.Inanim.Postv’ *[ba mini ya] ‘3Pl.Anim.Postv and 2Pl.Postv In this data, the “Nom. < Acc.” constraint would be the highest ranked in Dagbani (see (31c)), since in (31a) there are no attested forms that obey the coordination constraint “X and 1”. Also, the pronoun-before-full-NP constraint would be applicable in Dagbani (31b), since the form *[Azima mini ma], is ruled

Legon Journal of the Humanities Vol. 30.2 (2019) Page 135 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) out, along with any coordination structure in which the second conjunct is a singular 1st person pronoun, although full-NPs may precede 2nd and 3rd person pronoun forms. In terms of person ordering, Dagbani follows an order of 1 < 2 < 3 or 1 < 2 = 3, since both 2nd and 3rd person pronouns can be coordinated, with either of them occurring as first or second conjunct, even though the 2nd person pronoun being the first conjunct would seem more natural. For example, in 2 < 3 order, we can have [a mini o], ‘2Sg and 3Sg’, [yi mini ba], ‘2Pl and 3Pl’, and [yi mini li], ‘2Pl and 3Sg’. Whereas in 3 < 2 order, we can have [o mini a], ‘3Sg and 2sg’, [di mini o], ‘3Sg and 2Sg’, [di mini ya], ‘3Sg and 2Pl’, and [bɛ mini ya], ‘3Pl and 2Pl. Therefore, with the notion of “preverbal pronouns in postverbal position” and vice versa, the constraints regulating that, according to Grano’s (2006) PLP, will be “Nom. < Acc.”, which would further place Dagbani in Pattern III of Grano’s (2006, p. 49) pronoun case-marking typology, where coordinated pronouns are “Nom. < Acc.” both in subject and in object positions, which I take to be pre and postverbal positions, respectively. Concerning the quantifier modification, it could be argued that in Dagbani, only the “Nom.” pronominal forms may occur with quantifiers. Considering these observations, I propose that Dagbani nonemphatic pronoun distribution is influenced by case. Based on Agbedor’s (1996) analysis of weak pronouns in Ewe, the nonemphatic pronouns in Dagbani could be argued to be projected under IP where they are governed by INFL. Thus, avoiding complex syntactic processes, it could be stated that preverbal nonemphatic pronouns are generated at Spec IP and postverbal nonemphatic pronouns are the complement of the verb. Then, nominative and accusative cases are assigned to them, respectively. Furthermore, since Dagbani nonemphatic pronouns cannot bear contrastive focus and cannot stand alone as answer to a wh- question, they must be projected differently compared to emphatic pronouns, which behave like lexical nouns. Both emphatic pronouns and lexical nouns do not require case, and when they occur in focus contexts, they are usually followed by focus markers. In general, therefore, nonemphatic pronouns can only be projected under CP when they occur with a quantifier or in a coordinate structure, in which case they can also be focused or get topicalized in the syntax, as shown in (29c). Nonetheless, I leave this observation open for further research.

Page 136 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis Summary and conclusion This paper investigated the syntax of Dagbani personal pronouns along Cardinaletti and Starke’s (1994) typology. It also explored the notion of pre and postverbal categorization of Dagbani nonemphatic pronouns, which led to a few descriptive generalizations about Dagbani personal pronouns. First, all emphatic pronouns in Dagbani can occur in both pre and postverbal positions. The nonemphatic pronouns, on the other hand, have some syntactic distributional restrictions in sentences. Previous notions that some are preverbal while others are postverbal is supported. However, when nonemphatic pronouns occur in conjunction or quantifier constructions, some combinatorial restrictions are imposed, making preverbal forms to always occur as the first pronoun, and postverbal forms as the second pronoun. In such structures, the so-called “preverbal forms in postverbal positions” and vice versa are plausible. Additionally, whereas emphatic pronouns can bear contrastive focus, nonemphatic pronouns can only bear contrastive focus when modified by quantifiers or when they occur in conjunction structures. Dagbani nonemphatic pronoun distribution is also argued to be influenced by Grano’s (2006) PLP Constraints and pronoun case-marking typology. Based on Cardinaletti and Starke’s (1994) typology, Dagbani personal pronouns differ in several ways. On the one hand, unlike in Romance and Germanic languages, both emphatic and nonemphatic personal pronouns in Dagbani can be coordinated. They can also be modified by quantifiers. On the other hand, as in Romance and Germanic languages, there is a minimal containment relationship between the plural emphatic and plural nonemphatic pronouns in Dagbani, and only emphatic pronouns can be focused or topicalized. Due to the behaviour of the nonemphatic pronouns with conjunctions and quantifiers in the language, I propose that further research is needed to fully explore the nonemphatic pronouns in quantifier and conjunction phrases, as functional projections in Dagbani. In conclusion, I observe that Romance and Germanic languages’ pronouns and Dagbani pronouns have little in common with regards to their morphosyntactic behaviour.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 137 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) Abbreviations 1, 2, 3: first, second and third persons respectively; Anim. – animate; Def – definite determiner; Dem – demonstrative; Conj. – conjunctive marker; Emph. – emphatic pronoun; Foc – focus marker; Fut. – Future marker; Imperf – imperfective aspect; Inanim. – inanimate; Loc. – locative marker; NEG – negative particle; Nonemph. – nonemphatic pronoun; Num-Root – number root; Perf – perfective aspect; Pl – plural; Q – quantifier; QP – quantifier phrase; Sg – singular; Prev. – preverbal; Postv. – post verbal; ConjP. – Conjunctive phrase Acknowledgment I express my appreciation to the anonymous reviewers and the Journal administrator for their valuable comments and suggestions that helped to improve the analysis presented in this paper. I also thank Miss Sarah Caufield for proofreading the paper and making beneficial suggestions. All remaining errors are mine.

Page 138 Legon Journal of the Humanities Vol. 30.2 (2019) Abukari, K.The Syntax of Dagbani personal pronouns: an analysis References Abels, K. and Muriungi, P. (2008). The Focus marker in Kiitharaka: Syntax and Semantics. Lingua, 118, 687-731. Agbedor, P. (1996). The syntax of Ewe personal pronouns. Linguistique Africaine, 16, 19-51. Bodomo, A. (1993). Complex predicates and event structure: An integrated analysis of serial verb constructions in the Mabia languages of West Africa. Master's thesis. Trondheim: University of Trondheim. Working Papers in Linguistics, no. 20. Cardinaletti, A. & Starke, M. (1994). The Typology of Structural Deficiency: On the Three Grammatical Classes. University of Venice Working Papers in Linguistics, 4(2), 41-110. Grano, T. (2006). "Me and her” meets “he and I”: Case, person, and linear ordering in English coordinated pronouns. (Honors Thesis). California: Stanford University. Grohmann, K. K. (2000). Towards a Syntactic understanding of prosodically reduced pronouns. Theoretical Linguistics, 26(3), 175-210. Hudu, F. (2005). Number marking in Dagbani. Thesis. Alberta: University of Alberta. Hudu, F. (2010). Dagbani tongue-root harmony: A formal account with ultrasound investigation. (PhD dissertation). University of British Columbia, Canada. British Columbia. Hudu, F. (2012). Dagbani focus particles: A descriptive study. Journal of West African Languages, XXXIX(1), 97-129. Issah, S. A. (2008). Information Packaging in Dagbani, MPhil Thesis. Tromso: University of Tromso. Issah, A. S. (2011). Reflexive pronouns in Dagbani. Legon Journal of the Humanities, 22, 129-146. Issah, A. S. (2012). Phrasal identification and contrastive focus in Dagbani.Journal of West Arfican Languages, XXXIX(1), 75-96. Issah, A. S. (2013). The structure of the Dagbani simple noun phrase. South African Journal of African Languages, 33(2), 203-212. Issah, A.S. & Smith, P.W. (2018). Subject and non-subject ex-situ focus in Dagbani. Goethe-Universität, Frankfurt, 1-38.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 139 Abukari, K./Legon Journal of the Humanities Vol. 30.2 (2019) Kaiser, E. (2005). Effects of topic and focus on salience. In Chritian, E. & Coenelia, E. (Eds.) Proceedings of the Sinn und Bedeutung 10th annual meeting of the Gesellschaft für Semantik, 1, 139/154. Kiss, E. K. (1998). Identificational focus versus infromation focus. Language, 74(2), 245-273. Krifka, M. (2008). Basic notions of information structure. Acta Linguistica Hungarica, 55(3-4), 243-276. doi:10.1556/ALing.55.2008.3-4.2 Olawsky, J. K. (1999). Aspects of Dagbani grammar: with special emphasis on phonology and morphology. Munich: Lincom. Olawsky, J. K. (2002). What is a word in Dagbani? In Dixon R.M.W. & Aikhenvald, A.Y. (Eds.). Word: A Cross-Linguistic Typology. Cambridge: Cambridge University Press. Purvis, M. T. (2007). A reanalysis of non/emphatic pronouns in Dagbani. In Frederick Hoyt et al.( Eds.) Texas Linguistics Society IX: The Morphosyntax of Underrepresented Languages, 239-263. Rizzi, L. (1997). The fine structure of the left periphery. In Haegeman. L.Elements of Grammar: Handbook in generative syntax (pp. 281-337). Dordrecht, Netherlands: Kluwer. Schwarz, A. (2005). Copulative and Narrative Patterns in Gur Focus Constructions. Gur Conference “Between tone and text”, 12-14 October 2005, Bayreuth, (pp. 1-9). Bayreuth. Zimmermann, M. & Onea, E. (2011). Focus marking and focus interpretation. Lingua, 121, 1651-1670. doi:10.1016/j.lingua.2011.06.002

Page 140 Legon Journal of the Humanities Vol. 30.2 (2019) Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) DOI: https://dx.doi.org/10.4314/ljh.v30i2.7

Definiteness in the Zarma determiner phrase

Waheed Ayisa Jayeola Lecturer Department of Linguistics and African Languages Obafemi Awolowo University, Ile-Ife, Nigeria E-mail: [email protected]

Submitted: March 29, 2018/ Accepted: September 18, 2019/ Published: December 30, 2019

Abstract Adjectives in definite Determiner Phrases of Zarma, a Nilo-Saharan language trigger an additional (suffixed/base-merged) lexical determiner; in the event of adjectives modifying nouns, definite determiners can occur with either the nouns or the adjectives or both. In all of these cases, no different readings obtain. Structured interviews were conducted with Zarma native speakers to collect the data for this study. I analyse the phenomenon as a case of definite determiner doubling which does not bear on any form of agreement relations. I further suggest that definite determiner and its subsets – numeral, demonstrative, and quantifier do not overlap. However, each of these can occur alongside adjectives within the DP. Consequently, I consider the adjective as an exponent of the adjunct category. Based on Abney’s DP-Hypothesis and the restrictive theory of the Minimalist Program, the paper argues that the asymmetry in the surface realizations of elements/constituents within the Zarma DP is the effect of movement.

Keywords: definite determiner, adjective, functional head, determiner doubling, constituent order

Introduction Cross linguistically, variations exist in the way definite expressions are shown. Thus, it is possible to find languages where definite descriptions are realized through the combination of two words: a lexical word, mostly of the category N (Noun), and a functional word of the category D (Determiner), which provides reference to an individual marked as specific or known. It is possible to find in some languages what has been described in the literature as definite adjective; it seems like a vestige of this phenomenon exists in Zarma but it is not treated as such. It is also possible to find languages with a single word capable of dual interpretations either as a definite description or as an indefinite expression. It is interesting to

Legon Journal of the Humanities Vol. 30.2 (2019) Page 141 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) mention that the two possibilities are available to Zarma, a Songhai language of the Nilo-Saharan family spoken in some north western states of Nigeria and places in the Republics of Niger, Benin and Mali. A small number of nouns in Zarma are taken to be composites of two words (lexical and functional) whereas in a large number of its nominal expressions the particle that functions as the exponent of definiteness or referentiality is simply missing. However, it would be argued that such definite expressions contain a covert category that does not affect the forms of the words in which it is realized. Thus, definite expressions devoid of an overt definite marker are assumed to have phonetic forms with descriptive denotations as well as a null category with a referential value. This is done with a view to explaining the idiosyncrasy of Zarma in relation to its mapping of syntactic structures to Logical Form (LF) i.e. the interpretive interface. Thus, the purpose of the present article is to investigate the patterns of the distribution of definite determiner in relation to nouns and adjectives within the DPs of Zarma. And I shall put up the claim that the occurrence of the items recognised as the definite determiner marker either with nouns or adjectives is not an instance of agreement marking but a case of determiner doubling. Also, it shall be shown that the definite determiner is in complementary distribution with its subsets – numeral, demonstrative and quantifier. This shall be undertaken within the confines of the DP analysis (Abney, 1987; Longobardi, 1994, among others) where the determiner is treated as a functional category which necessarily projects its own phrase and takes a noun phrase (NP) as its complement. Relevant to this study are works on Arabic by Fassi Fehri (1999) and Gungbe by Aboh (2004a, 2013), where adjective and determiner (Specificity and Number) respectively occur after the head noun in their superficial syntax. The phonetic forms of the resultant structures are considered to have been so derived by assuming Kayne’s (1994) Antisymmetry Theory, where all movement is to the left and no headedness parameter is assumed. The paper proceeds as follows: the first section discusses the rudiments of the theoretical framework employed for this study. In the second section, I present empirical evidence for the presence of definite and indefinite articles which may not be realized overtly. I discuss noun-attributive modifiers in the third section. Here, I argue that relative clauses and adjectives are traditional modifiers of nouns but more emphasis is placed on the behaviours of adjectives in terms of their relationships with nouns within the DPs. The character of adjectives in relation to their interaction with some functional categories which prevent co-occurrence makes the proposal for double definiteness and the treatment of adjectives as an adjunct category worthwhile. The last section concludes the discussion.

Page 142 Legon Journal of the Humanities Vol. 30.2 (2019) Jayeola, W.A./Definiteness in the Zarma determiner phrase Theoretical Framework This study is undertaken within the tenets of the Minimalist Program, Chomsky (1993, 1995) and several other versions including the DP-hypothesis which is one of the major hypotheses that have been incorporated into the program. The minimalist program, based on compatibility, is strongly aided by Kayne’s (1994) Antisymmetry Theory. The Minimalist Program (MP) is considered as an attempt to reduce the levels of linguistic representations to two so that superfluous steps in the analysis of linguistic constructs can be easily eliminated. Thus, the two levels of linguistic representations that are postulated in the MP referred to as interfaces are the Phonetic Form (PF) and the Logical Form (LF). Similarly, the restrictive theory of the Minimalist Program incorporates a movement approach and focuses on the order of phrase structure composition by “operation merge”. On its part, the DP hypothesis recognizes determiner (D) rather than the head noun, as the head of its own phrase (DP) (Abney, 1987). The empirical and theoretical argument provided by Abney is that, Determiner Phrase, a functional category, is the maximal category that is projected by the class of determiner elements and which heads the noun phrase or has the noun phrase as its complement. The kind of complex constituent orders found in Zarma seems to be the focus of Kayne’s (1994) Antisymmetry Theory, which through the Linear Correspondence Axiom (LCA) seeks to place restriction on the possible syntactic representations that are available to human language. This approach favours the position that only one linear order is available for all languages which is Spec-Head-Complement. According to Kayne, languages that deviate from the order are assumed to involve/employ movement at one stage in the development of such languages. Zarma Definite/Indefinite Article Generally speaking, article is a term used in the grammatical classification of words, which refer to a subclass of determiners whose primary role is to differentiate the uses of nouns. According to Matthewson (1998, p. 25) articles do not have a unique semantic value across languages, thus, a distinction is usually made into definite and indefinite (or non-definite) types. It is a category that expresses the referential properties of nouns. In Zarma, article (definite/indefinite) appears consistently after the nouns. Definite article, from cross linguistic perspective, expresses definiteness. Thus, the association of a definite determiner to a noun in Zarma suggests that either the interlocutors share the referent of the noun in question or that the definite determiner limits the interpretation of the noun. Morphologically, the marking of

Legon Journal of the Humanities Vol. 30.2 (2019) Page 143 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) definiteness on nouns is not straightforward as it is possible to realize the item as a separate word or an inflected suffix or enclitic on the noun stem which within the spectrum of the minimalist program is understood as the output of operation merge. Consider in this connection the following illustrations in (1). 1a. i almárí - evening ii almáró - the evening b. i hárí - water ii háró - the water c. i má - name ii máá - the name d. i sáná - needle ii sáná á - the needle e. i kùsú - water pot ii kùsó - the water pot f. i bónkòní - king ii bónkònó - the king From the data in (1), definite determiner markers are essentially post nominal. They cause a change in the form of the vowels and the form or shape of the vowel seems to bear largely on some phonological accounts like agreement or vowel harmony. The English glosses provided for the data in (1) suggest that the definite determiner is singular, and could be marked by either o or a. The reason for this is that number in Zarma is a functional category that independently projects into a Number Phrase and is highly restricted in terms of co-occurring with definite article and other functional categories (see Jayeola, 2016). The realisation of the definite article marker as either an inflectional affix (enclitic) or a free standing morpheme may be sensitive to speech rate which could be spontaneous or slow. Baring other phonological/morphological facts, in normal or spontaneous speech, the definite article is often realised as a free standing morpheme as shown in (2). 2a. má ó berì name Def big ‘the big name’ b. hábú ó rá market Def P ‘in the market’

Page 144 Legon Journal of the Humanities Vol. 30.2 (2019) Jayeola, W.A./Definiteness in the Zarma determiner phrase A glance at the examples in (1&2) shows that the marking of definiteness on nouns alternates between the use of o and a. Tentatively, it is speculated that the alternation between the choice of o/a as the definite article marker could be explained in terms of assimilation or distribution. It is conceivable from the data in (1) and (2) that definite determiner markers are essentially post nominal. However, it is harder to capture in clear terms the distribution of [o] and [a] which are the concrete markers of the definite determiner because they could be realised as either an inflectional affix (enclitic) to demonstrate the effect of merge or as a free standing morpheme. The situation is made difficult by the irregularity observed in the speech rate effect which could be spontaneous or slow. It is common to realise the marker as a free standing morpheme in a slow speech but it does not mean that we cannot find it as an enclitic. This is the case as illustrated in (3). 3a. má-ó berì name-Def big ‘the big name’ b. hábú-ó rá market-Def P ‘in the market’ c. i kùsó berì water pot-Def big ‘the big water pot’ ii kùsú ó berì water pot Def big ‘the big water pot’ d. i chíní ó rá night Def P ‘in the night’ ii chínó rá night-Def P ‘in the night’ e. i Súsúba ó rá morning Def P ‘in the morning’ ii * Susubo ra morning-Def P

Legon Journal of the Humanities Vol. 30.2 (2019) Page 145 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) What has been shown of the definite determiner in (3) appears to defy any uniform phonological account. Particularly unclear is the reason for the non- convergence of the example in (e, ii) when in fact the expression in (d, ii) is suspected to have obtained from the same process. However, the matter is treated as the product of the grammatical properties of the lexical item súsúbà ‘morning’ which blocks or prevents it from being base-merged with the definite marker; it can only combine or merge externally with the definite marker to form a constituent, i.e. determiner phrase (DP), as indicated by the grammaticality of the example in (e, i). The implication of this claim is that, it may not be appropriate to use the orthographic forms of the lexical items as an index to recognise the definite article as a free standing morpheme. In spite of the irregularities contained in the phonological shapes or representations of the definite determiner, I assume that, in all instances, definiteness comes either through a replacement of the final vowel of the noun stem or an addition of either vowel [-o] or [-a] whose distribution can be explained as an effect of the nature of vowel harmony which the language allows. That is to say, the ATR value of the vowel in the noun stem determines the choice of an inflected vowel (affix) that gives the noun its definite reading. What has been said of the definite marker does not spread through all the nouns in Zarma because there are nouns that neither take overt inflections of the type just discussed nor seem to merge overtly with the items already identified as definite markers. For instance, the following nouns do not overtly mark for definiteness or specificity but one piece of semantic evidence is that they are appropriately interpreted at the LF by the interlocutors – i.e. they take default interpretations because the determiner particle is assumed to be covert. 4. i. zàma – knife ii. gùrí – egg iii. kòròfó – rope iv. labárì – talk v. hángá – ear vi. tóndí – stone vii. tása – plate viii. ánkóríbe – ear ring ix. bíná – heart x. rógó – cassava xi. fafa – breast xii. kòròkòti – maize

Page 146 Legon Journal of the Humanities Vol. 30.2 (2019) Jayeola, W.A./Definiteness in the Zarma determiner phrase A survey of the examples of nouns without overt definite marker as shown in (4) does not suggest any uniform features which can be said to be responsible for the way they tend to behave. Thus, a potent way of analyzing the situation in the data above which will be consistent with the syntax of Zarma is to propose a phonetically zero definite determiner for those nouns. This proposal is not novel because instances of this abound in the following English words – sheep, information and several others which do not have overt plural markers. One pertinent question that arises from our analysis is how to account for indefiniteness or unspecificity which lacks morphological marking/representation in Zarma if the zero definite determiner proposal is in the right direction. I start by explaining indefiniteness as an expression or construction in which the listener does not have any knowledge of the referent of the noun in question. The nominal expression kòròfó in each of the sentences in (5) is not ambiguous between two interpretations, i.e., it does not call for either definite or indefinite reading in spite of the absence of indefinite/definite determiner marker on it. 5a. áí de kòròfó 1sg buy rope Indef ‘I bought a rope’ b. áí na kòròfó de 1sg Perf rope-Def buy ‘I have bought/bought the rope’ It is not doubtful; going by the translations of the constructions in (5), that kòròfó ‘rope’ a bare noun may be interpreted as indefinite or definite without any corresponding change in its morphological form. However, syntactic variation in the order of the object DP is observed. This fact bears directly on a report by Cheng and Sybesma (1999), and Aboh (2004a, 2004b) that Sinitic and most Niger- Congo languages do not have definiteness articles of the Indo-European type, but encode definiteness by other syntactic devices that are arguably not expressions of D, i.e., pre- versus post-verbal position. By implication, in addition to the use of overt definite marker, Zarma also encodes definiteness by a syntactic device of pre- versus post-verbal position. Following Jayeola (2016) who treats proper names and pronouns as expressions of D and also asserts that relative clauses involve a D that embeds the relative complement clause in Zarma, it is the position of this paper that the examples in (5) involve the category D which does not have a reflex at Phonetic Form. This account is in tune with Abney (1987), Longobardi (1994), and Cardinaletti and Starke (1999).

Legon Journal of the Humanities Vol. 30.2 (2019) Page 147 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) Furthermore, the analysis pursued here suggests that there is no morphological or phonological constraint on the distribution of the definite marker. However, the semantic property of the determiner elements (covert/overt) gives the noun its definite or indefinite reading. Consequently, I treat the indefinite/definite article as an inherent part of the grammar of Zarma which may be realised overtly or covertly on the noun it modifies depending on the specific features of the noun in question. The intuition of Zarma native speakers, guided by convention makes it possible for them to determine or discriminate the choice of -o, -a and the null/zero variant of the determiner marker. One advantage of this proposal is that, it is not possible for personal pronouns and the definite article to co-occur, predicting that they are in complementary distribution. This explains the ungrammaticality of the following constructions. 6a. *írì kùsó 1pl water pot-Def b. * áí sánáá 1sg needle-Def The definite article inflection on the nouns is the cause of the non-convergent expressions in (6a & b); this is why the absence of such an inflection in (6c & d) makes the expressions to conform. 6c. írì kùsú 1pl water pot ‘our water pot’ d. áí sáná 1sg needle ‘my needle’ It is apparent from the illustrations in (6) that pronoun and determiner occupy the same structural position. Perhaps, it is also safe to say that determiner in Zarma is not considered as a nominal adjunct although it modifies nouns but does not co- occur with pronoun. This clearly indicates that determiner is essentially the element that projects and its property gives the resultant phrase its definite or indefinite reading. Thus, it is a functional category.

Page 148 Legon Journal of the Humanities Vol. 30.2 (2019) Jayeola, W.A./Definiteness in the Zarma determiner phrase Analysing the Basic Determiner Phrase The DP-hypothesis postulated by Abney (1987) as well as the LCA considers nominal category in the DP as a complement of D, however, the examples in (7) below suggest that the complement of D in Zarma occurs to its left contra LCA. This situation is not odd; it only shows language specific idiosyncrasies. Based on this fact, I recognise the variation in the linear order as the effect of movement. 7a. kùsó water pot -Def ‘the water pot’ b. hábó market-Def ‘the market’ c. ma ó name Def ‘the name’ d. sáná á needle Def ‘the needle’

The bare forms of the nouns in (7a, b) are kùsú ‘water pot’ and hábú ‘market’ respectively. Thus, it is speculated that the definite marker ó is base-merged with the nouns causing their pronunciations to be affected. On the other hand, examples (7c, d) show that definiteness externally merges with the noun to project into DP. Following from the aforementioned divisions, first, I consider the surface order of constituents within the basic DP in Zarma using examples (7c, d) as a movement of NP to the Spec DP because definiteness is realised as a free standing morpheme that externally merges with the noun. The analysis is presented in (8a) below using example (7d).

Legon Journal of the Humanities Vol. 30.2 (2019) Page 149 “The needle”

The bare forms of the nouns in (7a, b) are kùsú ‘water pot’ and hábú ‘market’ respectively.

Thus, it is speculated that the definite marker ó is base-merged with the nouns causing their

pronunciations to be affected. On the other hand, examples (7c, d) show that definiteness

externally merges with the noun to project into DP.

Following from the aforementioned divisions, first, I consider the surface order of

constituents within the basic DP in Zarma using examples (7c, d) as a movement of NP to the

Spec DP because definiteness is realised as a free standing morpheme that externally merges

Jayeola,with W.A./ Legonthe Journalnoun. of the The Humanities analysis Vol. 30.2 (2019) is presented in (8a) below using example (7d). 8a. 8a. DP

NP[+DEF] D’ sáná needle D[+DEF] á the

The configuration in (8a) shows that the NP sáná ‘needle’ starts as the The configuration in (8a) shows that the NP sáná ‘needle’ starts as the complement of the complement of the head D á ‘the’, but has to raise to a higher position than D to check its definite features in overt syntax. headThe D second á ‘the’ manner, but has of definitenessto raise to a markinghigher describedposition thanby the D data to checkin (7a, itsb) definite features in shows that the particle –ó ‘the’ is not suffixed to the nounkùsu ́ ‘water pot’ and hábú ‘market’overt butsyntax. seems to get fused (i.e. base-merged) with the nouns. This assumption relies on the fine-grained idea conceived within the minimalist syntax which posits that bound Themorphemes second (i.e. manner inflectional of definiteness and derivational) marking are describedcombined byby themerge. data in (7a, b) shows This is in line with the Inclusiveness Condition (IC) which states that a Logical Formthat object the particlemust be built–ó ‘the’ from theis not features suffixed of the to lexical the noun items k ofù stheú ‘water corresponding pot’ and hábú ‘market’ but initial numeration. It means that kùsó ‘the water pot’ for instance enters the derivationaboutseems definiteness fully to get inflected fused based (andi.e. carriesonbase its- merged) asemantic triple set with featureof features: the as nouns. wellsemantic, Thisas its syntactic assumptiphonological andon relies feature on which the fine is- phonological features. This being the case, kùsó ,for instance. has information about definitenessnotgrained identical idea based withconceived on itskù ssemanticú ‘waterwithin feature pot’the minimalistand as thewell likes.as itssyntax phonologicalFollowing which positsfrom feature the that whichtheoretic bound morphemesal explanations, (i.e. is not identical with kùsú ‘water pot’ and the likes. Following from the theoretical explanations,Iinflectional propose thatI proposeand N wouldderivational) that covertlyN would are covertlyraise combined to raiseD as bytoin D(8b)merge. as inusing (8b) This exampleusing is in example line (7a) with. The the justification Inclusiveness for (7a). The justification for our proposal comes from semantic evidence and the fact of morphologyourCondition proposal (IC) in comes syntax. which from states semantic that a evidenceLogical F andorm the object fact mustof mor bephology built from in syntax. the features of the 8b. lexical 8b.items of the corresponding DP initial numeration. It means that kùsó ‘the water pot’ for

instance enters the derivation fully inflected and carries a triple set of features: semantic, D[+DEF] N[+DEF]

kùsó syntactic and p honological features. This being the case, kùsó ,for instance. has information The analysis in (8b) describes how the syntactic component interacts with The analysis in (8b) describes how the syntactic component interacts with the Phonetic Form the Phonetic Form (PF) aspect of Zarma grammar. In10 summary, the descriptions (PF) aspect of Zarma grammar. In summary, the descriptions given have protected the

Pageposition 150 of the LCA which disallows right adjunctionLegon Journal and of the Humanitiesassumes Vol. 30.2 a (2019)head-complement as the

universal order.

Noun-Attributive Modifiers

Traditionally, adjectives and relative clauses are modifiers of nouns. As mentioned in the

previous section, relative clauses involve a D that embeds the relative complement clause.

Hornstein et al. (2005) describe relative clause as a sentence that can function as a kind of

giant adjective. This is shown in the examples below.

9a. zàárà kà Zouretou zèì à tórú cloth RelM Z steal Agr tear/damage ‘the cloth which Zouretou stole is torn/damaged’

b. àlùbòró kà ŋ̀wá dùndú zùrú kóí fúù man RelM eat yam run go house ‘the man who ate yam went away’

d. áí dí fúù kà Kadi cíná 1sg see house RelM K build ‘I saw the house which Kadi built’

It is clear from (9) that relative clauses are post nominal qualifiers in Zarma. However, it is

not the focus of this paper to go further on the properties or structural build-up of relative

clauses since the interest of this article is on the internal structures of adjectives and how they

11

Jayeola, W.A./Definiteness in the Zarma determiner phrase given have protected the position of the LCA which disallows right adjunction and assumes a head-complement as the universal order. Noun-Attributive Modifiers Traditionally, adjectives and relative clauses are modifiers of nouns. As mentioned in the previous section, relative clauses involve a D that embeds the relative complement clause. Hornstein et al. (2005) describe relative clause as a sentence that can function as a kind of giant adjective. This is shown in the examples below. 9a. zaára kà Zouretou zeì a tórú cloth RelM Z steal Agr tear/damage ‘the cloth which Zouretou stole is torn/damaged’ b. alùbòró ka ŋ̀wá dùndú zùrú kóí fúù man RelM eat yam run go house ‘the man who ate yam went away’ c. áí dí fúù ka Kadi cíná 1sg see house RelM K build ‘I saw the house which Kadi built’ It is clear from (9) that relative clauses are post nominal qualifiers in Zarma. However, it is not the focus of this paper to go further on the properties or structural build-up of relative clauses since the interest of this article is on the internal structures of adjectives and how they relate to the nouns they modify vis-à-vis the marking of definiteness. Consequently, the section that follows is devoted to the discussion of adjectives in Zarma. Zarma Adjectives One attributive modifier in Zarma is the adjective which in nominal expressions modifies the noun and describes some property of the thing referred to by the noun. Morphologically, adjective qualifiers in Zarma are all consonant initial and mostly consist of two syllables which have CVCV structure without any known derivational process. This is unlike Gungbe, Aboh (2004a), Turkish and Persian, Ghaniabadi et al. (2006), Chinese, Yang (2007), Yoruba, Ilori (2010), among others, where attributive adjectives are said to be derived. This category of modifiers has attributive characteristics that can be grouped into shape, colour, height, dimension etc. Attributive adjectives in Zarma essentially occur post nominally; a change of this order results in ill-formed or non-converging expressions. This is shown in (10).

Legon Journal of the Humanities Vol. 30.2 (2019) Page 151 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) 10a. i. tása berì plate big ‘a big plate’ ii. *berì Tása big Plate b. i. kùsú kéná water pot small ‘a small water pot’ ii. *kéná kùsú small water pot c. i. tírà chíre paper red ‘a red paper’ ii. *chíre tírà red paper d. i. kàmbà kúkù hand long/tall ‘a long hand’ ii. *kúkù kamba long/tall hand It is plausible to predict from the glosses of the data in (10) that indefiniteness as I earlier mentioned is not morphologically marked, although it is assumed to be present at the LF. Consequently, the merger of noun and adjective, using the examples in (10) is treated as a projection of two major categories which is headed by a null-D; reminiscent of what I said of indefinite article in the previous section. However, it is the position of this paper; as it will be argued later, the adjective, a class of modifier, is an adjunct because it can serve as a modifier of plural nouns. This is shown in the following examples. 11a. fafa-aí berì breast-Pl big ‘big breasts’ b. tása-aí Kená plate-Pl small ‘small plates’

Page 152 Legon Journal of the Humanities Vol. 30.2 (2019) Jayeola, W.A./Definiteness in the Zarma determiner phrase What is observed of the adjectives in (11) is not the same with categories such as quantifiers and numerals which Jayeola (2016) treats as functional items. As opposed to the situation in (11), quantifiers and numerals do not co-occur with plural morphemes within a DP. In another vein, numerals and quantifiers do not inflect whereas adjectives do. This phenomenon as it applies to Zarma willbe shown in the discussions below. A striking and intriguing scenario is created when the post nominal enclitic definite article occurs alongside the adjective within a DP. The gap between the noun and the adjective is widened because the determiner intervenes between the noun and the adjective. In this case, the head of the DP, enclitic D or a free standing D, assumed to occur phrase initially allows the raising of N, its complement, for phonological support. Consider the examples in (12). 12a. má ó berì name Def big ‘the big name’ b. tíra á berì paper Def big ‘the big paper’ c. wúndó tejí house-Def new ‘the new house’ I take the above examples (12a-c) as a projection of D because the definite article markers found in them actually account for the definite reading which the expressions carry. If for instance the free standing definite morpheme in (12b) or its suffixed variant in (12c) is removed, the resultant expressions will have an indefinite reading as shown in (13a and b) respectively. 13a. tíra berì paper big ‘a big paper’ b. wúndí tejí house new ‘a new house’ Noun-adjective constructions in Zarma still provide some other puzzling fact. The adjective presents a phenomenon where a suffixed determiner or determiner- like particle appears on it, translating to the possibility of adjective inflecting for

Legon Journal of the Humanities Vol. 30.2 (2019) Page 153 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) determiner. From this fact, it appears that Zarma attests to what has been described in the literature as determiner doubling. In this case, there is the double occurrence of either a free standing or suffixed determiner element which has superficially the same content. The uninflected form for bèrò is bèrì, it is yéná for yéyí and téjó for téjí. This is demonstrated in (14). 14a. má ó berò name Def big-Def. ‘the big name’ b. fafa berò breast big-Def. ‘the big breast’ c. háró yená water-Def. cold-Def. ‘the cold water’ d. wúndí tejó house new-Def. ‘the new house’ e. tíra berò paper big-Def. ‘the big paper’ A clear comparison of the examples in (12a-c) with the ones in (14b, d & e) presents a better understanding of the phenomenon of double definiteness. In those examples, only the adjectives inflect for definiteness making the appearance of the suffixed/enclitic definite article on the nouns optional. I do not regard this as an instance of agreement but a projection of another determiner phrase because the situation is not a case of an element taking on the morphosyntactic properties of another element. The omission of the two suffixed/enclitic determiners would give the expression an indefinite reading similar to the one earlier shown in (10 &12). Consequently, whenever a concrete reading is intended, one of the two enclitic definite articles is obligatory. Contrary to what I have said of the definite article, it is difficult to claim the existence of indefinite determiner doubling because it is not morphologically marked, this is unlike some Germanic varieties, Wood (2012), Northern Swedish, Delsing (1993). It is reasonable to mention that, in Zarma, the presence of double determiner does not license a change or reversal in the canonical order of Noun

Page 154 Legon Journal of the Humanities Vol. 30.2 (2019) Jayeola, W.A./Definiteness in the Zarma determiner phrase Determiner Adjective Determiner (NDAD). It is this condition that explains why the following expressions do not converge at spell-out. 15a. *berò kùsó big-Def water pot-Def b. *yená háró cold-Def water-Def What is indicated in (15) is quite different from the situation in Greek, (Alexiadou, 2006, and Alexiadou & Wilder, 1998), where both post-nominal and pre-nominal positions are possible; in the absence of Determiner spreading however, the order of constituents in its noun phrases is fixed or rigid.1 According to Aboh (p.c.), what is being taken as a case of determiner doubling in Zarma may be considered or viewed as an instance of plural agreement found in French. However, it is not the conviction of this study that the behaviours of the two (determiner doubling and plural agreement) bear some sense of resemblance with each other. A close look at the two instances predicts some degree of variations: in French, determiner, adjective and noun agree in terms of number as indicated at the footnote whereas Zarma does not display noun-adjective agreement both in terms of number as well as definite article indication. Therefore, it may not be plausible to see the two instances as products of an identical process2. One plausible approach to the treatment of N-article and Adjective-article expressions in Zarma is to analyse the enclitic article on the noun as nominal agreement marker or the spell-out of an agreement relation between a noun and its specifier in AgrP. However, analysing the inflection of noun and adjective as agreement relations will be an inconsistent representation of Zarma syntax, because it is understood from the discussion so far that Endocentric Concordial Pattern (Stockwell, 1977, p. 9) does not exist between nouns and their modifiers. Similarly, agreement of this nature does not also occur even in the IP-clause. Consider the examples below.

1 See Alexiadou and Wilder (1998: 1) for the detailed discussion of how determiner spreading can relax the order of elements – rigid or flexible. 2 The following examples represent the fact of plural spreading in French: a. Le grand garçon- ‘the big boy’ b. les grands garçons- ‘the big boys’ c. La belle maison- ‘the fine house’ d. les belles maisons- ‘the fine houses’

Legon Journal of the Humanities Vol. 30.2 (2019) Page 155 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) 16a. hárí ga yeyí water be cold ‘water that is cold’ b. háró ga yeyí water-Def. be cold ‘the water is cold’ c. * hárí ga yenó water be cold-Def. d. * háró ga yenó water-Def. be cold-Def. e. Kùsó ga berì water pot-Def. be big ‘the water pot is big’ f. * Kùsó ga berò water pot-Def. be big-Def. A close study of the examples in (14) and (16) shows that the two types of AP-modifier in Zarma viz: attributive (DP-internal modifier) and predicative (DP- external AP) are all generated to the right of the N, thus no divide is made between pre- and post-nominal APs. One other feature of adjectives conceivable from the glosses in (16) is that zero copula for predicate adjectives may not be possible. A plausible claim/proposal which I contemplate about the crashed expressions in (16) is that agreement is not allowed in predicative constructs. It seems that the auxiliary element (copula) gà ‘be’ blocks what could have been considered as the morpho-syntactic agreement relation between the modified noun and its modifier AP. This analysis predicts that the enclitic definite determiner on adjective is not a trait of agreement, if it were to be, the expressions ought to have converged. In conclusion, I recognise the situation as a case of double definiteness where the post-nominal definite determiner projects and heads the DP such that there is a second lower determiner projection that hosts the article-enclitic/ suffixed to the adjective. The derivation of this structure is analysed in the following section. Adjectival Modification and Determiner Doubling in Zarma What has been shown as regards adjectives and nouns inflecting for definite article in Zarma is not the same with what has been reported in the literature for many Niger-Congo languages, especially the Bantu sub-family which display noun-

Page 156 Legon Journal of the Humanities Vol. 30.2 (2019) Jayeola, W.A./Definiteness in the Zarma determiner phrase adjective agreement in terms of number. Thus, it may be out of place to consider the situation in Zarma as a case of definiteness agreement because empiricism does not show that the definite article should agree in its + definite feature (specification) since either of the noun or the adjective can be bare and the expression still clearly and unambiguously interprets as definite. Consider the following examples. 17a. kùsú berò water pot big-Def ‘the big water pot’ b. kùsó bèrì water pot-Def big ‘the big water pot’ c. tíraá bèrì paper-Def big ‘the big paper’ d. hárí yená water cold-Def ‘the cold water’ e. háró yeyí water-Def cold ‘the cold water’ From the data in (17), there appears to be no clue as to whether the uninflected/ bare nouns/ adjectives have [+ definite] specification or not because indefiniteness as earlier mentioned is not morphologically marked in Zarma. Invariably, the situation does not seem to be as clear as it is in Modern Greek, (cf. Alexiadou & Wilder, 1998; Simpson, 2001). The examples from Greek as presented by Simpson (2001, p. 130) illustrate the distinction and this is shown in (18). 18a. to meghalo to Ghermaniko to piano the big the German the piano ‘the big German piano’ b. *ena meghalo to Piano a big the Piano The phenomenon in (18) can be marked as an instance of definiteness agreement. This is because a noun with a definite determiner should always select

Legon Journal of the Humanities Vol. 30.2 (2019) Page 157 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) an adjective with a definite determiner also. This explains the reason for the ill- formedness of (18b). The situation here is largely different from the one found in Zarma. Of interest to the analysis pursued here is the fact that adjectival modification appears to tolerate a greater measure of freedom in the ordering of multiple modifier strings. This is strictly in line with Hawkin’s (1983) prediction. Consider the following examples. 19a. kùsó berò híbí water pot-Def big-Def black ‘the big black water pot’ b. kùsú híbí berò water pot black big-Def ‘the big black water pot’ c. tántábale kùare kená pigeon white small ‘a/the small white pigeon’ d. tántábale kùare beró pigeon white big-Def ‘the big white pigeon’ e. tántábale berò kùare pigeon big-Def white ‘the big white pigeon’ The expressions in (19) have not helped to show whether each adjectival modification can be recognised as an instance of adj-DEF modification because definiteness is not considered as being marked on some of them, at least overtly. However, in the spirit of Simpson (2001), I would like to postulate that definiteness or indefiniteness as the case may be, is present but not overt on some of the adjectives which probably explains why a greater degree of freedom in ordering is tolerated. This is similar to the situation in Mandarin reported by Simpson (2001, p. 150) and Greek, reported by Alexiadou and Wilder (1998, p. 325). These studies serve as a veritable support for adjectival modification as DP-hypothesis and a good guide to establish determiner doubling as a phenomenon in Zarma. Interestingly, it is obvious that attributive adjectives entirely follow

Page 158 Legon Journal of the Humanities Vol. 30.2 (2019) overtly at all. Since I have not been able to establish the existence of noun-adjective agreement nor assume definite article spreading in Zarma, it is reasonable to consider determiner doubling (definite) as a thriving phenomenon in the language.

An Analysis of Double Definiteness

Following from the foregoing assumption, two possible approaches to the analysis of adjectival modification are available: they can be analysed as a category of adjunct or as a Jayeola, W.A./Definiteness in the Zarma determiner phrase functionalthe modified head. DP and the adjectives optionally inflect for definiteness whereas indefiniteness is not marked overtly at all. Since I have not been able to establish the existenceThe surface of noun-adjective order of double agreement definiteness nor assume is better definite analys ed article using thspreadinge adjunction in Zarma, it is reasonable to consider determiner doubling (definite) as a thriving approachphenomenon because in the adjectives language. in this kind of projection cannot be considered as functional heads, rather I assumeAn an Analysis invocation of of Doublethe LCA whichDefiniteness makes it possible for the adjective to left adjoinFollowing to the from lower the D. foregoingThe assumption assumption, in this two regard possible is that approaches the post nominal to the article analysis of adjectival modification are available: they can be analysed as a category projectsof adjunct and or heads as a functionalthe DP where head. NP raises to D and that there is a second lower determiner The surface order of double definiteness is better analysed using the projectionadjunction thatapproach hosts thebecause definite adjectives article encliticin this tokind the of adjective projection where cannot the adjectivebe is considered as functional heads, rather I assume an invocation of the LCA which assumedmakes it topossible originate for at the the adjective complement to left position adjoin of to the the lower lower D. D. This The phenomenon assumption presents in this regard is that the post nominal article projects and heads the DP where NP anraises instance to D ofand DP that recursion there iswhere a second the upper lower DP determiner dominates projection the lower DPthat (cf. hosts Collins the , 2016, definite article enclitic to the adjective where the adjective is assumed to originate pp.at the 14 -complement15). In another position vein, t heof structurethe lower shows D. This that phenomenon the particle –presentsó ‘the’ is an not instance suffixed to the of DP recursion where the upper DP dominates the lower DP (cf. Collins, 2016, pp. adjective bèrì ‘big’; it rather gets fused (i.e. base-merged) with the adjective. Similar to an 14-15). In another vein, the structure shows that the particle –ó ‘the’ is not suffixed earlierto the adjectiveproposal, bbèèrìrò ‘big’; ‘the big’ it rather enters gets the fused derivation (i.e. base-merged) fully inflected with and thecarries adjective. a triple set of Similar to an earlier proposal, bèrò ‘the big’ enters the derivation fully inflected and features:carries a semantic, triple set syntactic of features: and semantic,phonological syntactic features. and As phonological suggested by features.the analysis As in (20), suggested by the analysis in (20), the head ADJ must be marked with definiteness thewhen head D isADJ definite must beand marked that a withdefinite definiteness D must have wh aen host. D is It definite means thatand ADJthat a covertly definite D must raises to D. have a host.20. It means that ADJ covertly raises to D. 20. DP

NP[+DEF] D' má DP D'

D[+DEF] D[+DEF] ADJ[+DEF] bèrò 20

Legon Journal of the Humanities Vol. 30.2 (2019) Page 159 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) The analysis provided in this regard does not involve a merger (internal/ external) of redundant elements; it rather makes it possible for every symbol of grammar to be interpreted. This is in strict compliance with the Principle of Full Interpretation. The compatibility of this analysis or assumption depends largely on: (a) the noun and its enclitic/suffixed article occupy the DP initial position, while (b) the adjectives follow the noun or occur in post-nominal position. The order in Zarma contrasts with the fact of Swedish where the adjectives occur pre-nominally and the adjectival article occupies the DP-initial position as shown in (21). 21 det store huset the big house-the ‘the big (Bernstein, 2001, p.549). house’ The situation in Zarma is not the exact mirror image of what obtains in the Swedish example. However, as previously emphasised, the proposal for a lower determiner is in line with the prediction of the LCA, Kayne (1994) as well as the position held by Delsing (1988, 1993), Santelmann (1993), among others. The non-marking of definiteness on some nouns and/or adjectives notwithstanding, the analysis is not affected; this study is guided by the assumption that a null functional head for either of the higher or lower determiner projections or both exist as definiteness may be encoded by other devices other than morphological as earlier explained. Finally, it is pertinent to mention that double definiteness account cannot be extended to constructions in which either of numeral, demonstrative or quantifier occurs with adjective to modify nouns. This situation accords with the Split-D- Hypothesis assumed in Aboh (2004a, 2004b, 2013). In line with this proposal, Giusti (2015, p. 70) asserts that the split DP provides an escape hatch for any kind of element, either AP (Adjective Phrase) or genitive DP. Thus, this present study contemplates a different account for the treatment of this type of construction which I take up in the following sub-section. Numeral/Demonstrative/Quantifier + Adjectival Modification The decision to provide a different account for constructions in which adjectives and numerals/demonstratives/quantifiers occur is premised on the following reasons: (a) the proposal that each of the definite determiner, numeral, demonstrative and quantifier is a functional projection and also in complementary distribution, (b) numerals, demonstratives and quantifiers are all free standing

Page 160 Legon Journal of the Humanities Vol. 30.2 (2019) Jayeola, W.A./Definiteness in the Zarma determiner phrase morphemes and not inflectional as definite determiner often tends to be. Given the above reasons, a different account is desirable in terms of how the expressions are derived as well as how they interact. The occurrence of an adjective between the noun and numeral/ demonstrative/ quantifier at surface structure as well as the absence of the enclitic definite determiner on the noun and the adjective as shown in the asterisked examples in (22) further confirms the claim earlier made that definite article and its subsets – numeral, demonstrative, quantifier do not overlap. In the light of this assertion, a different analysis is required to account for the surface linear order. 22a. tásà bèrì fó plate big one ‘one big plate’ b. kùsú kéná ísìtákí water pot small four ‘four small water pots’ c. wóí féjí híbí this sheep black ‘this black sheep’ d. hánsì bèrì híbí ódì dog big black That ‘that big black dog’ e. hánsì híbí bèrì ódì dog black big That ‘that big black dog’ f. *kùsó bèrì ódì water pot-Def big that g. *kùsú bèrò fó water pot big-Def one As I have earlier explained, examples (22f & g) in the above expressions crash because there are competitions among the different subsets of determiner elements (i.e. definite determiner and numeral) that occur within the same phrase (functional projection). To treat the adjective as an adjunct to the noun in line with the LCA, which predicts that the adjective is left adjoined to the noun and that the entire NP moves to Spec of NumrlP/DemP as the case may be, will not yield the correct linear order which Zarma exhibits. This is exemplified in (23b) using example (23a).

Legon Journal of the Humanities Vol. 30.2 (2019) Page 161 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) 23a. tásà berì fó/ódì plate big one/that ‘one/ that big plate’ b.

The structure in (23b) presents [Adjective-Noun-Numeral/Demonstrative] order contrary to the [Noun-Adjective-Numeral/Demonstrative] which Zarma tolerates. It therefore seems that the manner of adjunct approach which the LCA assumes is inappropriate. Another possible approach in this direction would be to treat adjectives like other modifiers which form a closed class in some languages as functional projections that introduce a DP complement. This approach has been adopted in works like Aboh (1999, 2004a); Cinque (2005); Ajiboye (2007) among others. In the spirit of their analyses, the DP first moves to Spec ModP and the ModP raises to Spec DemP; an approach that translates to what Aboh (1999, 2004a) regards as ‘Snowball/Roll-up’ movement. Snowballing, which allows pied-piping of a maximal projection continues until it reaches the Spec of the highest phrase because it depends on the number of intervening functional heads. Unfortunately, the snowball/roll-up approach fails to account for the peculiarity observed in the Zarma data; several intervening functional projections are not tolerated within its D-system, which is the reason for the adoption of the roll-up/snowball approach. In the sense of the fact of Zarma data, the functional approach does not seem to serve any useful purpose; therefore, I contemplate a revisit of the adjunct-based approach proposed by Dechaine (1993). I start the defence of this choice by mentioning that the reading provided by Numeral/Demonstrative plus adjective modification is not tied to or dependent on the adjective and so, it is not justifiable to treat adjectival modification asa

Page 162 Legon Journal of the Humanities Vol. 30.2 (2019) Jayeola, W.A./Definiteness in the Zarma determiner phrase functional head, similar to and in the spirit of the treatment of double definiteness. Instead of the functional approach, I shall in the spirit of Dechaine (1993), treat adjectives in this construction as a based-generated adjunct where the surface linear order is derived through the movement or raising of the entire NP to the Spec of NumrlP/DemP as the case may be. This is structurally represented in (24).

The resolve to retain the base-generated adjunct analysis is informed first, by its simplicity, a requirement that the data of Zarma also satisfies. In this regard, no movement of the assumed NP complement to Spec-AP is required to have the noun-adjective order which the language attests. It therefore does not seem to serve any useful purpose to appeal to what Chomsky (1995) refers to as Generalized Pied-Piping or Snowballing movement (Aboh, 2004a). Another justification for the refined analysis is linked to language internal evidence. Unlike what the roll-up/ snowball approach assumes, this paper does not deem it fit to stipulate a specific modifier sequence in Zarma. Considering the examples in (22), it is apparent that adjectives of different types; dimension and colour, for instance do not require any specific sequence of order. This is not the case with Gungbe, Aboh (1999, 2004a) and Buli, Sulemana (2012). Finally, it is not possible to analyse the situation under consideration in Zarma as a case or an instance of syntactic movement of the AP in the surface linear order. Besides, adopting the functional approach will be inconsistent with the account which has been given of the double definiteness and at the same time, it will indeed be unnatural. Conclusion The attempt in this paper has been to explain that definite article in Zarma is realised sometimes as an enclitic item on the noun and at other times as a freestanding item. Zarma definite articles may also be realised as a covert or a null

Legon Journal of the Humanities Vol. 30.2 (2019) Page 163 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) category. However, it is taken as the element that heads the projection in each case where the noun originates as its complement. The data used in this study show that definite determiner marker can also be base-merged with post nominal adjectives. The post-nominal position of the head of the DP on the one hand is understood as a product of phonological support or order of merge. On the other hand, the strong specifier feature of the head D’ must be checked before spell-out for the derivation to converge. On the basis of syntactic form and semantic interpretation, the study shows that Zarma definite determiners can occur with either nouns or adjectives or both. This means that adjectives in definite DPs make possible an additional (suffixed/base-merged) lexical determiner. In all of these cases, no different readings obtain. Similarly, it is understood that the definite determiner and its subsets numeral, demonstrative, and quantifier do not overlap. I consider the discussion of determiner doubling constructions in this paper as a great contribution to the ongoing debates on the syntax and semantics of the definite determiner and its subsets and how they interact which will likely translate into a technical proposal.

Page 164 Legon Journal of the Humanities Vol. 30.2 (2019) Jayeola, W.A./Definiteness in the Zarma determiner phrase References Abney, S. P. (1987). The English noun phrase in its sentential aspect. (Doctoral thesis). MIT. Distributed by MITWPL. Aboh, E. O. (1999). From the syntax of Gungbe to the grammar of Gbe. (Doctoral thesis). University of Geneva. Aboh, E. O. (2004a). The morphosyntax of complement-head sequences: Clause structure and word order patterns in Kwa. New York: Oxford University Press. Aboh, E.O. (2004b). Topic and focus within D. In L. Cornips, & J. Doetjes (Eds.), Linguistics in the Netherlands, 21, 1-12. Amsterdam: Benjamins. Aboh, E. O. (2013). The morphosyntax of the nominal left periphery. Handout, African Linguistics School III, Nigeria, July, 2013. Ajiboye, O. (2007). The syntax and semantics of Yoruba nominal expressions. The Linguistic Association of Nigeria in collaboration with M & J Grand Orbit Communications Limited, Port Harcourt, Nigeria. Alexiadou, A. (2006). On the cross-linguistic distribution of (in)definiteness spreading. ӦLT syntax workshop, Universitӓt Klagenfort. Alexiadou, A. & Wilder, C. (1998). Adjectival modification and multiple determiners. In A. Alexiadou, & C. Wilder (Eds.), Possessors, predicates and movement in the DP. (pp. 303-332). Amsterdam: John Benjamins. Bernstein, J. B. (2001). The DP hypothesis: identifying clausal properties in the nominal domain. In M. Baltin, & C. Collins (Eds.), The handbook of contemporary syntactic theory. (pp.536-561). Malden, MA: Blackwell Publishers. Cardinaletti, A. & Starke, M. (1999). The typology of structural deficiency: A case study of the three classes of pronouns. In H. V. Riemsdijk (Ed.), Clitics in the languages of Europe. (pp. 145-233). EUROTYP 20-5, Mouton de Gruyter, Berlin - New York. Cheng, L. L, & Sybesma, R. (1999). Bare and Not-So-Bare nouns and the structure of NP. Linguistic Inquiry, 30, 509-542. Chomsky, N. (1993). A minimalist program for linguistic theory. In K. Hale, & S. J. Keyser (Eds.), The View from Building 20: Essays in Linguistics in Honour of Sylvain Bromberger. (pp. 1–52). Cambridge, MA: MIT Press. Chomsky, N. (1995). The minimalist program. MIT. Cambridge, Mass. Cinque, G. (2005). Deriving Greenberg’s universal 20 and its exceptions. Linguistic Inquiry, 36 (2), 315-332. Collins, C. (2016). Introduction to minimalist syntax. The African Linguistics School Lectures. Cote-d’Ivoire, July, 2016.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 165 Jayeola, W.A./Legon Journal of the Humanities Vol. 30.2 (2019) Dechaine, R-M. (1993). Predicates across categories: towards a category – neutral syntax. (Doctoral thesis). UMass Amherst. Delsing, L. O. (1988). The Scandinavian noun phrase. Working Papers in Scandinavian Syntax 42, 52-79. Delsing, L. O. (1993). The internal structure of noun phrases in the Scandinavian languages. (Doctoral thesis). University of Lund. FassiFehri, A. (1999). Arabic modifying adjectives and DP structures. Studia Linguistica 53, 105–154. Ghaniabadi, S., Ghomeshi, S. & Sadat-Tehrain, N. (2006). Reduplication in Persian: A morphological doubling approach. Proceedings of the 2006 Annual Conference of the Canadian Linguistic Association. Giusti, G. (2015). Nominal syntax at the interfaces: A comparative analysis of languages with articles. Cambridge Scholars Publishing. Hawkins, J. (1983). Word order universals. New York, Academic Press. Ilori, J. F. (2010). Nominal constructions in Igala and Yoruba. (Doctoral thesis). Adekunle Ajasin University, AkungbaAkoko. Ondo State, Nigeria. Jayeola, W. A. (2016). A movement approach to word order variations in Zarma, a Nilo-Saharan language. (Doctoral thesis). University of Ibadan, Ibadan, Nigeria. Kayne, R. S. (1994). The antisymmetry of syntax. Cambridge, Mass. MIT Press. Longobardi, G. (1994). Reference and proper names: A theory of N-movement in syntax and logical form. Linguistic Inquiry, 25 (4), 609-665. Matthewson, L. (1998). Determiner systems and quantificational strategies: Evidence from Salish. Holland Academic Graphics, The Hague. Santelmann, L. (1993). The distribution of double determiners in Swedish: Den support in Do. Studia Linguistica, 47, 154-176. Simpson, A. (2001). Definiteness, agreement and the Chinese DP. Language and Linguistics, 2 (1), 125-156. Stockwell, R. P. (1977). Foundations of syntactic theories. New Jersey: Englewood Cliffs Inc. Sulemana, A-R. (2012). The structure of the determiner phrase in Buli. (M. PHIL. Dissertation). University of Ghana, Legon. Wood, J. (2012). Indefinite determiner doubling. Clauses/nominals workshop. University of Aarhus, December 14, 2012. Yang, G. (2007). Reduplication, atelicity and pluractionality. Paper presented at the City University of Hong Kong, August 2007.

Page 166 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019) DOI: https://dx.doi.org/10.4314/ljh.v30i2.8

Documenting Simpa: Advances in language documentation1

Nana Ama Agyeman Lecturer Language Centre University of Ghana, Legon, Ghana E-mail: [email protected]

Submitted: January 23, 2019/ Accepted: August 18, 2019/ Published: December 30, 2019

Abstract Documentary linguistics,2 also known as language documentation, a relatively new branch of Linguistics, advocates for the fundamental need to collect records of language use and practices in various forms from diverse genres for multiple purposes. Such purposes include language description, language development, language maintenance, and language revitalisation. Such a record of a language serves to feed not only linguistic research but also research in other disciplines, such as anthropology, history, and ethnography. Language documentation is recognized as an ultimate response to language endangerment. This paper explores language documentation with specific reference to Simpa, an under-described, minority language of Ghana. The paper reviews theories, approaches, methods, and tools of language documentation to highlight how they were employed and attuned to take care of the Simpa context. Thus, the discussion dilates on specific field methods and tools adapted for obtaining a balanced set of data from three complementary event types, viz., natural communicative events, staged communicative events, and elicitations, to build a language documentation corpus. Data processing, data annotation, and data management practices applied in building the corpus, as well as dissemination of the research outcomes are also addressed. Furthermore, fieldwork ethics used in the study are discussed. Finally, for consideration in future research, the paper reflects on some challenges that were encountered in documenting Simpa.

Keywords: language documentation, language description, language endangerment, Simpa, Efutu, Ghana

1 This paper is partly based on chapters from Agyeman (2016), a PhD thesis by the author. 2 Abbreviation: ELAR=Endangered Languages Archive, ELDP=Endangered Languages Documentation Programme, HRELP=Hans Rausing Endangered Languages Project, L1=first language/mother tongue, L2=second language, OCE=observed communicative event, OLB= observable linguistic behaviour, SCE=staged communicative event. UNSD=United Nations Statistics Division

Legon Journal of the Humanities Vol. 30.2 (2019) Page 167 Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019) Introduction This study discusses a documentation of Simpa, a South Guan language spoken in Winneba, a coastal town in the Central Region of Ghana. The paper attempts to define language documentation and its associated elements to relate them to the Simpa documentation experiences. The discussion begins with a definition and a detailed characterisation of language documentation. This is followed by a sociolinguistic depiction of Simpa. After that, a detailed description of the documentation of Simpa in terms of its contents, methods, tools, and outcomes is presented. Then, the paper outlines apparent constraints that were encountered. Finally, the paper concludes the discussion with some recommendations. Language Documentation

Definition Language documentation, also referred to as documentary linguistics, is a fairly new sub-discipline of linguistics. Discussions in language documentation dates back to the 1990s. Major research publications in the area include Himmelmann (1998), the Language Documentation and Description volumes which started in 2003, Gippert, Himmelmann, and Mosel (2006), Grenoble and Furbee (2010), and Boerger et al. (2018), among other publications. The field of language documentation could be said to be still evolving. Austin (2003, p.6) asserts that "language documentation as a research area and activity is not yet understood", especially "as distinct from collecting data for linguistic description". A similar view from Furbee (2010) suggests that it will be premature to consider language documentation as having a theory yet, because current definitions emphasize conducts and contents, and urge exemplary ethical and best practices instead of theory-building. Nevertheless, there is a sound delineation of what language documentation entails. For instance, language documentation is described as "a lasting, multipurpose record of a language" (Himmelmann, 2006, p.1). Furthermore, language documentation is distinguished from language description which is conceived as "a system of abstract elements, constructions and rules that constitute the invariant underlying structure of the utterances observable in a speech community" (Himmelmann, 1998, p. 166). Rather, language documentation primarily involves "the making and keeping of records of the world's languages and their patterns of use (Woodbury, 2003, p. 35); it "deals with creating multipurpose records of languages through audio and video recording of speakers and signers and with annotation, translation, preservation, and distribution of the resulting materials" (Austin, 2012).

Page 168 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Documenting Simpa: Advances in language documentation Language documentation thus, is concerned with the creation and maintenance of a comprehensive record of a language and language use in all conceivable genres and contexts (Himmelmann, 1998, 2006; Austin, 2003; Woodbury, 2003). Such a record may include all varieties of a language, depending on the specific goals of the documentation. Ideally, all facets of a language should be included in the record, nevertheless, the goals of a research may require a documentation of specific aspects of a language. Essentially, language documentation takes on ethnographic approach in the sense that language is observed through the lens of culture (Austin, 2010a; Hill, 2006; Franchetto, 2006; and Harrison. 2005). Furthermore, language documentation is fundamentally multi-disciplinary in nature (see again Austin, 2010a; Hill, 2006; Franchetto, 2006; and Harrison, 2005; see also Widlok, 2005; and Barwick, 2005). Above all, language documentation adheres to ethical standards (Austin, 2010a; Thieberger, 2007; and Dwyer, 2006). Language documentation is of immense value. Most significantly, it is a response to language endangerment (Hale, 1998; Crystal, 2003; Foley, 2004; Matras, 2005; Himmelmann, 2006). Many of the world's languages, especially minority ones, face endangerment, a situation where a language ceases to be in use because it is replaced by another language or all of its speakers have died. Language death could result from a natural disaster, such as an earthquake, or unnatural causes, such as war and migration, or as a result of repression from a dominant language (Kraus, 1992; Hale, 1998; UNESCO, 2003; Crystal, 2000; Dalby, 2003). Reasons for language documentation Although language endangerment is not a new phenomenon, concerns about it have been raised increasingly in recent times, especially in the last three decades. For instance, Hale (1998) projected that out of the estimated 6000-plus languages in the world, 3000 will perish and further 2400 will be near extinction in a couple of decades. A similar prediction states that at least a third of the world's languages will disappear in the 21st century (SIL, 2018). There is therefore a call to document languages, before any misfortune befalls them, because once a language becomes extinct due to any factor, such a language can no longer be documented. Languages, endangered or otherwise, should therefore be documented for posterity and for other good causes. But why at all is it necessary to document an endangered language; why document a language if it is going to die anyway? The answer to these questions is not farfetched. Indeed, linguistic diversity is crucial to human existence, similar to the way in which diversity in species is critical to the ecosystem. Arguably, the knowledge and belief systems of a people are encoded in their language. Their

Legon Journal of the Humanities Vol. 30.2 (2019) Page 169 Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019) beliefs and knowledge about governance, commerce, subsistence, socialization, healing, recreation, and indeed about every other aspect of life, are rooted in their language. Language is said to be the bedrock of a culture's world view and values, a key to individual and community identity (SIL, 2018). A similar view on the uniqueness and importance of individual languages is echoed in Hale (1998). Thus, when an undocumented language becomes extinct, the implication is that all such knowledge of the speakers die with it with no possibility of recovering such information. But when a language receives documentation, even if that language becomes extinct, the record from the documentation could be used for its revitalisation if so desired (see Raymond, 2007; Asfaha, Kurvers, & Kroon, 2007; Nathan & Csato, 2007; and Nathan & Fang, 2009). Besides revitalisation, the record of a language documentation can be used for language description, language planning, and language support and development activities. Furthermore, data from documentation corpora could be used in diverse linguistic enquiries, for instance, to test the validity or otherwise of theories. Beyond the linguistic arena, the record of a language can feed research in other disciplines, including the fields of anthropology, history, ethnography, and music, among other fields. Forms of data As previously stated, the main traditional methods of capturing primary data in language documentation are audio and video recording (Woodbury, 2003, p. 36; Austin, 2006, p. 89; Himmelmann, 2006, p. 1; Nathan, 2010b; 2009, 2006, Austin & Grenoble, 2007; Thiebberger & Musgrave, 2007, p. 29; Ashmore, 2008). Besides audio and video recording, still photos taken with a digital camera and field notes taken with pencil or pen on paper also form an important part of the primary data of any language documentation (Himmelmann, 2006, p. 1). Audio and video recorders, and digital cameras therefore, are indispensable equipment in language documentation (Woodbury, 2003). The quality of the primary data has to be of a high standard to the extent that the use of particular types of audio and video recorders are recommended, making the choice of equipment a crucial responsibility in language documentation (see Ashmore, 2008 and Nathan, 2010b). Some important recommendations in capturing primary data regard the use of external microphones rather than reliance on a recorder's internal microphone, as well as the use of a set of closed headphones for both audio and video recording. Overall, the format of the recordings should be of the right standard in order to achieve long preservation, ease of retrieval, and easy accessibility. Data from the records of a language may be classified into event types. A classification by Himmelmann (2006, pp. 7-11) provides two formats, namely,

Page 170 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Documenting Simpa: Advances in language documentation observable linguistic behaviour (OLB) and metalinguistic knowledge. The OLB data comprises samples of natural occurrences of "all kinds of communicative activities in a speech community" (Himmelmann, 2006, p. 7), including narratives, conversations, arguments, fights, gossips, sermons, songs, dispute settlement, political debates, selling and buying, and teaching and learning, among other discourse types. The metalinguistic knowledge format on the other hand records "the tacit knowledge speakers have about their language" mainly through elicitation (Himmelmann, 2006, p. 8). Metalinguistic knowledge data basically represents the native speakers’ "interpretations and systemizations of linguistic units and events" (Himmelmann, 2006, p. 8). Data on information regarding grammaticality and grammatical paradigms, phonological and morphological paradigms, varieties (for example, young speakers version versus older speakers'), usage and appropriateness, as well as folk taxonomies, all fall under metalinguistic knowledge. The two types of data, viz., data from OLB and that from metalinguistic knowledge are complementary: while the former represents evidence for how speakers actually communicate naturally in their everyday settings (Himmelmann, 2006), the latter provides an avenue for recovering certain aspects of the language that are not accessible from the analysis of OLB (Himmelmann, 2006. Another classification of primary data from language documentation recognizes three formats, namely observed communicative events (OCE), elicitations, and staged communicative events (SCE) (Lüpke, 2009). The main variation in the three types of primary data is attributed largely to their methods of collection, especially with regard to the level of influence of the researcher (Lüpke, 2009, p. 60). In collecting OCE data, the researcher has no influence whatsoever on the events being captured, except for their presence. The OCE data therefore consists of naturally occurring communicative events, including planned and spontaneous monologues and interactive discourse (Lüpke, 2009). The OCE data is comparable to Himmelmann's (2006) OLB data in terms of its naturalness and least influence of the researcher. Data from the elicitation method is "heavily influenced linguistically by and only created for the sake of the researcher" (Lüpke, 2009, p. 60). The elicitation method employs carefully designed linguistic tools, including questionnaires and other types of stimuli to collect wordlists, paradigms, and acceptability judgements (Lüpke, 2009). The elicitation data is comparable to Himmelmann's (2006) metalinguistic knowledge data in terms of contents and the heavy influence from the researcher. Lastly, data from the SCE derive from a blend of elicitation method with OCE method, with some influence from the researcher but not as much as in elicitations. The SCE method, as the label suggests, is staged

Legon Journal of the Humanities Vol. 30.2 (2019) Page 171 Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019) through the use of non-linguistic prompts such as pictures, videos, and games to obtain responses (Lüpke, 2009). Data annotation and metadata The primary data of a language documentation usually comes with annotations (Schultze-Berndt, 2006; McGill, 2009); raw data without any form of annotation may be meaningless and difficult to work with (Berge, 2010, p.55). The form of annotation to accompany the data is determined by the research goals, and in some cases, the archiver3. Common forms of annotation in language documentation include transcriptions and translations. Annotations could also be done with the help of various types of linguistic software (see McGill, 2009 and Schultz-Berndt, 2006). Metadata is another obligatory constituent of a language documentation (see Bergqvist, 2007; and Nathan & Austin, 2004). Metadata gives detailed description and explanation as well as access restrictions of the primary data and the annotations. Thus, without a metadata, the usefulness of a primary data is hugely reduced. Archiving is another obligation in language documentation (see Nathan, 2010a, 2008; Trilsbeek and Wittenburg, 2006; Johnson, 2004). In order to ensure proper preservation and accessibility, the language documentation corpus which comprises all the primary data with its annotations and metadata is archived. Players in language documentation may include several parties, each of whom assumes specific roles, responsibilities and expectations, based on the research goals. Austin (2003, p. 8) identifies at least six parties, namely, funding agency, archiver, researcher or research team, speaker community, user and general public. Sociolinguistic background of Simpa Simpa is spoken in Winneba, a coastal town in the central region of Ghana. It is one of Ghana's close-to-fifty4 indigenous languages5, and a minority one. It is officially called Efutu6 (also spelt Effutu), such that all documents, formal and informal, government or otherwise, use the name 'Efutu'. Together with Awutu and Senya, Efutu is classified as a dialect of Awutu which is further classified asa South-Guan language belonging to the Kwa branch of the Niger-Congo family of languages (Hall, 1983; Eberhard, Simons & Fennig, 2019). Absolute number of speakers of Simpa (and indeed for each of the two sister-dialects) is not known, however, a total number of speakers of all three dialects is estimated at 129,000, based on a 2013 UNSD survey (Eberhard, Simons & Fennig, 2019). 3 Further discussion on archive appears in section 4.5. 4 Ethnologue estimated the number of indigenous languages of Ghana at 68 and 73 in 2013 and 2016, respectively. However, the figure is drastically reduced to 48 in the 2019 update. It appears that in the recent publication of Ethnologue, some forms are considered as dialects and therefore are not counted separately. 5 ‘Language’ is used loosely here, although Simpa is considered as a dialect, as indicated later in this paragraph. 6 ISO 639-3: afu; 50 12' 0" North, 10 19' 0" West.

Page 172 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Documenting Simpa: Advances in language documentation Although speakers call their language as well as their township, clan, and people 'Simpa', outsiders call it Efutu, as indicated earlier. When asked for the reason for the difference in language name, speakers explained that the term 'Efutu' is an Akan7 expression with the meaning 'mixed up', and that non-Simpa people call the language so because they perceive the language as containing vocabulary from other languages. For this reason, the researcher decided to use the name Simpa, (alongside Efutu) as an attempt to bring attention to how the speakers call their language and also as a form of respect for the speakers. Simpa could be said to be somewhat threatened. In Winneba, the dominant language is Fante, a dialect of Akan, which is spoken in the inland parts of the town, while Simpa is spoken in the suburbs along the coast by predominantly fisher folks. (Welmers, 1974, p. 11; Agyeman, 2016, p. 31). Speakers of Simpa are bilingual in Simpa and Fante which they speak as a second language (L2). In the Simpa speech community and in Winneba in general, English, the official language of Ghana, and Fante are the languages used in primary and secondary schools both as media of instruction and as examinable subjects.8 Simpa however is prohibited in schools to the extent that its use, whether inside or outside the classroom, attracts punishment (see Agyeman, 2013, p. 268; Ansah & Agyeman, 2015, p. 97; Agyeman, 2016, p. 29). Simpa-Speaking children are therefore obliged to suppress their mother tongue in favour of the curricular languages in school. Not only is Simpa threatened but it is also under-described. Studies in the language include Welmers (1973), Forson and Gingiss (1977), Dolphyne and Kropp- Dakubu (1988), Boafo et al. (2002), Abaka (2006), Obeng (2008), Agyeman (2011, 2013a, 2013b, 2015, 2016), Ansah and Agyeman (2015), and Agyeman and Akrofi- Ansah (2013, 2016), with Agyeman (2016) representing the most detailed study of the language so far. The fact that Simpa is vulnerable in terms of language vitality has been pointed out in Welmers (1973), Abaka (2006), Agyeman (2013a, 2013b, 2016), and, Agyeman and Akrofi-Ansah (2013). Indicators for measuring language vitality include absolute number of speakers, population of speakers within a total population, intergenerational transfer, language policy and attitudes, availability of literacy materials, domains of use, and response to new domains and media, among other criteria (Kraus, 2000; UNESCO, 2003). In consideration of these indicators, Simpa fails to meet the standards in many regards. For example, the absolute number of Simpa speakers cannot be that substantial, considering the fact that the estimated figure of 129,000 includes speakers of all three dialects of Awutu. Moreover, the

7 Akan (ISO 639-3: aka), also a Kwa language, is a linguistic neighbour of Simpa, and a dominant language in Winneba and indeed, in Ghana. 8 This is as a result of Ghana's language of education policy which prescribes the use of English, the official language, in combination with one indigenous language for basic education. For this reason, 11 dominant languages from various regions have been developed for the purpose. See Agyeman (2013), Ansah (2014), and Ansah and Agyeman (2015).

Legon Journal of the Humanities Vol. 30.2 (2019) Page 173 Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019) population of Simpa speakers, in comparison with the total population of Winneba is marginal, as intimated above. Furthermore, although there is intergenerational transfer, children ultimately reduce their Simpa speaking rate or stop speaking it altogether in favour of the school languages, due to school language policy and attitudinal factors (again, see Agyeman, 2013a, 2013b, 2016; Ansah & Agyeman, 2015). Furthermore, there is not much response to new domains and media, for instance, Simpa songs are rare, and the only available public media experience is a one-hour-per-week radio programme by the community radio station, Radio Peace 88.9 FM. To add to the above points, there are no materials for Simpa language education and literacy. All these outlined factors work together to render Simpa vulnerable, hence the need for documentation. Documentation of Simpa

Preamble The Simpa documentation project was undertaken in the Linguistics Department of the School of Oriental and African Studies (SOAS), University of London, as part of a PhD study. The research was funded mainly by the Endangered Languages Documentation Programme (ELDP)9, and partly supported by the Office of Research, Innovation and Development (ORID), University of Ghana. The fieldwork was carried out in Winneba, Ghana, the home of the Simpa speakers, and participants were mostly fish workers, including fishermen and fish sellers, all of whom were native speakers. Other participants included school teachers and heads of schools, and traditional leaders. The immediate research goal was to document Simpa and use the annotated corpus to describe aspects of the grammar of the language, as well as depositing the annotated corpus in an archive. Ultimate goals included further descriptive analyses of the language and the development of practical materials, including wordlists, primers, and a dictionary, for the speech community as well as the linguistic research community. The sections that follow discuss various aspects of the documentation of Simpa. Language Consultants As noted by Austin (2003), the speaker community constitutes a key player in language documentation, as language consultants are recruited from this community. In this study, the term language consultants, sometimes described as participants or respondents, refers to speakers who participated in the fieldwork. Their role included supply of data and its annotation, as well as corroboration of the annotations. They were selected randomly and recruited through negotiation. Most

9 The project award ID is 'ELDP: IGS-0096'

Page 174 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Documenting Simpa: Advances in language documentation importantly, their informed consent was sought before engaging their services, in conformity with ethical standards (see Dwyer, 2006; Austin, 2010). In this process, the researcher explained their exact role and the prospect of earning remuneration upon participation. Also, the fact that their acceptance to participate is voluntary and that they could withdraw their participation at any point was explained. Besides, the intended uses of the data and other possible uses were made clear. In this research, consent was mostly verbal and was recorded on audio or video10. The interactions with the consultants were conducted in Fante. Equipment In line with recommendations11, equipment choice in this study was based not only on the research goals but also on the fieldwork environment. Thus, the windy and the noisy nature of the Simpa fishing beach where most of the recordings took place, as well as the frequent interruption of electricity power supply12, all informed equipment choice. Specific equipment included Toshiba satellite laptop, Asus Eee netbook laptop, Zoom h4n audio recorder, Canon hg10 video camera, Canon Ixus 220 digital camera for still pictures, three different microphones, Rode Deadcat windshield, Rode blimp windshield, and a set of close headphones. The microphones included a Rode NTG2 Shotgun, a cardioid microphone and a lavalier microphone. These varieties of microphones were included to cater for different recording situations. The choice of the Zoom h4n13 and the Canon hg10 was helpful because they had outlets for external microphone and headphones14. The blimp windshield was added especially because of the Simpa recording environment; it was used for recordings at the beach and other outside environment in order to reduce the extreme background noise in the recordings. The Asus Eee netbook was chosen especially for its long battery life, in case of electricity power cut, and also for its portability. Other equipment included a tripod for the video camera and a mini tripod for the audio recorder. Essential consumables, such as batteries and SD cards for the recorders, a card reader, and an external hard drive were not left out. A life jacket was acquired for precaution. A peli case was included for transporting and protecting the equipment.

10 See Dwyer (2006, p. 39) for advice on verbal consent. 11 See discussion in section 2. 12 The fieldwork was done at a time when Ghana had a major electricity power crisis. 13 See a review of the Zoom h4n by Bernard Howard at www.hrelp.org/archive/review/zoom_h4n_review.html 14 See section 2 for discussion of recommendations on the use of external microphones.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 175 Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019) Data: Types, tools, methods, and annotation Issues on data have been briefly outlined. In this section, the discussion focuses on the types of data as well as their collection tools and methods in the documentation of Simpa. Primary data in this study included four main types of media materials: video, audio, still photos, and field notes. In terms of event types, the data included all three traditional types, namely, OLB or OCE, metalinguistic knowledge or elicitation, and SCE15. The following subsections expatiate on data. Elicitation Elicitation or metalinguistic knowledge data in this research relied on four main tools: (i) Ibadan wordlist of 400 basic items16, (ii) tense, aspect, mood, and polarity (TAMP) questionnaire, (iii) pronoun questionnaire, and (iv) serial verb construction (SVC) questionnaire. These elicitation questionnaires were administered with the help of trained speaker-consultants. The Ibadan wordlist contained 400 English vocabulary items which are common to most West African cultures such that equivalent vocabulary exists in their languages. The vocabulary items were from different domains, including body parts, household objects, food ingredients, farm tools, kinship terms, cardinal numbers, verbs, and adjectives. The main method of administration of the wordlist was by mentioning the word in English first, followed by its Akan equivalent, after which the respondent supplied the Simpa equivalent. Thus, Akan was used as an intermediary language while English and Simpa functioned as the source and target languages, respectively. The elicitation sessions were recorded on audio. A couple of respondents who could read and write17 opted to complete the list on their own, where they used Akan writing system. In addition, the list was administered to them orally for audio recording18. The second elicitation tool, viz., the TAMP questionnaire, was designed based on a framework by Dahl (1985). Dahl (1985) hypothesizes the existence of a set of cross-linguistic TAMP categories to which language-specific categories can be assigned, and proposes prototypical contexts in which a given TAMP category may occur cross-linguistically. In creating the TAMP questionnaire for this research, 31 verbs were selected from the Ibadan wordlist and some elicitation data19 to compose sentences in English which were then translated into Simpa with the help of speaker-consultants. In administering the TAMP questionnaire, respondents 15 See section 2 for discussion of the various types of data. 16 This is a common but an unpublished wordlist used by many language documentation researchers. Further description of the wordlist appears in the next paragraph. 17 Both consultants had tertiary education. 18 The oral method was useful for ascertaining tone in words 19 This was data from the video watching and description.

Page 176 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Documenting Simpa: Advances in language documentation were tasked to reproduce each sentence based on a specified context. Each verb was used in all the various contexts. For each sentence, the negative equivalent was also elicited. Each response was repeated at least twice for confirmation. These elicitations were recorded on audio, alongside note-taking with pencil and paper. The third elicitation tool which collected data on pronouns was self-designed. The pronoun questionnaire included a set of carefully constructed English sentences with highlighted nouns and noun phrases. In this elicitation, respondents were tasked to initially produce a Simpa translation of the English sentence. Once the Simpa translation was rendered, the respondent was further asked to repeat the Simpa sentence in which they substituted pronouns for highlighted nouns and noun phrases. These elicitations were audio-recorded. The fourth tool, that is the SVC questionnaire, listed SVCs in Akan20. In this elicitation, respondents were tasked to render Simpa equivalents of the Akan sentences. The sessions were audio-recorded. For the various elicitation methods, metadata was recorded at the beginning and also at the end of each elicitation session. Metadata at the beginning included location and date of recording, equipment used for the recording, consultant's name, age, occupation, hometown, L1, L2, and other languages. Metadata at the end included access restrictions where the consultant was asked to indicate whether any part of the data could not be published, and whether any persons or groups may not access the data. Observable Linguistic Behaviour (OLB) In this study, OLB or OCE21 included conversations, arguments, jokes, and other forms of interactions among speakers. Some of the OBL recordings involved more activities with minimal or no speaking; they included activities like fish preparation (including cleaning, washing, salting, drying, smoking, etc), net mending, boat mending, boat painting, and repairing of outboard motors. One major event capture in the OLB data was a celebration of a local festival, viz., Petu festival, which involved various activities, including durbars, processions, drumming, singing, dancing, gun firing, funfairs, and several rituals. The OLB data were recorded on video and audio; events involving less speech and more activities were recorded on video while those involving more speech, such as conversations and arguments, were recorded on audio, sometimes coupled with video. For the OLB data, all metadata were collected only after the recording. Another type of naturally occurring data included in the corpus were recordings from a radio programme by the community radio station, Radio Peace, 20 The Akan SVCs were mainly from Agyeman (2002; 2003), Kambon (2013), and Osam (1997). 21 See section 2 for definition of OLB/OCE.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 177 Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019) 88.9 FM. The programme, which was done in the Simpa language, discussed social and political issues affecting the community. The discussion was carried out by a host and a panel of guests, and it included listeners' participation. This data was classified as natural data because of its occurrence being independent of the research. Staged communicative events (SCE) One of the tools used to collect SCE data was a non-verbal stimulus in the form of a picture book by Mayer (1969) with the title Frog Where Are You? The aim of this exercise was to construct a story based on the pictures in the book. In this method, the consultant initially previewed all the pictures. After that, he went over the pictures again from the beginning to construct a story about the pictures. The whole process was recorded as one session in both video and audio. Another method for collecting SCEs yielded relatively short narratives termed as 'prompted narratives'. In this method, consultants were asked to talk about a given topic, for instance, to describe the various fishing equipment and their functions, or how to process salted fish, etc. In some cases, consultants talked about their preferred topic rather than the suggested one. Data from the SCE methods were predominantly monologues. Another SCE tool which involved video watching and description, was self- designed and it targeted complex predication, especially, verbal constructions, including serial verb constructions (SVCs)22. In this method, some of the videos from the natural events (OLB/OCE)23 were selected based on their contents. The selection included videos with high content of activities, such as videos of women cleaning, cutting, washing, and arranging fish on a mesh for smoking; children playing different games at the beach; fishermen conversing, sewing or mending nets; fishermen launching or docking a boat, among such other videos. These videos were played back for consultants to watch and describe the activities. The sessions were recorded on audio and video. Finally, folk songs and stories were also recorded and they were classified under SCEs. Each of the various SCEs was recorded in both video and audio. Metadata was collected at the end of the session. The above-described SCE methods could be differentiated from the OLB and the metalinguistic knowledge methods. Unlike OLB which occurred naturally without any prompts from the researcher, data from the SCEs were produced upon the request or prompts from the researcher. Nevertheless, there was a level of flexibility with the SCE methods in the sense 22 One of the research goals was to analyse SVCs in Simpa hence, their elicitation. 23 See section 2., above, and also section 4.4.2., below, for discussion of OLB/OCE.

Page 178 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Documenting Simpa: Advances in language documentation that the consultant chose or manipulated the type of responses to give, unlike in metalinguistic knowledge method in which consultants were somewhat restricted to supply only certain kinds of responses. As mentioned earlier, still photographs were included in the corpus. Images in the photographs included artefacts, fishing tools, fishing boats, fish species, consultants and other participants, and scenes from places like the fishing beach and the fish market. In addition, interviews were conducted among community members, including traditional leaders, school teachers, and church leaders. Such interviews investigated matters on topics like language use, language attitudes, language ideologies, aspects of Simpa history, and linguistic neighbours. Data annotation In this study, for practical reasons, not every data in the corpus received annotation. For instance, data with unclear sound quality were excluded. Moreover, due to time factor, only a selection based on the researcher's discretion was annotated. Thus, data that would serve the immediate needs of meeting the research goals were given priority. Furthermore, for technical reasons, annotated data included only audio recordings (see further details below). Data annotation mainly involved transcription and translation. Linguistic software used for data annotation included Transcriber (a tool for annotating speech signals), ELAN (a tool annotating audio and video recordings), and Toolbox (a tool for building lexical data, and for parsing and interlinearizing texts)24. On data from elicitation, phonetic transcription involving the use of the International Phonetic Alphabet (IPA) with tone marking, and free translation were done manually with pencil and paper by the researcher. For all other data types, i.e., OLB and SCE, transcription was initially done electronically in Transcriber by a trained native speaker from the community. In this process, the annotator listened to the Simpa recordings in Transcriber through a set of closed headphones to transcribe them with the use of Akan writing system since Simpa has no standard writing system. This was followed by a free translation of the transcribed Simpa data into English with pencil and paper by the trained native speaker, viz, the annotator. After that, a second level transcription of the Simpa data in the form of phonetic transcription was done manually with pencil and paper by the researcher, through listening from a set of closed headphones and with the help of the native speaker's transcription. This was necessary in order to get the right pronunciations of words for accurate transcription. The next level of annotation involved interlinearisation in Toolbox after importing the Transcriber file into Toolbox for interlinear glossing

24 See SIL international for further description of these linguistic tools.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 179 Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019) into English. This was made possible because the Toolbox project had earlier been entered with lexical data of Simpa with their English glosses. Thus, it was possible for Toolbox to parse the Simpa sentences into their basic constituents with their corresponding English glosses. ELAN was used minimally, mainly for metadata rather than data annotation. To ensure accuracy, all transcriptions were read out to native speaker consultants for confirmation. A summary of the various data types and subtypes in audio files and their associated Transcriber files, as well as their respective total length in minutes is presented in Table 1. Table 1: Summary of Audio Data

Event Type Audio Files Transcribed Files

No. of Files Total No. of Total Length in Files Length in Mins. Mins. Elicitations Ibadan wordlist 5 218 TAMP questionnaire 46 152 Pronoun 12 42 questionnaire SVC questionnaire 36 302 Observable Natural events 16 152 2 13 Linguistic Radio programme 2 109 1 16 Behaviour (OLB) Stages Frog story 1 8 1 8 Communicative Prompted narratives 40 283 14 111 Events (SCE) Video watching and 11 133 2 12 description Folk songs and 10 27 5 19 stories

Archiving and dissemination Archiving and dissemination of the documentation outcomes are major responsibilities in language documentation. In this study, the corpus from the Simpa documentation, which consisted of all the primary data and their annotation25 was deposited with the endangered languages archive (ELAR), a digital archive. As a requirement from ELAR, the only acceptable format of metadata is CMDI maker

25 All transcriptions in Transcriber were migrated into ELAN, another linguistic software, before archiving. This became necessary, as CMDI maker, the only acceptable metadata format does not support Transcriber files.

Page 180 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Documenting Simpa: Advances in language documentation metadata, a web tool for generating a scientific metadata (see ELAR website for further description of CMDI maker)26. Thus, all data in the entire corpus were formatted in CMDI maker metadata for digital archiving. Such a digital archive enhances accessibility of the deposit and constitutes a significant way of dissemination. Besides ELAR, the researcher is making arrangements to deposit copies of the primary data elsewhere, especially in the speech community. This will be done in consultation with speech community members to ensure that the materials are accessible to them in the most convenient format. Besides the archive deposit, data from the corpus was used to write a descriptive grammar of Simpa as a PhD thesis (see Agyeman, 2016). In addition, in consultation with the speech community, data from the corpus was used to produce a booklet on Simpa fishing beach for the community members. The booklet contains pictures of the fishing beach and fishing tools, with short sentences in Simpa describing the pictures. Other practical materials for the community include an ongoing project of a tri-lingual dictionary – a Simpa-Fante-English dictionary. Furthermore, findings from the research have been disseminated through publication in the form of journal articles and also through conference and seminar presentations (See Agyeman, 2011, 2013a, 2013b, 2015, 2016, 2017; Ansah & Agyeman, 2015). Also, with support from the College of Languages Education of the University of Education, Winneba (UEW), a couple of language and literacy workshops which advocate the use of Simpa in schools have been organised for teachers in basic schools in Winneba. Field ethics Linguistic research, like research in any other discipline, must observe ethical principles (see Dwyer, 2006; Austin, 2010; Rice, 2010). In documenting Simpa, the researcher observed field ethics. Before starting the fieldwork, permission was sought and obtained from leaders in the community. These leaders included the chief of Winneba, the leader of the fishermen, the founder and manager of the community radio station (Radio Peace FM), and the dean of the faculty of Ghanaian languages of UEW. In conducting the fieldwork, informed consent was sought and obtained from all consultants. Besides, all consultants were duly remunerated upon completion of their tasks. Furthermore, as a form of giving back to the community, multiliteracy workshops were organized for a group of school teachers in the community.27 Moreover, the production of the booklet on Simpa fishing was also a form of giving back to the community.

26 All Transcriber files, video files and audio files were imported into CMDI maker through ELAN software. 27 The multiliteracy workshops constituted a way of addressing the issue of prohibition of Simpa speaking in schools. They were organized in conjunction with Dr Ari Sherris, a Fulbright visiting scholar and expert in multiliteracy matters.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 181 Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019) Reflections There were a few matters that are worth reflecting on. One of such matters involved an apparent language barrier in certain cases. Although most speakers speak and understand Fante, as intimated earlier in the paper, there were a few participants with very low proficiency and comprehension in Fante. This necessitated the hiring of an interpreter who helped with translation and interpretation in those cases. Another point of reflection concerns some belief systems in the community. One of such beliefs relates to the prohibition of fishing on Tuesdays. It was and is still believed that engaging in fishing on a Tuesday is a taboo, a violation of which will lead to a misfortune. Nevertheless, one could go to sea to catch crabs on Tuesdays. Another taboo regarded filming or photographing a pregnant woman. Thus, the researcher was confronted at one point for violating this taboo28. These and other belief systems in the community could have implications for or even pose a challenge for researchers. For instance, the researcher became more cautious in videoing and photographing women, in case they were unnoticeably pregnant. On the positive side, the researcher chose Tuesday as one of the days for visiting the fishermen for interviews, consultation, and other interactions. Another point of reflection regards software, specifically, ELAN. Although the software had very useful features, overall, it was complicated to use, in comparison with other software. It will be worthwhile updating future versions of the software to make it more user-friendly. Another point of reflection regarded the nature of the environment in relation to the equipment, including the recorders and laptops. As a result of the sandy and humid nature of the beach, the researcher was constantly concerned about the protection of the equipment.29 It would be worthwhile to invest in, say, water resistant equipment in such an environment, if one’s project budget allows it; or, at least, in a less costly alternative, such as a plastic cover to protect the equipment. Conclusion The field of documentary linguistics is still evolving. Language documentation, as conceived in the literature and in practice attests to be a starting point of addressing language endangerment, besides other gains. Nevertheless, language endangerment is widespread in many parts of the world, including Ghana. Out of the approximately fifty indigenous languages in Ghana, many of them are in a similar or even worse situation as Simpa in terms of vitality. Meanwhile, most indigenous languages remain undocumented. Even, the dominant ones, including the

28 The researcher explained to the accusers that it was only a myth and went further to show them from her laptop pictures of herself during pregnancy and pictures of the child from that pregnancy as a form of assurance. 29 The equipment survived the project anyway, nevertheless, maintaining its security created a constant anxiety for the researcher.

Page 182 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Documenting Simpa: Advances in language documentation 11 institutional languages have not received adequate documentation, as conceived in documentary linguistics. Notwithstanding, the importance of linguistic diversity cannot be overemphasized. Although these facts paint a gloomy picture, all hope is not lost, as it may seem. For instance, support for language documentation is available from some governments and organizations. One such organisation is ELDP which provides support in the form of training through workshops in various parts of the world30. ELDP also provides various categories of research grants for documentation of endangered languages across the world (see Austin, 2010b). Another way of approaching the issue is for language and linguistics departments to include the study of language documentation in their programmes to train students to take up responsibilities of documentation. Besides, community members could be mobilized to document their language. With the outcomes of the documentation, further actions could be taken to develop languages in order to reverse the trend of language endangerment.

30 A number of such workshops have been organized in West Africa, with three of them taking place in Ghana in the past decade.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 183 Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019) References Abaka, E.. N. (2006). The Efutu vocalic phonology. A paper presented at the Legon Trondheim Linguistics Project Colloquium on the Typology, Lexicography and Development of the Languages of the Volta Basin. Accra, Linguistics Department, University of Ghana. 9th – 13th Jan. 2006. Agyeman, N. A. (2016). A descriptive grammar of Efutu (southern Ghana) with a focus on serial verb constructions: a language documentation study. (Doctoral Dissertation). SOAS, University of London. Agyeman, N. A. (2015). Agreement in Efutu (Simpa) - a language documentation study. Paper presented at the first School of Language Conference (SOLCON-1), University of Ghana, Legon, 27-29 October 2015 Agyeman, N. A. (2013). Language use in Winneba - some preliminary observations. In Youngberg, Connor & Laura Kipp (Eds). SOAS Working Papers in Linguistics, 16. 259-277. Agyeman, N. A. (2013). Language Policy of Education in Ghana: Implications for minority languages. Paper presented at the Language Endangerment: Language Policy and Planning Conference. CRASSH, Cambridge. 26th July 2013 Agyeman N. A. (2011). Serial verb constructions in Efutu. Paper presented at the 2011 International Workshop on the Description and Documentation of the Ghana-Togo-Mounting Languages, Ho, Ghana Agyeman, N. A. (2003). On object sharing and referent sharing in Akan serial verb constructions. Studies in the Languages of the Volta Basin, 1, 3-8. Agyeman, N. A. (2002). Serial verb constructions in Akan. (MPhil thesis). Department of Linguistics, NTNU, Trondheim. Agyeman, N. A. & Akrofi-Ansah, M. (2013). Language use in two South-Guan language communities in Ghana: Implications for language documentation. Paper co-presented at the Institute of African Studies Seminar Series, University of Ghana Agyeman, N. A. & Akrofi-Ansah, M. (2016). Vowel harmony systems in two South-Guang languages (Efutu and Leteh): A case of dominant language influence. Paper presented at the 9th Linguistics Association of Ghana (LAG) Conference, GILLBT, Tamale, 26-28 July 2016 Ansah, G. N. (2014). Re-examining the fluctuations in language in-education policies in post-independence Ghana. Multilingual education, 4 (1), 12.

Page 184 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Documenting Simpa: Advances in language documentation Ansah, M. A. & Agyeman, N. A. (2015). Ghana language-in-education policy: the survival of two South-Guan minority dialects. Per Linguam, 31(1), 89-104. Asfaha, Y. M., Kurvers, J. & Kroon, S. (2007). Literacy use and acquisition in multilingual Eritrea. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 4, pp. 195-206). London: SOAS. Ashmore, L. (2008). The role of digital video in language documentation. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 5, pp. 77- 102). London: SOAS. Austin, P. (2012). Language Documentation. obo in Linguistics. DOI: 10.1093/ obo/9780199772810-0075. Austin, P. K. (2010a). Communities, ethics and rights in language documentation. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 7, pp. 34-54). London: SOAS. Austin, P. K. (2010b). applying for a language documentation research grant. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 7, pp. 285-299). London: SOAS. Austin, P. K. & Grenoble, L. A. (2007). Current trends in language documentation. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 4, pp. 12-25). London: SOAS. Austin P. K. (2006). Data and language documentation. In Jost Gippert, Nikolaus P. Himmelmann and Ulrike Mosel (Eds.). Essentials of Language Documentation (pp. 87-112). Berlin: Mouton de Gruyter. Austin, P. K. (2003). Introduction. In Peter K. Austin (Ed.). Language Documentation and Description, (Vol. 1, pp. 6-14). London: SOAS. Barwick, L. (2005). A musicologist's wishlist: come issues practices and practicalities in musical aspects of language documentation. In Peter K. Austin (Ed.). Language Documentation and Description, (Vol. 3, pp. 53-62). London: SOAS. Bergqvist, H. (2007). The role of metadata for translation and pragmatics in language documentation. In Peter K. Austin (Ed.). Language Documentation and Description, (Vol. 4, pp. 163-173). London: SOAS. Boerger, Brenda H.; Moeller, Sarah Ruth; Reiman, Will; Self, Stephen (2018). Language and culture documentation manual

Legon Journal of the Humanities Vol. 30.2 (2019) Page 185 Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019)) Crystal, D. (2003). Endangered languages: What should we do now? In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 1, pp. 18-34). London: SOAS. David, C. (2000). Language death. Cambridge: Cambridge University Press. Dahl, O. (1985). Tense and aspect systems. Oxford: Blackwell. Dalby, A. (2003). Language in danger: The loss of linguistic diversity and the threat to our future. Columbia University Press. Dolphyne, F. A. & Kropp-Dakubu, M. E. (1988). The Volta-Comoe languages. In Kropp- Dakubu, M. E. (Ed.) The languages of Ghana (pp. 50-90). London: Kegan Paul International for the International African Institute. Dwyer, A. M. (2006). Ethics and practicalities of cooperative fieldwork and analysis. In Jost Gippert, Nikolaus P. Himmelmann and Ulrike Mosel (Eds.). Essential of Language Documentation (pp. 31-66). Berlin: Mouton de Gruyter. Eberhard, D. M., Simons, G. F., & Fennig, C. D. (2019). Ethnologue: Languages of the world. 22nd edition. Dallas, Texas: SIL International. Online version: http://www.ethnologue.com. Foley, W. A. (2004). Language endangerment, language documentation and capacity building: Challenges from New Guinea. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 2, pp. 28-38). London: SOAS. Forson, B. & P. J. Gingiss. (1977). Efutu. In Kropp-Dakubu, M. E. (Ed.). West African Language Data Sheets 1. (pp. 180-185). Furbee, N. L. (2010). Language documentation: Theory and practice. In L. Grenoble & N. L. Furbee (Eds.). Language documentation: Practice and values (pp. 3-24). Amsterdam: John Benjamins. Gippert, J., Himmelmann, N. P., & Mosel, U. (Eds). (2006). Essentials of Language Documentation. Berlin: Mouton de Gruyter. Grenoble, L. & Furbee, N. L. (2010) Language documentation: Practice and values. Amsterdam: John Benjamins. Hall, E. (1983). Ghanaian languages. Accra: Asempa Publishers. Hale, K. (1998). On endangered languages and the importance of linguistic diversity. In Lenore A. Grenoble and Lindsay J. Whaley (Eds.), Endangered Languages; Language Loss and Community Response (pp. 192-216). Cambridge: Cambridge University Press.

Page 186 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Documenting Simpa: Advances in language documentation Harrison, K. D. (2005). Ethnographically informed language documentation. In Peter K. Austin (Ed.), Language Documentation and Description (Vol. 3, pp. 22-41). London: SOAS. Hill, J. H. (2006). The ethnography of language and language documentation. In Jost Gippert, Nikolaus P. Himmelmann and Ulrike Mosel (Eds.), Essential of Language Documentation (pp. 113-128). Berlin: Mouton de Gruyter. Himmelmann, N. P. (1998). Documentary and descriptive linguistics. Linguistics (Vol. 36, pp. 161- 195). Himmelmann, N. P. (2006). Language documentation: What is it and what is it good for? In Jost Gippert, Nikolaus P. Himmelmann and Ulrike Mosel (Eds.), Essentials of Language Documentation (pp. 1-30). Berlin: Mouton de Gruyter. Franchetto, B. (2006). Ethnography in language documentation. In Jost Gippert, Nikolaus P. Himmelmann and Ulrike Mosel (Eds.), Essentials of Language Documentation (pp. 183-212). Berlin: Mouton de Gruyter. Johnson, H. (2004). Language documentation and archiving, or how to build a better corpus. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 2, pp. 140-153). London: SOAS. Kambon, O. (2013). Serial verb nominalisation in Akan. PhD Thesis. Department of Linguistics, University of Ghana. Krauss, Michael (2000) Preliminary Suggestions for Classification and Terminology for Degrees of Language Endangerment. Krauss, M. (1992). The world’s languages in crisis. Language, 68 (1), 4-10. Lüpke, F. (2009). Data Collection methods for field-based language documentation. In Peter K. Austin (Ed.), Language documentation and description (Vol. 6, pp. 53-100). The Hans Rausing Endangered Languages Project (HRELP). London: SOAS. Matras, Y. (2005). Language contact, language endangerment, and the role of the 'salvation linguist'. In Peter K. Austin (Ed.), Language Documentation and Description (Vol. 3, pp. 225-251). London: SOAS. McGill, S. (2009). Documenting grammatical tone using Toolbox: An evaluation of Buseman's interlinearisation technique. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 7, pp. 236-250). London: SOAS.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 187 Agyeman, N.A./Legon Journal of the Humanities Vol. 30.2 (2019) Nathan, D. (2010a). Language documentation and archiving: From disk space to My space. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 7, pp. 172-208). London: SOAS. Nathan, D. (2010b). Sound and unsound practices in documentary linguistics:towards an epistemology for audio. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 7, pp. 262-284). London: SOAS. Nathan, D. (2009). Audio responsibilities in endangered languages documentation and archiving. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 6, pp. 101-116). London: SOAS. Nathan, D. (2008). Digital archives: Essential elements in the workflow for endangered languages documentation and revitalisation. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 5, pp. 103-119). London: SOAS. Nathan, D. (2006). Thick interfaces: Mobilizing language documentation. In Jost Gippert, Nikolaus P. Himmelmann and Ulrike Mosel (Eds.). Essential of Language Documentation (pp. 363-379). Berlin: Mouton de Gruyter. Nathan, D. & Austin, P. K. (2004). Reconceiving metadata: Language documentation through thick and thin. In Peter K. Austin (Ed.). Language Documentation and Description, (Vol. 2, pp. 179-188). London: SOAS. Nathan, D. & Csato, E. A. (2007). Multiliteracy, past and present, in the Karaim communities. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 4, p. 207). London: SOAS. Nathan, D. & Fang, M. (2009). Language documentation and pedagogy for endangered languages: a mutual revitalisation. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 6, pp. 32-60). London: SOAS. Obeng, Samuel Gyasi. (2008). Efutu Grammar. Munchen: Lincom Europa. Osam, E. K. (1997). Serial verbs and grammatical relations in Akan. Typological studies in languages (35, pp. 253-280). Raymond, M. (2007). Literacy work in Papua Guinea: The accidental and the planned. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 4, pp. 174-193). London: SOAS. Rogers, C. & Campbell, L. (2015). Endangered languages. Oxford research encyclopaedias. Oxford University Press.

Page 188 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Documenting Simpa: Advances in language documentation Schultze-Berndt, E. (2006). Linguistic annotation. In Jost Gippert, Nikolaus P. Himmelmann and Ulrike Mosel (Eds.). Essential of Language Documentation (pp. 213-253). Berlin: Mouton de Gruyter. SIL International. (2018). Why care about endangered languages? Ethnologue: Languages of the world. Online. Retrieved on 10th October 2018 Simons, G. F. & Fennig, C. D. (2018). Ethnologue: Languages of the world, 21st Ed. SIL International. Online version. Retrieved on 10th October 2018. Thieberger, N. & Musgrave, S. (2007). Documentary linguistics and ethical issues. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 4, pp. 26-37). London: SOAS. Trilsbeek, P. & Wittenburg, P. (2006). Archiving challenges. In Jost Gippert, Nikolaus P. Himmelmann and Ulrike Mosel (Eds.). Essentials of Language Documentation (pp. 311-336). Berlin: Mouton de Gruyter. UNESCO Ad Hoc Expert Group on Endangered Languages. (2003). Document submitted to the International Expert Meeting on UNESCO Programme Safeguarding of Endangered Languages. Online. Retrieved: January 2013. Welmers, W. E. (1974). African language structures. University of California Press. Widlok, T. (2005). Cross-disciplinarity in the documentation work of an anthropologists and linguists. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 3, pp. 12-21). London: SOAS. Woodbury, A. (2003). Defining documentary linguistics. In Peter K. Austin (Ed.). Language Documentation and Description (Vol. 1, pp.35-51). London: SOAS.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 189

Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) DOI: https://dx.doi.org/10.4314/ljh.v30i2.9

Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana

Bernard Afiik Akanpabadai Akanbang Senior Lecturer Department of Planning University for Development Studies, Wa Campus, Ghana E-mail: [email protected] Gordon Dugle Lecturer Department of Management Studies School of Business and Law University for Development Studies, Wa Campus, Ghana [email protected] Millicent Awialie Akaateba Lecturer Department of Planning University for Development Studies, Wa Campus, Ghana [email protected]

Submitted: May 31, 2018/ Accepted: December 2, 2019/ Published: December 30, 2019

Abstract This paper uses Q Methodology to ascertain the views of project staff on how they could be involved in evaluation in order to enhance process use. Structured interviews were conducted with twenty-five project staff who participated in two project evaluations within the context of participatory evaluation in Ghana. It emerged that the use of Q Methodology in evaluation studies enhances the mainstreaming of process use by helping to focus on the critical issues in participatory evaluation practice. Process use is enhanced when ample time is provided for intensive interaction with evaluation stakeholders from the formative stage of evaluation. Process use also requires evaluators to create an environment that is supportive of mutual interaction and closeness in working relations with stakeholders. The research concludes that process use in evaluation is attained through well-planned evaluation that gives voice to project staff and emphasises shared learning. The study recommends critical attention to active participation of project staff, group processes and shared learning, if process use is to be achieved by evaluation. Further research is needed to clarify which form(s) of stakeholder involvement in evaluations contributes more positively to process use.

Keywords: evaluation, stakeholder involvement, process use, Q methodology

Legon Journal of the Humanities Vol. 30.2 (2019) Page 191 Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) Introduction The involvement of multiple stakeholders in evaluation has been widely acknowledged (Akanbang, Darko & Atengdem, 2013b; Smits & Champagene, 2008). For instance, participatory evaluation is noted to facilitate: stakeholder empowerment (Harnar & Preskill, 2007; King, 2007; Suárez-Herrera et al., 2009), increased use of evaluation results (Bryson, Patton, & Bowman, 2011; Cornachione, Trombetta, & Nova, 2010; Patton, 2008; Podems, 2007; Smits & Champagne, 2008; Suárez-Herrera et al., 2009); sense of ownership and commitment to the evaluation and provision of opportunities for learning about effective evaluation practice (Preskill, Zuckerman & Matthews, 2003). What is however unclear in the participatory evaluation discourse is the form of stakeholder involvement in evaluation that is required for it to positively influence process use ( Harnar & Preskill, 2007). From the early 2000s, Ghana began to promote evaluation as part of its poverty reduction efforts. The national approach to evaluation is anchored on participatory monitoring in which ensuring a high level of awareness and participation by the citizenry in evaluation processes is advocated. Thus, participatory data collection, collation and use are integral to the monitoring and evaluation process. Citizenry- service providers’ interfaces have been provided and implemented as part of pilot projects in order to improve upon service delivery (International Monetary Fund, 2009). The prominence given to process use of evaluation in the past two decades has created renewed interest in exploring stakeholder involvement in evaluations (Akanbang, Darko & Atengdem, 2013a; Amo & Cousins, 2007; Baptiste, 2010; Bryson, Patton & Bowman, 2011; Harnar & Preskill, 2007). Process use of evaluation is one of the key ways of enhancing evaluation utilisation (Amo & Cousins, 2007; King, 2007; Patton, 2008, Podems, 2007). Its emergence at a time that evaluation has come under criticism for not contributing enough to the attainment of project objectives (Bambeger, 2009; House, 2008) makes it a contemporary discourse. This paper presents evidence on the form of stakeholder participation that is required to enhance process use in evaluation. It contributes to the literature and the search for ways of enhancing process use by examining the views of project staff involved in participatory evaluation in Ghana. The question explored in the study is “what processes of stakeholder involvement in evaluation are required for it to positively impact on process use?” As underscored by Harnar and Preskill (2007), there is consensus on the need for stakeholder involvement in evaluation, but the form and process of involvement required to attain process use remain unclear. Many evaluations do not commit ample time and financial resources to cater for

Page 192 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana the painstaking consultations and dialoguing needed for stakeholder participation (Bamberger, Rugh, & Mabry, 2006; Bryson & Patton, 2010; Bryson, Patton & Bowman, 2011). The rise in results-based management coupled with the desire to show value for money has also jeopardized substantial involvement of stakeholders in evaluation (Bryson et al., 2011). There is therefore the need for evidence not just to support stakeholder involvement in evaluation but to reveal the processes and forms of involvement that will yield optimal impact on process use. By focusing on the perspective of project staff (demand side of evaluation), the study brings new perspectives to the discussion which has largely reflected the perspectives of evaluators (supply side of evaluation). Theoretical Framework

Stakeholder involvement in evaluation Traditionally, monitoring and evaluation (M&E) have been viewed as important processes for improving development planning, service delivery, demonstrating results, and for learning (UNDP, 2009). Since the 1980s, efforts to involve stakeholders directly affected by development activities to make an input into the evaluation have intensified (Ofosu & Ntiamoah, 2016; Estrella, 2000; Vernooy, 2005). The introduction of participation into M&E is traced to participatory research traditions (Matsiliza, 2012; Vernooy, 2005). The growing interest in performance-based accountability beyond financial reporting; as well as the shift towards decentralisation and devolution of central government responsibilities and authority to lower levels of government have contributed to the interest in participatory monitoring and evaluation (Hope, 2008; Speer, 2012; Estrella, 2000). Participatory M&E is largely based on two perspectives – participation as a means (process of participation) and participation as an end (effect or benefits of participation) (Vernooy, 2005). According to Matsiliza (2012), Participatory M&E is an active engagement and judgment process through which different levels of stakeholders engage in monitoring and evaluation of a project. Among the key features of participatory monitoring and evaluation (PM&E) are: building capacity of local people to analyse, reflect and take action; enabling people to reflect on their experiences together; being flexible and adaptive to local contexts; encouraging stakeholder participation beyond data gathering and strengthening people's capacities to take action and promote change (Estrella, 2000; Mascia, Pailler, Thieme, Rowe, Bottrill, Danielsen & Burgess, 2014; Matsiliza, 2012). In this paper, we adopt the definition of stakeholder involvement in evaluation by Jones, Young and Stanley (2004). They define stakeholder involvement in evaluation as ‘making key stakeholders to become integrally involved in setting up the systems for measuring and reporting on results, reflecting on results achieved, proposing solutions and responding to challenges, and promoting the implementation of evaluation recommendations’ (p 30).

Legon Journal of the Humanities Vol. 30.2 (2019) Page 193 Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) In spite of the benefits associated with stakeholder involvement in evaluation, the level of involvement of stakeholders in evaluation is still a matter of debate and controversy among the evaluation community (Akanbang, Dugle & Yakubu, 2016; Donaldson, Patton, Fetterman. & Scriven, 2010). Process use The goal of PM&E is not simply the creation of an M&E capability, but the use of performance information to assist in improving project planning and management. Evaluation use is thus central to participatory monitoring and evaluation. Evaluation use, defined as the application of evaluation processes, products, findings to produce an effect (Johnsonet al., 2009), has been a long subject of interest to evaluation researchers, policy makers and practitioners (Akanbang, Yakubu & Akaateba, 2015; Hyyrylainen & Viinamaki, 2008). Evaluation use is considered a central outcome of any evaluation, because without it, evaluation cannot contribute to social betterment (Henry & Mark, 2003; Patton, 2008). Research on evaluation use has traditionally been limited to use of findings divided into three types: instrumental, conceptual and symbolic use (Akanbang, Yakubu & Akaateba, 2015; Amo & Cousins, 2007; Fleischer & Christie, 2009; Hyyrylainen & Viinamaki, 2008). Patton, in 1997, is believed to have coined process use as “use category of evaluation”. Since then, a lot of interest has grown over the subject (see Akanbang, Osei-Darko &Atendem, 2013a & b, Forss et al., 2002; Amo & Cousins, 2007; Harnar & Preskill, 2007; Podems, 2007; Henry & Mark, 2003; Johnson et al., 2009; Baptist, 2010). Patton, later in 2008, defined process use as including: cognitive, attitudinal and behaviour changes in individuals, and programme or organisational changes resulting, either directly or indirectly from engagement in the evaluation process and learning to think evaluatively (e.g., increased evaluation capacity, integrating evaluation into the programme, goals clarification, conceptualizing the programme’s logic model, setting evaluation priorities, improving outcomes measurement). (Patton, 2008, p. 156) Forss, Rebien and Carlsson (2002) argued that evaluation commissioners and evaluators should work explicitly to increase process use as the most cost effective way of strengthening the overall utility of an evaluation. Process use makes an evaluation to become part of the treatment, rather than just being an independent assessment of effects. The definition of process use by Akanbang et al. (2013b) is adopted in this paper. They define process use as ‘uses that occur and result in programme enhancement and organisational learning as a result of programme participants’ involvement in and learning from the evaluation process. It is indicated by changes in individuals’ knowledge and behaviour, changes in programme design

Page 194 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana and operations and changes in organisational procedures resulting directly from evaluation process and not the evaluation findings’ (p. 2). Context to the study This paper is based upon the experiences of project staff of two project evaluations, Small Towns Water Supply and Sanitation Project (STWSSP) and the Community Based Health Planning Services CHPS), in a zone that piloted the decentralised M&E system developed by the National Development Planning Commission of Ghana. The STWSSP is a US$ 26 million Project that was funded by the World Bank. The Project initially spanned November, 2004 and April, 2009 but with some amendment and additional funds to the tune of US$10m in 2007 and US$ 15m in 2009, the project was extended by one year. The Project was implemented in seventy-two communities in forty-two districts in six regions of Ghana – Upper East, Upper West, Brong-Ahafo, Ashanti, Central and Western Regions. An evaluation in the first quarter of 2011 was conducted in four districts (Bongo and Garu in the Upper East and Wa and Lawra in the Upper West) out of twelve districts in the six project regions. Monitoring and evaluation was an integral part of project management. Project progress monitoring was undertaken mainly in the form of field visits to sites of projects to ensure compliance and immediate feedback for project improvement. At the technical level, monitoring tracked the extent to which the project delivered the expected quantity and quality of outputs. At the administrative level, monitoring tracked capacity of staff to deliver results, use of resources and equipment and the extent to which workplans and schedules were adhered to. Quarterly reviews, mid-year reviews and annual reviews were an integral part of the project evaluation and provided platforms for actors to assess the progress of the project, identify obstacles to project implementation and reach consensus on the way forward for successful project completion. CHPS project is a US$ 4.8 million project with funding support from the Japanese International Development Cooperation (JICA). Spanning March 2006 and February 2010, it aimed at strengthening community participation in health planning and service provision; and strengthening of monitoring and supervision at all levels of health services delivery in the Upper West Region. Evaluation under the project had a two-fold purpose: learning for project enhancement, and upward accountability to the financier. This study limited itself to process use in evaluation in the two projects in the Upper West and East Regions because the two regions belonged to a context in which NDPC framework for decentralised participatory monitoring and evaluation was piloted.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 195 Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) Methods The study was conducted between April 2011 and October, 2012. A survey of all ongoing and completed projects in the Upper East and West Regions between the period 2000 and 2010 was undertaken. Two projects were then purposively selected because they adopted participatory evaluation process; and had at least ten participants in the evaluation still at post to participate in the study. The study participants were all staff who were involved in the implementation of the projects and participated in their evaluation and were still at post. In Q Methodology, participants are selected not to provide an accurate mirror of the population but to enable them communicate their views (Watts & Stenner, 2012). The study participants for the CHPS project consisted of three staff at the project office, three regional deputy directors, two district health directors and two staff in-charge of the CHPS compounds who were directly involved in the implementation and evaluation of the project. Respondents from the STWSSP included two extension staff, two MIS officers, and one of the accountants of the Community Water and Sanitation Agency in the Upper East and West Regions; four district schedule officers for water and sanitation; four team leaders of the district water and sanitation teams; and two water board chairmen. Respondents had been working on the projects for five years and participated in at least three evaluations. The study employed Q methodology in collecting and analysing the data. Q Methodology is a mixed-methodological approach rooted in constructivism (Boroş, Visu-Petra & Cheie, 2007; Danielson, 2009; Ward, 2010). The methodology provides flexible procedures for the examination of subjectivity within an operant framework. In an operant approach, the subjects of the study in their natural state are allowed to speak for themselves. Q Methodology leans more towards qualitative methods at the point of data collection but gets more quantitative at the point of analysis (Brown, Durning, & Selden, 2008; Watts & Stenner, 2012). Q Methodology is suitable for in-depth analysis of small number of cases. It is these unique strengths of the methodology that recommended itself to this study in relation to other qualitative methods or even R Methodology. We were interested in capturing the subjective experiences of staff involved in the implementation and evaluation of community development projects. The views of project beneficiaries were not considered in the study because we wanted to avoid complications associated with Q Methodology, such as the administration of the Q Sort to illiterate populations. The study also found it complicated to use pictures to represent the different statements that constitute the Q Sample to make the sorting process easier for illiterate populations.

Page 196 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana Besides, the study participants were few and it was inappropriate to use conventional factor analysis which requires a large sample size for statistical power (Watts & Stenner, 2012). The study is focused on the identification of themes in stakeholder evaluation approaches and not on consensus, as is the case with traditional factor analysis. In addition, Q Methodology was preferred over other qualitative methods such as focus group discussions and in-depth interviews in this study because it allowed for quantitative analysis of the data and for easy identification of patterns across persons which is not possible with the other qualitative methods. Subjectivity runs through all the stages of a Q study, yet the factors are usually reliable and easily replicated, and are subject to statistical summary which facilitates more careful description and comparison. Q Methodology is distinct from R Methodology in that the latter studies relationships among objective variables while the former studies relationships between persons. Notwithstanding these strengths of Q Methodology, it is not a very popular methodology in participatory evaluation studies, even though, it has grown over time and is widely used in psychology, political science and in the social and health sciences (Danielson, 2009). Its use in this study thus expands the methodological scope of evaluation studies. The following Q Methodology procedures were used: generation of ideas about the research topic (Concourse), generation of Q Samples, generation of Person Samples, development of Q sort and analysis of results (van Exel & de Graaf, 2005; Webler, Danielson, & Tuler, 2009). These stages and procedures are discussed in detail in BoroŞ, Visu-Petra and Cheie (2007), van Exel and de Graaf (2005), and Webler, Danielson and Tuler (2009). The concourse of this paper was derived from reported statements in published literature about what constitutes the form of stakeholder involvement processes required for process use of evaluation. Specifically, the works of Bryson et al. (2011), Cornachione et al. (2010), Estrella and Gaventa (1998), Rossi et al. (2004), Smits and Champagne (2008) were the main sources of information for this concourse. In all 25 statements were generated from the literature to constitute the concourse for this study. Our Q Sample was generated by subjecting the concourse statements to field test among five staff of the two projects to find out how the statements represented the perception of evaluation stakeholders. The field test helped to refine statements that were initially not clear to evaluation stakeholders. All the 25 refined statements constituted the Q Sample. The question of bias does not arise in the constitution of the Q Sample in the study because the concourse is representative of all viewpoints on the subject. The Q Sample is represented in Table 1.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 197 Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) Table 1: Q Sample

No. Statement 1 The evaluator needs not consult with stakeholders to describe the evaluation, their willingness to participate and their potential involvement in order to promote process use 2 Evaluation managers need to inform stakeholders about the evaluation, describe its broad parameters, identify the evaluator and initiate dialogue in order to enhance process use 3 Brainstorming solutions with participating organisations/individuals promote process use 4 Stimulating sharing and expressing multiple standpoints benefit collective comprehension 5 Following-up to determine whether decision-making is informed by evaluation constitutes an interference with management’s work and is not required to ensure process use 6 Organising the evaluation at a time that allows it to affect project decisions encourages participation and facilitates process use 7 Constantly keeping stakeholders informed of their expectations and developments promotes process use 8 A well formulated strategy for stakeholders’ participation is enough to ensure process use 9 The neglect of cultural sensitivity in an evaluation imperils process use 10 The evaluator’s ability to build and sustain interests of stakeholders in the use of the evaluation determine whether process use occurs or not 11 When stakeholders are provided with a guided journey of discovery that emphasises the learning process, process use is guaranteed 12 Assessing the information needs of stakeholder groups and gauging their potential level of commitment does not bring any difference to the use of evaluations 13 Ensuring that all key stakeholders have access to information on the evaluation at every stage enhances process use occurrence 14 When findings are presented using theatre, skits or video, evaluation findings are trivialised and therefore lose their value and seriousness 15 When friendly, open and frank discussions, while respecting differing opinions and individual sensitivities are encouraged by the evaluator, process use is enhanced 16 Evaluators need not reach decisions with the staff and management about the implications of the analysed information in order to promote process use 17 The evaluator from the design stage of the evaluation engages stakeholders on obstacles and opportunities to enhance process use

Page 198 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana

18 Packaging and disseminating evaluation findings bearing in mind the needs of the different stakeholders enhances process use 19 The use of pre-departure debriefings on initial findings is as weak as the fectef of the main evaluation report in influencing process use 20 Use of focus groups as medium of data collection will not yield any learning as only a few people will capture the discussions 21 Evaluators without people skills (e.g. building relationships, facilitating groups, managing conflicts) cannot facilitate process use 22 Involving stakeholders and participants in an evaluation process, focusing on issues that are relevant for them, increases process use 23 Involving stakeholders who have an interest and want to know something more about the programme enhances process use of evaluations 24 Helping key stakeholders to avoid attaching their egos to the outcome of the evaluations encourages genuine participation and process use 25 Discussion of evaluation findings as a group enhances the process use of the evaluation Source: Authors’ Construct, 2011 The target population is composed of all evaluation participants of the two project evaluations in the Upper West and East regions. All participants of the population were expected to be interviewed. Snowballing was used to reach participants as a sample frame of participants was not readily available. A total of twenty-five (10 from CHPs and 15 from STWSSP) respondents participated in the study. A Q sort is a special questionnaire used in Q studies. The shape of the Sort is a quasi-normal distribution while the kurtosis of the distribution depends on the nature of controversy surrounding the topic. Studies where involvement, interest or knowledge of respondents is expected to be low, the distribution is steeper in order to leave room for ambiguity, indecisiveness or error in the middle of the distribution. However, in studies where respondents have well-articulated opinions on the topic, the distribution is flatter in order to provide more room for strong (dis)agreement with statements. In this study, steeper distribution was selected because process use is still an emerging concept and relatively new to most evaluation participants. Our Q Sort consisted of a pack of 25 randomly numbered cards (number of cards in a pack corresponds to the number of statements in the Q Sample). Each card contains one statement of the Q sample. The cards were what participants sorted. Cards were printed on heavy paper using a legible font type and size and cut into uniform sizes. They were then bound together with an elastic band. The pack also

Legon Journal of the Humanities Vol. 30.2 (2019) Page 199 Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) included a guide, a score sheet containing the distribution for the Q sorting task, and a list of instructions on how respondents were to rank the statements. The score sheet was a single page containing the question that was to be addressed at the bottom and answer spaces above as shown in Table 2. The Q Sort includes a rating scale across the top that can range from –3 to 3 to –6 to 6 (Akhtar-Danesh et al., 2008). Tracking questions such as name, age, educational qualification, and experience with evaluation, among others were put on a different page. The Q Sorts were administered to individual participants in a solo setting. Up to 30 minutes was allocated for the completion of each Q Sort. A sample of the Q-Sort Table is shown in Table 2.

Principal Component Analysis (PCA) within the Q Methodology software (PQ Method) was used to analyse the data. PCA produces eigenvalues for each factor, which helps in deciding on the number of factors to consider for further analysis. Eigenvalues are a measure of the relative contribution of a factor to the explanation of the total variance in the correlation matrix. For a factor to be significant, it must have an eigenvalue of 1 and above, have at least two people significantly loading on it. A factor loading is significant at 0.01 level, if it exceeds 2.58 of the Standard Error (SE) where standard error (SE) is defined as SE = 1/√N, where ‘N’ is the Q Sample (Brown, 1980 cited in Akanbang, 2013a). A factor loading expresses the extent to which each Q sort is associated with each

Page 200 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana factor. Factor interpretation is about generating descriptions of social perspectives contained in each factor. Three outputs of the analysis (distinguishing statements, normalised factor scores, and consensus statements) are relevant in interpreting the perspectives. Each perspective is interpreted using the distinguishing statements (Watts & Stenner, 2012). A distinguishing statement for a perspective is a statement that its score on that perspective is significantly different from its score on any other perspective. The highest-ranking statements are those that loaders on the perspective said were most like how they think, while the lowly-ranked statements indicate the issues that the participants felt were less like how they think (‘3’ were highly ranked, ‘0’ neutral and ‘-3’ lowly ranked). These high and low ranked statements were weaved together into an explanation of how people who loaded highly on this factor saw the subject matter. A distinguishing table or a consensus table has three main components as shown in Table 5 (the numeric number assigned to the statement in the Q Sample, the statement itself and the ranking given to the statement). Comments made by the Q participants during the Q sorting process were used to aid in the factor interpretation. The accuracy of the interpretation was verified by asking the relevant (significantly loaded) participants to comment on their views about the theme of the study. Results Three perspectives on the form and processes of involving stakeholders in evaluation to enhance the occurrence of process use emerged from the study. These perspectives were found to be acceptable because each of them had more than the minimum number of loadings required to meet the composite reliability of 0.95 (Akhtar-Danesh et al., 2008). The correlations between the factors were also low, an indication that the factors were uniquely defined. Table 3 shows the factor characteristics while Table 4 shows the correlation between the factors. Table 3: Factor Characteristics Factors 1 2 3 No. of Defining Variables 7 8 5 Average Rel. Coef. 0.8 0.8 0.8 Composite Reliability 0.966 0.97 0.952 S.E. of Factor Scores 0.186 0.174 0.218

Legon Journal of the Humanities Vol. 30.2 (2019) Page 201 Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) Source: Analysis of Field Data, 2011

Table 4: Correlations between Factor Scores

1 2 3 1 0.3392 0.3366 2 0.4001 3

Source: Analysis of Field Data, 2011 Factor 1, labelled as “early involvement and information flow perspective” had seven respondents identifying with it while factor 2, defined as “who to involve and use of group processes perspective” had eight respondents loading on it. Factor 3, labelled as “evaluator as a facilitator” group had five respondents loading on it. Factor 1: Early involvement and information flow perspective The perspective held by this factor is illustrated in Table 5. Out of seven respondents who loaded on the factor, five were from the STWSSP. As captured in Table 5, this group perceives the stage of the evaluation process at which stakeholders were involved as very important to the occurrence of process use. In addition, there was the need for flow of information in an environment of respect, openness and frankness in the evaluation process. At the onset of the evaluation, project managers were expected to brief staff of the evaluation on its purpose and processes. Table 5: Statements that illustrate early involvement and information flow perspective

No. Statement Rank Evaluation managers need to inform stakeholders about the 2 evaluation, describe its broad parameters, identify the evaluator and initiate dialogue in order to enhance process use 3 Constantly keeping stakeholders informed of their expectations 7 and developments is a way of promoting process use 2 Organising the evaluation at a time that allows it to affect project 6 decisions encourages participation and facilitates process use 2 When friendly, open and frank discussions, while respecting 15 differing opinions and individual sensitivities are encouraged by the evaluator, process use is enhanced 3

Page 202 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana Evaluators without people skills such as building relationships, 21 facilitating groups, managing conflicts and interpersonal communication cannot facilitate process use 1 Packaging and disseminating evaluation findings bearing in mind 18 the needs of the different stakeholders enhances process use -1 Ensuring that all key stakeholders have access to information on 13 the evaluation at every stage enhances process use occurrence -1 Discussion of evaluation findings as a group enhances the process 25 use of the evaluation -2 The use of pre-departure debriefings on initial findings is as weak 19 as the effect of the main evaluation report in influencing process use -3

Source: Analysis of Field Data, 2011 Comments from three participants who identified with this perspective clarify why they think stakeholders should be involved at the early stage of evaluation. The whole evaluation process should be discussed with management for the evaluator to understand issues from the onset. Evaluators need to know the cultural values of the key stakeholders so as to obtain their cooperation and meaningful contributions in the process. If stakeholders do not understand the purpose of the evaluation, they may not cooperate and key information may not be discussed” (Deputy Regional Director of Health Services, Public Health, 03.05.2011) “Stakeholder consultation is very important and needs to be done early to ascertain their willingness and ability to participate in the process. (Extension Officer, CWSA, 11.05.2011) Besides the stage of involvement and manner of information flow, respondents also preferred that evaluation be organised at a time that is suitable to participants as well as meet their information needs. The use of pre-departure debriefings was important to this perspective. Flow of information during the evaluation was also important to this group. It was recognised by the group that this flow of information goes beyond the beginning and end stages of the evaluation. Information flow should be institutionalised throughout the evaluation process. The following comments from respondents who identified with the perspective illustrate their views. Openness and frankness is a prerequisite to a successful evaluation and learning process. There should not be secrecy in the process. Access to information brings about success in whatever endeavour you want to do. (Chairman, Water and Sanitation Development Board, 16.05.2011)

Legon Journal of the Humanities Vol. 30.2 (2019) Page 203 Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) In an open and frank environment, learning takes place and all contributors and listeners go home with something new. Also, some salient points that can advance project implementation are brought up. (Information Technology Specialist, Community Water and Sanitation Agency, 10.05.2011) Cordial relationship with stakeholders and adequate information about the evaluation clears suspicion such as perception of the evaluation being a witch hunting activity. District Budget Officer/District Water and Sanitation Team Schedule Officer, 02.06.2011) The need for openness as expressed in the above quotes is an indication of the need to provide information to disabuse the minds of participants of their wrong notions about evaluation, and the need to build trust and confidence of the staff in the evaluation process. This way staff would be forthcoming with vital information critical to the success of the evaluation mission. Factor 2: Who to involve and use of group processes perspective The concern of this group was that, the type of evaluation participants and the processes of involving them are important to having an evaluation that can contribute to enhancing process use. Statements contained in Table 6 illustrate the perspective of the group. Stakeholders with strong interest in the project should be involved. To ensure meaningful involvement of stakeholders, it is important to assess the information needs of participants at the beginning of the evaluation. Armed with the information needs of the different stakeholders, the evaluation is better designed, incorporating the identified needs and conducted in a manner that takes into consideration the different contexts and sensibilities of participants. The use of group processes was recognised as essential to evaluation’s ability to contribute to process use by the group. Two respondents of this group gave the following comments to buttress the need for the employment of group processes in evaluations. Use of group processes enhances cross fertilisation of ideas and it is good for the evaluation. (Water and Sanitation Engineer, CWSA, 11.05.2011) Discussions of findings as a group is very important as it will afford all stakeholders access to the findings and therefore enhances the use of the evaluation to impact on development outcomes. (Senior Environmental Health Officer, 16.05.2011)

Page 204 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana Table 6: Statements Illustrating Factor 2 Perspective

No. Statement Rank Involving stakeholders who have an interest and want to know 23 something more about the project enhances process use of evaluations 3 25 Discussion of evaluation findings as a group enhances the process use 3 When friendly, open and frank discussions, while respecting differing opinions and individual sensitivities encouraged by the evaluator, 15 process use is enhanced 3 Ensuring that all key stakeholders have access to information on the 13 evaluation at every stage enhances process use occurrence 2 Involving stakeholders and participants in an evaluation process, 22 focusing on issues that are relevant for them, increases process use 1 9 The neglect of cultural sensitivity in an evaluation imperils process use 1 The use of pre-departure debriefings on initial findings is as weak as the 19 effect of the main evaluation report in influencing process use 1 Assessing the information needs of stakeholder groups and gauging their potential level of commitment does not bring any difference to the use 12 of evaluations -2 Following-up to determine whether decision-making is informed by evaluation constitutes an interference with management’s work and is 5 not required to ensure process use -2 Evaluators need not reach decisions with the staff and management about the implications of the analysed information in order to promote 16 process use -3

Source: Analysis of Field Data, 2011

Factor 3: Evaluator as a facilitator perspective In this group, the evaluator’s role is appreciated not in terms of the perceived conventional role as an independent technical person, but as a facilitator. Evaluators should provide participants with a guided journey of discovery in which learning and not judgement is the cardinal objective. Evaluators’ exit strategy is also very vital while their knowledge and skills in the use of modern techniques in the presentation of findings is also much recognised. Their show of respect for culture and context and the ability to ensure flow of information is critical to carrying the people along in the evaluation process. Similarly, their skills in stimulating sharing and expressing multiple viewpoints are required for facilitating collective understanding and action. The following comments by some of the members of the

Legon Journal of the Humanities Vol. 30.2 (2019) Page 205 Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) group shed more light on the skills required by evaluators in order to facilitate the conduct of evaluations in ways that impact positively on process use. People skills such as building relationships with stakeholders are vital to facilitating evaluations that are used to improve upon projects. The dynamic nature of the evaluator as well his/her non-judgmental attitude is essential to the conduct of an evaluation that will affect development outcomes positively. (Deputy Director of Regional Health Service, Administration, 03.05.2011) Evaluators should know and be sensitive to the cultural values of the key stakeholders if their cooperation and meaningful contribution to the evaluation is to be assured. (Deputy Director of Regional Health Service, Public Health, 03.05.2011) Table 7 sheds more light on the perspectives of the group.

Table 7: Statements Illustrating Factor 3 Perspective

No Statement Rank Helping key stakeholders to avoid attaching their egos to the outcome of 24 the evaluations encourages genuine participation and process use 3 When stakeholders are provided with a guided journey of discovery that 11 emphasises the learning process, process use is guaranteed 1 Ensuring that all key stakeholders have access to information on the 13 evaluation at every stage enhances process use occurrence 3 Discussion of evaluation findings as a group enhances the process use of 25 the evaluation 3 9 The neglect of cultural sensitivity in an evaluation imperils process use 2 Stimulating sharing and expressing multiple standpoints tend to benefit 4 collective comprehension of the program 2 Involving stakeholders who have an interest and want to know something 23 more about the project enhances process use of evaluations -1 The use of pre-departure debriefings on initial findings is as weak as the 19 effect of the main evaluation report in influencing process use -1 Evaluators without people skills (e.g. building relationships, facilitating 21 groups, managing conflicts) cannot facilitate process use -2 When findings are presented using theatre, skits or video, evaluation 14 findings are trivialised and therefore lose their value and seriousness -3 Source: Analysis of Field Data, 2011

Page 206 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana Perspectives common to the groups The need for early consultations with evaluation stakeholders in order to impact process use was shared by all three groups of perspectives. Another common area is the recognition that the stakeholders to be involved in an evaluation must be people who are interested in evaluation. Discussion of findings Who to involve in an evaluation process: This study clearly reveals that it is not enough to know which group(s) of stakeholders to involve, but more crucially, who among a group of stakeholders to involve in an evaluation. It is important to involve staff who have particular need for the evaluation information. The involvement of such stakeholders is very important especially among public organisations where evaluation is seen as an add-on activity as observed by Akanbang, Yakubu and Dugle (2016). Evaluation guidelines of organisations should therefore state the specific individuals within a category of stakeholders to be involved inan evaluation. Evaluation research on identifying these specific individuals is required to aid evaluation policy of organisations. Early involvement of staff in the evaluation process: early involvement in the process is important in ensuring buy-in and ownership of the evaluation process. Initial briefings to staff about the evaluation will ignite a deep thinking process in the staff about the project being evaluated. During the briefing process, new thoughts and insights may emerge to facilitate the evaluation. Early consultations require explaining the purpose of the evaluation, the role of evaluation in the project and how the evaluation ought to be conducted. The question, however, is how to make ample time for early involvement in evaluation in contexts where it is largely donor driven, time bound, and accountability oriented. Critical as this question is, the study believes that emerging arrangements where project implementers have control of the budget for evaluation as was revealed by Akanbang et al.(2016), offers an opportunity for them to assert themselves and demand an evaluation process that puts them at the centre at the beginning of the evaluation. Donor recipient organisations must mainstream evaluative thinking in all aspects of their work so that they can actively participate and own project evaluation. The employment of group processes emerged as central to the occurrence of process use. As observed by Preskill et al. (2003), the more occasions project staff have for sharing information, challenging assumptions, asking questions, and publicly reflecting on past experiences, the more likely they are to experience process use. The employment of group processes also allows the evaluator to gain a comprehensive understanding of the project and forge consensus with

Legon Journal of the Humanities Vol. 30.2 (2019) Page 207 Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) stakeholders on key decisions required for the enhancement of the project. To overcome the low uptake of group processes in evaluation as noted by Akanbang et al. (2016), evaluation policies and guidelines should require projects to form an all-encompassing taskforce for evaluation; as well as group discussion of findings. Ensuring a functioning communication throughout the evaluation process: this finding of the study highlights the importance of trust in an evaluation that supports process use. A functioning communication would generate continuous interest in participants, thus causing them to reflect on emerging issues from the evaluation. Besides, it would also help to deal with issues of misrepresentations, antagonism and consequently promote acceptance and ownership of the evaluation process and its findings and recommendations. In addition to the usual debriefing meetings with evaluation participants, exchange of email and telephone contacts as well as the establishment of a social media platform purposefully for the evaluation should be encouraged in evaluation. The role and skill of the evaluator in ensuring process use is highlighted by this paper. Evaluators should not continue to distance themselves from the evaluation process as neutral technical persons. They should be endowed with skills in stimulating, sharing and expressing multiple viewpoints, that are essential to ensuring collective comprehension and action. This implies new calibre of evaluators and therefore professionalization of evaluation in the developing world. Evaluation terms of reference should also clearly state people skills as a requirement of the evaluator. Conclusion In contributing to the evolving literature on the form of stakeholder involvement in evaluation relevant for process use, the study argued, based on a Q Methodology, that involving project staff at the early stages of evaluation, ensuring a functioning communication with them as the evaluation progresses and employing group processes in the evaluation enhance process use. Even though the findings are not significantly new to the evaluation community, the use of Q Methodology has enabled the narrowing of the wide range and diverse perspectives on the stakeholder involvement processes to a few focus areas enabling the formulation and targeting of evaluation processes that facilitate process use. The study concludes that for enhanced process use, a well-planned evaluation process that gives voice to project staff and places emphasis on shared learning is essential. The study therefore recommends critical attention to active participation of project staff, group processes and shared learning, if process use is to be achieved by evaluation. Further research is needed to clarify which form(s) of stakeholder involvement in evaluations contributes more positively to process use. Page 208 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana References Akanbang, B. A. A., Darko, R. O., & Atengdem, P. B. (2013a). Programme Implementers Experiences of Factors Affecting the Occurrence of Process Use in Evaluation in Northern Ghana. Journal of Human Subjectivity, 11(1), 5–27. Akanbang, B. A. A., Darko, R. O., & Atengdem, P. B. (2013b). Programme implementers’ experiences of process use types in three evaluation contexts in Northern Ghana. Operant Subjectivity, 36(4), 270–292. Akanbang, B. A. A., Yakubu, I. & Akaateba, M. (2015) Programme Implementers ‘experiences of evaluation use in Northern Ghana. Annals of Humanities and Development Studies, 6(1), 38-63 Akanbang, B. A. A., Yakubu, I., & Dugle, G. (2016). Evaluation Inquiry in Donor Funded Programmes in Northern Ghana: Experiences of Programme Staff. Ghana Journal of Development Studies, 13(1), 18–40. Akhtar-Danesh, N., Baumann, A., & Cordingley, L. (2008). Q-Methodology in nursing research a promising method for the study of subjectivity. Western Journal of Nursing Research, 30(6), 759–773. Amo, C. and Cousins, B. J. (2007). Going through the process: an examination of the operationalisation of process use in empirical research on evaluation. In B. J. Cousins, (Ed.), Process use in theory, research and practice: new directions for evaluation (pp. 5-26). San Francisco, California. Bamberger, M. (2009). Enhancing the utilization of evaluations for evidence- based policy-making. In M. Segone (Ed.). Bridging the gap: The role of monitoring and evaluation in evidence-based policy making (pp 99-120). Geneva: Evaluation working papers, No 12 Bamberger, M., Rugh, J. & Mabry L. (2006). Real world evaluation: Working under budget, time, data and political constraints, Thousand Oaks, CA: Sage. Baptiste, L. J. C. (2010). ‘Process use across evaluation approaches: An application of Q methodology in programme evaluation’: (Doctoral dissertation) Retrieved from rave.ohiolink.edu/etdc/view?acc_num=kent1271093193. Accessed 10.06.2018 Barbosa, J. C., Willoughby, P., Rosenberg, C. A., & Mrtek, R. G. (1998). Statistical methodology: VII. Q‐methodology, a structural analytic approach to medical subjectivity. Academic Emergency Medicine, 5(10), 1032–1040.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 209 Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) BoroŞ, S., Visu-Petra, L., & Cheie, L. (2007). A Q-Sort analysis: Investigating the social perception of a chronic disease: Between sympathy and stigma. Cognitie, Creier, Comportament/Cognition, Brain, Behavior, 11(2). Brown, S. R., Durning, D. W., Selden, S. C. (2008). Q methodology. In G. J. Miller & K. Yang (Eds.), Handbook of research methods in public administration (pp 721-763). CRC Press. Bryson, J. M. & Patton, M. Q. (2010). Analysing and engaging stakeholders. In Wholey, J. S., Hatry, H. P. & Newcomer, K E. (Eds.). Handbook of practical programme evaluation (3rd ed., pp 30-54). San Francisco: John Wiley and Sons. Bryson, J. M., Patton, M. Q., & Bowman, R. A. (2011). Working with evaluation stakeholders: A rationale, step-wise approach and toolkit. Evaluation and Program Planning, 34(1), 1–12. Cornachione, E. B., Trombetta, M. R., & Nova, S. P. C. (2010). Evaluation use and involvement of internal stakeholders: The case of a new non-degree online program in Brazil. Studies in Educational Evaluation, 36(1), 69–81. Danielson, S. (2009). Q Method and Surveys: Three Ways to Combine Q and R. Field Methods, 21(3), 219–237. Donaldson, S.I., Patton, M.Q., Fetterman, D. & Scriven, M. (2010). The 2009 Claremont debates: The promise and pitfalls of utilisation focused and empowerment evaluation. Journal of Multi-Disciplinary Evaluation, 6 (13), 15-57. Estrella, M. (2000) An introduction. In M. Estrella, et al. (eds.), Learning from change: Issues and experiences in participatory monitoring and evaluation (pp 1-15). Intermediate Technology Publications Estrella, M., & Gaventa, J. (1998). Who counts reality? Participatory monitoring and evaluation: A literature review Brighton: Institute of Development Studies, University of Sussex. IDS Working Paper 70. Fetterman, D. M. (1998). Empowerment evaluation and the Internet: A synergistic relationship. Current Issues in Education, 1(4), 1–20. Fleischer, D. N & Christie, C. A (2009). Evaluation use: Results from a survey of U.S. American Evaluation Association members. American Journal of Evaluation, 30(2), 158-175 Forss, K., Rebien, C. C., & Carlsson, J. (2002). Process use of evaluations: Types of use that precede lessons learned and feedback. Evaluation, 8(1), 29–45.

Page 210 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana Friedman, M. (2005). Trying hard is not good enough: How to practice measurable improvements for customers and communities. Victoria, B.C: Trafford Publishing. Greene, J. G. (2005). Stakeholder Involvement’. In S. Mathison (Ed.), Encyclopaedia of evaluation. Thousand Oaks: Sage. Harnar, M. A., & Preskill, H. (2007). Evaluators’ descriptions of process use: An exploratory study. New Directions for Evaluation, 2007(116), 27–44. Hatry, H. P., Wholey, J. S., & Newcomer, K. E. (2010). Evaluation challenges, issues and trends. In K. E. Newcomer, H. P. Hatry, & J. S. Wholey (Eds.), Handbook of practical program evaluation (3rd ed., pp 668-680). San Francisco: John Wiley & Sons. Henry, G. T. and Mark, M. M. (2003). Beyond use: understanding evaluation’s influence on attitudes and actions. American Journal of Evaluation, 24(3), 293–314. House, E. R. (2008). Blowback consequences of evaluation for evaluation. American Journal of Evaluation, 29(4), 416-426. Hyyrylainen, E. & Viinamaki, O. P. (2008). The implications of the rationality of decision-makers on the utilization of evaluation findings. International Journal of Public Administration, 31(10-11), 1223-1240. International Monetary Fund. (2009). Ghana: Poverty Reduction Strategy Paper—2006 Annual Progress Report (No. IMF Country Report No. 09/237). Retrieved from https://www.imf.org/en/Publications/CR/Issues/2016/12/31/ Ghana-Poverty-Reduction-Strategy-Paper-2006-Annual-Progress- Report-23154 Johnson, K., Greenseid, L. O.., Toal, S. A., King, J. A., Lawrenz, F. & Volkov, B. (2009). Research on evaluation use; A review of the empirical literature from 1986 to 2005. American Journal of Evaluation, 30(3), 377-410. Jones, R., Young, V., & Stanley, C. (2004). CIDA evaluation guide. Ottawa: Evaluation Division Performance & Knowledge Management Branch. Canadian International Development Agency. Kállay, E. (2007). Q Methodology the investigation of meaning systems in different life-conditions through proverbs. Cognitie, Creier, Comportament/Cognition, Brain, Behaviour, 11(2), 353–369. King, J. A. (2007). Developing evaluation capacity through process use. New Directions for Evaluation, 2007(116), 45–59.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 211 Akanbang, B. A. A., Dugle, G. & Akaateba, M. A./Legon Journal of the Humanities Vol. 30.2 (2019) Mascia, M. B., Pailler, S., Thieme, M. L., Rowe, A., Bottrill, M. C., Danielsen, F.,& Burgess, N. D. (2014). Commonalities and complementarities among approaches to conservation monitoring and evaluation. Biological Conservation, 169(2014), 258-267. Matsiliza, N. (2012) Participatory monitoring and evaluation: Reviewing an inclusive approach in the South Africa’s Government wide monitoring and evaluation. Africa’s Public Service Delivery and Performance Review,1(2), 67-83. Miller, W., & Lennie, J. (2005). Empowerment evaluation: A practical method for evaluating a national school breakfast program. Evaluation Journal of Australasia, 5(2), 18-26. Newcomer, K. E., Hatry, H. P., & Wholey, J. S. (2010). Planning and designing useful evaluations. In J. S. Wholey, H. P. Hatry, & K. E. Newcomer (Eds.), Handbook of practical program evaluation (3rd ed., pp 5-29). San Francisco: John Wiley & Sons. Nicholls, J., Mackenzie, S., & Somers, A. (2007). Measuring Real Value: A DIY guide to social return on investment. London: New Economics Foundation. Ofosu, S., & Ntiamoah, E. B. (2016). Assessing community involvement in monitoring and evaluation of development projects. The case of the Kwahu West Municipal Assembly, Ghana. British Journal of Education, Society and Behavioural Science, 14(1), 1-12. Patton, M. Q. (2008). Utilization-focused evaluation. Sage publications. Podems, D. (2007). Process use: A case narrative from Southern Africa. New Directions for Evaluation, 2007(116), 87–97. Preskill, H., Zuckerman, B., & Matthews, B. (2003). An exploratory study of process use: Findings and implications for future research. American Journal of Evaluation, 24(4), 423–442. Rossi, P. H., Lipsey, M. W., & Freeman, H. E. (2004). Evaluation: A systematic approach. Thousand Oaks: SAGE Publications, Inc. Smits, P. A., & Champagne, F. (2008). An assessment of the theoretical underpinnings of practical participatory evaluation. American Journal of Evaluation, 29(4), 427–442. Speer, J. (2012). Participatory governance reform: a good strategy for increasing government responsiveness and improving public services? World Development, 40(12), 2379-2398.

Page 212 Legon Journal of the Humanities Vol. 30.2 (2019) Agyeman, N.A./Q methodological study on stakeholder involvement processes that support process use of evaluation in the Upper Regions, Ghana Suárez-Herrera, J. C., Springett, J., & Kagan, C. (2009). Critical connections between participatory evaluation, organisational learning and intentional change in pluralistic organisations. Evaluation, 15(3), 321–342. United Nations Development Programme (2009) Handbook on planning, monitoring and evaluation for development results. United Nations Development Programme, One United Nations Plaza New York, NY 10017, USA van Exel, J., & de Graaf, G. (2005). Q Methodology: A sneak preview. Retrieved from http://blogs.nd.edu/lapseylab/files/2014/10/vanExel.pdf onth 9 June, 2010 Vernooy, R. (2005) Participatory monitoring and evaluation: Readings and resources nd (Volume 8, 2 edition). The Rural Poverty and Environment Program Initiative, IDRC, Ottawa, Ontario, Canada. Ward, W. (2010). Q and you: The application of Q Methodology in recreation research. In Watts, C. E., Jr.; Fisher, C. L., (Eds). Proceedings of the 2009 Northeastern Recreation Research Symposium (pp. 75-80). Gen. Tech. Rep. NRS-P-66. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northern Research Station. Watts, S., & Stenner, P. (2012). Doing Q Methodological research: Theory, method and interpretation. Sage. Webler, T., Danielson, S., & Tuler, S. (2009). Using Q method to reveal social perspectives in environmental research. Greenfield MA: Social and Environmental Research Institute, 54, 1–45.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 213

Kambon, Ọ. B. & Asare, Y. M./Legon Journal of the Humanities Vol. 30.2 Kambon, B. O. & Asare, Y. M./(2019)Legon DOI: Journal https://dx.doi.org/10.4314/ljh.v30i2.10 of the Humanities Vol. 30.2 (2019) DOI: https://dx.doi.org/10.4314/ljh.v30i2.10

Humanities and Sciences as Complementary Aspects of an Afrikan=Black Whole: Evidence from Archeoastronomy Ọbádélé Bakari Kambon Senior Research Fellow Language, Literature and Drama Section Institute of African Studies, College of Humanities University of Ghana - Legon Email: [email protected] Yaw Mankatah Asare Graduate Assistant Institute of African Studies, College of Humanities University of Ghana, Legon, Ghana Email: [email protected]

Submitted: January 14, 2019/ Accepted: September 24, 2019 /Published: December 30, 2019

Abstract In ancient Afrika1, science, technology, engineering and mathematics were not seen as separate from or at odds with what is now referred to in English as the Humanities. Focusing on archeoastronomy of Kmt ‘land of Black people (i.e. Ancient Egypt)’, we demonstrate that the scientific principles used to build pyramids, temples, and other edifices were rooted in and fused with Afrikan systems of deep thought and spirituality. The method adopted in this study examines alignments of structures of Kmt ‘land of Black people’ with respect to solar and celestial phenomena in an attempt to establish systematic patterns and correlations between architectural alignments and astronomical phenomena. This method is adopted to show the relationship between structures constructed and knowledge of the oneness of humanities and sciences. Therefore, in this study, we find that the dichotomization and fragmentation of knowledge and disciplines should move towards a more holistic Kmtyw ‘Black people, people of Kmt’ perspective.

Keywords: archeoastronomy, complementarity, archeogeography, STEAM, Kmt

1 As I have argued elsewhere, while not ideal, Afrika (n.)/Afrikan (adj.) is preferred to “Africa” as the word is consistently spelled in various Afrikan languages with a /k/ (cf. Yorùbá Áfríkà; Akan Afrika; Kiswahili Afrika; isiZulu iAfrika; Kikongo Afelika; Hausa Afirka; Kirundi Bufirika; Gĩkũyũ Abĩrika; Igbo Afrịka; Luganda Afirika; Lingála Afríka; Malagasy Afrika; Sesotho sa Leboa Afrika; Oromoo Afrikaa; Fulfulde Afirik; Setswana Aferika; Tsivenda Afurika; Tsisonga Afrika; Siswati Í-Afríka; Soomaaliga Afrika; Kinyarwanda Afurika, etc.) noting that Afrikan languages by-and-large do not use a /c/ for a hard /k/ sound. Because of the aforementioned consideration, this spelling has been advanced and preferred since the 1960s by various Afrikan-Centered individuals, movements and organizations of the continent and diaspora such as Llaila O. Afrika, the Republic of New Afrika, Afrika Youth Movement, the Afrikan-Centered Education movement, the Afrikan-Centered Psychology movement, the New Afrikan Prison Struggle, AfrikanWorldAnalysis.com, Inspire Afrika, Step Afrika!, etc.

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 215 215

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019)) Introduction and Background In this paper, we will articulate the view that, from an Afrikan-centered perspective, the Humanities and Sciences were seen as complementary aspects of a whole. By way of definition, the term complementary denotes “Combining in such a way as to enhance or emphasize the qualities of each other or another” (Stevenson, 2010, p. 356). From another definition, it refers to “completing something else or making it better: serving as a complement – used of two things when each adds something to the other or helps to make the other better: going together well: working well together” (Merriam-Webster, 2018a). It is in this vein that we will examine the Afrikan precedents of what are now disparately categorized as Humanities and Sciences. Qualitative in nature, traditionally the Humanities as a collective comprises the following: “Liberal arts, arts, literature; classics, classical studies, classical languages, classical literature; Latin – literae humaniores” (Waite, 2009, p. 429). According to Merriam Webster, The Humanities relates to “the branches of learning (such as philosophy, arts, or languages) that investigate human constructs […] and concerns as opposed to natural processes (as in physics or chemistry) and social relations (as in anthropology or economics)” (Merriam-Webster, 2018b). However, this type of definition begs the question of where is the intersection of these branches of learning? Not to deprecate the analogy, branches stem from a common tree with common roots. It also may be noted that for many, science may not be so much of a field of knowledge, but rather a ‘systematized’ way of knowing (i.e., that which conforms to the scientific method). Used traditionally, however, the sciences tend to take a quantitative approach towards knowing. This includes Algebra, Agriculture, Chemistry, Biology, Physics, and Mathematics. Looking at the term from an etymological perspective, science denotes "what is known, knowledge (of something) acquired by study; information;" with a strong correlation with the scientific method. It is derived from the Latin scientia “knowledge, a knowing; expertness,” from sciens (genitive scientis) "intelligent, skilled," present participle of scire "to know," probably originally "to separate one thing from another, to distinguish," related to scindere "to cut, divide," from PIE root *skei- "to cut, split" (source also of Greek skhizein "to split, rend, cleave," Gothic skaidan, Old English sceadan "to divide, separate") (Etymonline, 2018). In modern times, the sciences and the humanities are seen and treated as divided and separated from each other conceptually as well as in implementation for the purposes of intellectual categorization. For example, at the University

Page 216 216 Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole:Whole: EvidenceEvidence fromfrom Archeoastronomy Archeoastronomy of Ghana, the College of Humanities is separate from the College of Basic and Applied Science. However, in classical Afrika, Kmt “land of Black people”, contemporarily known as so-called Ancient Egypt and its predecessor civilizations deeper into the interior of Afrika in the south this was not the de facto state of 2 affairs. To put things in perspective in terms of trajectory of Afrikan thought and practice in this regard, it is necessary to look at the origins of these studies and how they persist throughout Afrikan space and time. In this vein, according to Diop, “The history of Africa will remain suspended in air and cannot be written correctly until African historians connect it with the history of Egypt [...] The African historian who evades the problem of Egypt is neither modest or objective, nor unruffled, he is ignorant, cowardly, and neurotic” (Diop & Cook, 1989, p. xiv). Therefore, it is of the utmost importance for us to understand the relationships between Kmt 3 “land of Black people” and its parent civilizations to the south. In Ancient Afrika and more recent pre-conquest-by-Eurasians Afrika, the Humanities and the Sciences were demonstrably seen as complementary aspects of an Afrikan whole. Further, those pursuits, which are now referred to as Sciences and Humanities in English, coexisted and were in no way separated has been discussed by a few other scholars (Browder, 1992; Finch, 1998; Obenga, 2004; Van Sertima, 1983). Separation and dichotomization, however, form the conditio sine qua non, of Eurasian thought. Ani (1994) defines dichotomization in Eurasian thought as: A mechanism which accompanies objectification. It is the splitting of [a] phenomenon into confrontational, conflicting parts. It facilitates the pursuit of power over other, and is therefore suited to the European Asili (Ani, 1994, p. xxviii). Writing further with regard to the Eurasian thrust for dichotomization, in this process: All realities are split into two parts. This begins with the separation of self from “other,” and is followed by the separation of the self into various dichotomies (reason/emotion, mind/body, intellect/nature). The process continues until the universe is composed of disparate entities (Ani, 1994, p. 105). The above leads us to the underlying premise of our article where we hone the idea that it is un-Afrikan to compartmentalise Afrikan knowledge, thought and practise as they mutually complement each other. In fact, Kreamer (2012) attests to this fact when writing on the cosmology of Afrikans as echoed by the arts. She notes in the introduction of the book African Cosmos that Afrikans 2 The use of “so-called” here implies that this is not how the indigenes referred to the country, it is rather a name that has been imposed upon the land by non-Afrikans. 3 See Van Sertima, I. (1994). Egypt: Child of Africa. New Brunswick, NJ: Transaction Publishers. (Van Sertima, 1994).

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 217 217

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019)) take a “holistic approach toward comprehending reality” (p. 15). This is because bodies of knowledge are not compartmentalised but rather integrated thus serving as complements. Ideas of complementary opposites are also attested by Obenga (Obenga, 2004) speaking about the key aspect highlighted in the Ahmes (Rhind) papyrus, which is: the necessity for a correct method (tp-hsb), to be used in directing human intelligence in its effort to understand reality. The method envisaged is obviously the science of mathematics. For the ancient Egyptian conception of mathematics required that the abstract sciences, for which correct method is a prerequisite, be complemented by the applied and technical sciences (Obenga, 2004, p. 434) (bold emphasis added). Thus, a major difference between Afrikan thought and Eurasian thought, therefore is the distinction between complementary opposites and dichotomous diametric opposites. In this paper we lay bare many manifestations of Afrikan conceptions of complementary opposites drawing from examples from throughout the global Afrikan world. We maintain that the Afrikan Science of Astronomy and associated mathematical calculations utilized analogical reasoning to inform Architecture, Philosophy, Spirituality and Literature and vice-versa as the abstract complemented the physical. Analogical reasoning allows for one to grasp knowledge and extrapolate knowledge from one field to another to solve problems. Thus, at the core of religious symbols, rituals, and arts laid a complex understanding of astronomy (Hilliard, 1998; Womack, 2013). With regard to astronomy, the Celestial Sphere – characterised and replicated on Earth – was and is still conceptualised as Mȝˁt ‘Maat’ in the Afrikan worldview. This profound concept whose core principles form the foundation of Afrikan society connotes the concepts of Truth, Justice, Righteousness/Propriety, Harmony, Balance, Order, Reciprocity. However, Beatty (1997) warns us about seeing Mȝˁt ‘Maat’ in fragmented perspectives as opposed to a holistic concept signalling ‘universal order’. For Beatty, “Because Maat is not an object, it cannot be known as an object. Maat is not a Newtonian machine with isolated or separate parts interacting by law. Maat does not have distinct parts or entities, but possesses interrelated and interconnected manifestations of a cosmic whole. It is a we awareness that does not divide up the world into separate and self-contained units. Distinctions are made, yet there is never any fragmentation (p. 216). Mȝˁt ‘Maat’ is thus a fractal principle of a microcosm reflecting the macrocosm of universal order that sustains/governs the cosmos and, as such, it manifests in various ways in the physical world. Beyond this, in the language of mdw nṯr ‘Divine Words/Hieroglyphs’, all nouns are masculine or feminine.

Page 218218 Legon Journal of the HumanitiesHumanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole:Whole: EvidenceEvidence fromfrom Archeoastronomy Archeoastronomy

Mȝˁt ‘Maat’ is represented as a woman with an ostrich feather. As such, to even refer to Mȝˁt ‘Maat’ as “it” may miss the point of Mȝˁt ‘Maat’ as “she” in the worldview of the Kmtyw ‘Black people, people of Kmt’. We get a sense of three levels of the manifestations of Maat by reading Obenga (2004): ● The first manifestation is on a Universal level where all things in the cosmos are aligned. ● The second manifestation is political where the ruler organises his people to cast out isft for the restoration of balance, Ma’at. On the political level, the concept of Ma’at works against injustice. It is in the name of Ma’at that the pharaoh subjugates rebels and dominates foreign lands. ● On the individual level, “Maat embraces specific rules for living in concert with moral principles.” Whoever lives according to these rules and principles achieves universal order in his or her own life, in practical terms, and lives in harmony with the ordered Whole. The most accomplished, useful and appropriate human actions are circumscribed in the cosmological order, as symbolized by the way the pharaoh’s name is written inside a circular cartouche, a perfect geometrical shape representing the vitalizing Sun (Obenga, 2004, pp. 191-192). While all three levels of this multi-layered view are clearly interrelated, in this paper, we focus on the first two aspects of Mȝˁt ‘Maat’ in the form of evidence from archaeoastronomy and other related areas that demonstrate the science related to humanistic endeavours such as “temple-building.” Desires to fulfil the ‘harmonious interaction’ and ‘co-existence’ of the physical and cosmological plane are manifested in the architectural genius of Kmt ‘Land of Black people.’ Architecture is a concrete manifestation of the vision of a people. As people in Kmt sought to align themselves with the sun and stars, architecture was constructed to reflect this vision. However, for this vision to be materialised it demanded that one had an intimate knowledge of the movement of the stars and the sun in order to actualize the concept of “as above, so below” (Chandler, 2000, p. 28; Martin, 2008, p. 957). It is worth noting that far from alignments of monumental and structures being an invention in the nation of Kmt ‘Land of Black people,’ deep in the southern interior of Afrika there was a megalithic calendar circle aligned with

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 219 219

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019)) certain stars at a place now referred to as Nabta Playa. According to Brophy and Rosen (2005): The report that a primary megalithic alignment was constructed to point to the bright star Sirius, circa 4,820 BC, is reconsidered in light of the satellite data, new field data, radiocarbon, lithostratigraphic and geochronologic data, and the playa sedimentation history. Other possible archaeoastronomical interpretations are considered for that alignment, including the three star asterism (of Alnitak, Alnilam, and Mintaka) circa 6,270 BC that is also implicated in the small Nabta Playa ‘calendar circle’ (p. 15). Interestingly, the calendar circle of Nabta Playa was located in the area known as Nubia and which now forms part of southern Egypt and was created a few thousand years before the most conservative dates given for the unification of Kmt ‘land of Black people’ under ḥr nˁr mr ‘Heru Narmer,’ the acknowledged unifier of upper (southern) and lower (northern) Kmt ‘land of Black people’. The stone megaliths found in the Nubian desert, Nabta Playa, also echo a strong sense of astronomical knowledge, requiring scientific study, observation and acute scientific calculation. Malville, et al. (2007) discuss the astronomy of Nabta Playa pointing to three pieces of evidence of astronomical knowledge: 1. the repeated positioning of “megaliths, stele, human burials and cattle burials toward the northern regions of the sky” 2. the cromlech on the site as shown by Wendorf and Schild also show an orientation of “two sightlines toward the north and toward the rising sun at the June solstice” 3. the orientation of the stele towards the brightest stars, Nabta, Sirius, α centauri, Arctus and Canopus in the celestial realm of the Nabta skies (McK Malville et al., 2007, p. 6) Thus, while we will look extensively at temple orientations of Kmt ‘land of Black people’, it is imperative for us to understand that Kmt ‘land of Black people’ should not be seen as isolated from other Black people of its parent civilizations in the interior of Afrika to the south or restricted to ancient pre-history. In fact, a more recent archaeo-astronomical site organized along similar lines is known as Namoratunga II. Namoratunga II is located to the west of Kenya’s Lake Turkana in Kenya and is thought to have been created around 300 BCE. Namoratunga II, which translates to “people of stone” in the Turkana language prominently features a formation of stone encircling 19 basalt pillars

Page 220220 Legon Journal of the HumanitiesHumanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole:Whole: EvidenceEvidence fromfrom Archeoastronomy Archeoastronomy arranged in alignment with the star systems of Aldebaran, Bellatrix, Central Orion, Pleiades, Saiph, Sirius and Triangulum. The basalt pillars are thought to be linked to the constellations associated with the 12-month lunar calendar of 354 days used by Cushitic speakers of southern Ethiopia (Lynch & Robbins, 1978). Cosmological symbolisms of directional orientations are also evident among the Dogon people of Mali as found in Conversations with Ogotemmeli, a collection of philosophical teachings reported by Griaule from Ogotemmeli in the 1930s (Griaule, 1934 (1970)). Implicit in the structural positioning of a Dogon community is the knowledge of cardinal positions as illustrated by Griaule. In his illustration of the village, we see that the village is oriented towards the north. The village is personified as a person laying on their back with the head facing north. Griaule reports that Dogon beds are oriented towards a N-S position where the head always faced North. This cardinal positioning also informed the sleeping positions and the burial positioning of both genders. Griaule reports “The man lies on his right side facing west, and the woman on her left side facing east, which are the positions they will occupy in the grave” (Griaule, 1934 (1970), p. 95). This intentional structural positioning of the bed, grave and village reflects, as noted above, the incorporation and complementarity of scientific knowledge with that of cultural practices as well as stressing analogical reasoning as a human is geographically and architecturally defined. While the orientations are different, we find the shared worldview of ensuring specific placement with regard to cardinal directions in the grave transcends time dating back to the late Pleistocene era. Indeed, the Jebel Sahaba cemetery, a Mesolithic site dated to 14,340 to 13,140 years and located in the Nile Valley in Nubia (Northern Sudan) is notable for Afrikans buried in the fetal position facing south and with their heads oriented toward the east as shown in Figure 1 (Graham, 2016).

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 221 221

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019))

Figure 1: Map of cemetery 117 at Jebel Sahaba: Afrikans buried in the fetal position facing south and with their heads oriented toward the east (Image Credit British Museum, 2019) Orientations toward significant cardinal directions is also evinced in other parts of the continent as we will show below. With regard to the south of Kmt ‘land of Black people’ it is also important to note that the orientation of the map for Afrikans was South-Up. In mdw nṯr, it is clear that iȝbt, the word for East is the same as the word for left, and the word imn ‘West’, is the same as the word for right as well as all other derivative words. Further, šmˁw ‘Upper Kmt’ was synonymous with rsy ‘the South’ and mḥw ‘lower Kmt’ was associated with mḥtt ‘North’ (Carruthers, 1989, pp. 22-23). This spatial orientation can be seen in what is regarded by cartographers as the oldest map of topological significance known as the Turin Papyrus Map created by scribe imnnḫt ‘Amennakhte’ ca. 1150 BCE. This map has a South-Up orientation showing a 15- kilometre stretch to the mines of Wadi Hammamat (Harrell & Brown, 1992a, 1992b; McMahon, 1992). According to Carruthers (1999):

Page 222222 Legon Journal of the HumanitiesHumanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole:Whole: EvidenceEvidence fromfrom Archeoastronomy Archeoastronomy Our ancestors, the ancient Kemites (i.e., Egyptians)4 equated up with South and down with North. Upper Kemet (Egypt) is in Southern Egypt and Lower Egypt is in the North. The Nile flows down from the Equator to the Mediterranean Sea. Pharaoh Jehewty-Mose asserted: "My southern boundary is at the top of the world." Quite possibly the Kemites considered the Equator as the top of the world (p. 262). This reflects a Ra-centered view based on the ecliptic in that “from the perspective of a person in the northern hemisphere facing the equator, the sun would appear to rise on the left and set on the right. From the position of a person in the southern hemisphere, however, the sun would appear to rise on the right and set on the left” (Kambon, 2017, p. 24). Archeoastronomy as Evidence of Complementarity between “The Humanities” and “The Sciences” The method adopted in this study examines alignments of structures of Kmt ‘land of Black people’ with respect to solar and celestial phenomena in an attempt to establish systematic patterns and correlations between architectural alignments and astronomical phenomena. This method is adopted to show the relationship between structures constructed and knowledge of the oneness of humanities and sciences. According to Seba Montgomery, “In Kemet, the temples stand as testimony to the importance and the functionality of their star knowledge (Montgomery, 2017, p. 4). In a similar vein, Tata Obenga states that “Egyptian temples, like all such places of worship as lakes, shrines, pyramids, obelisks and palaces, were also expected to conform to the cosmic order of Maat. So the axes of sacred buildings were aligned with the energy vectors of the universe. Ancient Egyptians were always driven by a passionate desire to achieve unity with the entire cosmos” (Obenga, 2004, p. 604). Indeed, we find reflected in all aspects of the life of the Kmtyw ‘Black people, people of Kmt’ the intimate knowledge of the movement of the stars and sun. This is particularly clear when we examine various architectural orientations vis-à-vis astronomical events. These astronomical events have been grouped into seven families by Belmonte et al. (2009) after an analysis of about 330 temples. This categorization is helpful for understanding these orientations as a whole complex which grew out of a worldview that sought to be in alignment with the cosmos as opposed to viewing each building as an isolated and disconnected incidence of architectural anomalies. Indeed, the orientational complex including equinoctial, solstitial, seasonal, Sopdet, Canopus, and the (mid) quarter-cardinal family accounts for the majority of the orientations of temples and other structures of ancient Kmt ‘land of Black people’. Thus, we will look

4 Better translated literally as Black people.

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 223 223

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019)) at each of these families below providing a brief description before giving a more detailed account with examples following the typology developed by Belmonte et al. (2009): 1. The “equinoctial” or eastern family: Buildings in this group are oriented towards the “equinoctial sun when the disk has completely risen above the horizon” (Belmonte et al., 2009, p. 227). The equinox occurs on March 20th or 21st and September 22 or 23. For one to orient architecture towards the equinoxes it implies the knowledge of the cardinal points which, according to Belmonte et al., stems from the explicit “observation of certain configuration of stars and constellations in the northern skies, notably Meskhetyu” (p. 227). 2. The solstitial family: In this group, structures are oriented towards

sunrise during ˁȝpy ‘aApi/Summer Solstice’ and

nn ‘Nen/Winter Solstice’. Aapi occurs during June 20 or 21 whereas Nen occurs during December 21 or 22. The orientation of structures like the mrw ‘pyramids’ of Giza and the temple of ipt swt ‘Karnak’ towards the sun during the Aapi and Nen solstices is evidence of the astronomical knowledge of Afrikans=Black people. Figure 2 gives a useful illustration of the aforementioned equinoxes and solstices.

Figure 2: Equinoxes and Solstices (Weather.gov, 2018)

Page 224224 Legon Journal of the HumanitiesHumanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole:Whole: EvidenceEvidence fromfrom Archeoastronomy Archeoastronomy 3. The seasonal family: Buildings in this group were specially constructed to mark and record seasonal changes. ȝḫt ‘Inundation Season’ marked the flooding of the Nile (mid-July-mid-November), Prt ‘Planting Season’ marking the time for sowing (mid-November-mid-March) and Smw ‘Harvest Season’ marking the time for harvesting in the calendar of the Kmt(yw) ‘Black people, people of Kmt’ (Mid-March-Mid-July).

Figure 3: ḥr m ȝḫt ‘Great Sphinx’ at the Summer Solstice (Photo Credit Belmonte & Shaltout, 2010, p. 535) The placement of ḥr m ȝḫt ‘Great Sphinx’ at the mrw ‘pyramids’ of sȝ rˁ ḫwfw ‘Khufu’ and sȝ rˁ ḫˁ.f rˁ ‘Khafre’ during Aapi

or the June solstice captured in an alternative pictorial representation of ȝḫt ‘Akhet/horizon’ as depicted in mdw nṯr, the sun rising between two mountains.

Allen (2015, p. 10) traces the etymology of the word ȝḫt ‘Akhet/horizon’ to ȝḫ, ‘an effective spirit’. Thus, he argues that the eastern horizon, was known as the Akhet, meaning “Place of Becoming Effective.” This view is expounded upon by Browder (1992) as he articulates that: Viewed from the Sphinx at the time of the summer solstice when the sun is at its greatest distance north of the celestial equator, it sets directly between the pyramids of Khufu and Khafre, thus writing across the horizon an akhet ‘on the scale of acres’. This is a clear reference to Horus [Heru] the god and the akhet sign, because anyone approaching from ancient Memphis would have seen the head of the Sphinx silhouetted between the two pyramids (p. 115).

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 225 225

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019))

Figure 4: Light of the rising sun enters the sancta sanctorum of the main temple of Abu Simbel (Belmonte et al., 2009, p. 233) Indeed, the indigenous term describing this monumental carving out of solid stone in mdw nṯr is ḥr m ȝḫt ‘Great Sphinx’, literally ‘Heru in the horizon’ displaying one of the world’s oldest and largest manifestations of hierophany which can still be seen today as shown in Figure 3. Another such instance of hierophany can be viewed in the temple at Abu Simbel dedicated to Rˁmssw mry imn ‘Ramessu Mery Amen’, which provides an example of a building oriented towards the heliacal rising, signalling the beginning of and šmw seasons as shown in Figure 4 and as documented by Belmonte, et al. (2009), worth quoting in full below: At dawn on 22 February 2004, the light of the rising sun enters the sancta sanctorum of the main temple of Abu Simbel. The first rays illuminate the figures of Amon- Re, the divinized king and the shoulder of Re-Horakhty, all of them gods of solar character, while the figure of Ptah, god of the netherworld, stays in darkness. When the light progresses, the figure of Re-Horakhty receives first the full and afterwards the last rays of the sun. This marvellous hierophany may have occurred at the beginning of the Peret and Shemu seasons of the ancient Egyptian calendar, during the first decades of the reign of Ramses II, the temple builder (Belmonte et al., 2009, p. 231). 4. The Sopdet family: Buildings in this family were oriented towards

the rising of the brightest star in the Kmtian sky, spdt ‘Sopdet/Sirius.’ Belmonte, et. al (2009) put in this category 3 3 structures with a declination of -17 ⁄4° and -53 ⁄4° an example being

Page 226226 Legon Journal of the HumanitiesHumanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole:Whole: EvidenceEvidence fromfrom Archeoastronomy Archeoastronomy

the temple dedicated to Ist ‘Aset/Isis’ located behind the temple of ḥwt ḥr ‘Het-Heru/Hathor’ at Dendara. 5. The Canopus family: Belmonte, et. al (2009) include in this family buildings oriented towards Canopus, the second brightest in the

sky of Kmt ‘land of Black people’ after spdt ‘Sopdet/ Sirius’. An example of such a building with this orientation is the

temple of Ist ‘Aset/Isis’, a part of the complex of temples at Philae. 6. The Meridian family: Buildings in this family were aligned with “certain configurations of stars near the celestial pole” (Belmonte et al., 2009, p. 237). We are informed that the circumpolar constellation of msxtyw ‘Meskhetyu’ received a great deal of attention through the orientation of the buildings, making it almost inseparable from the first family (Belmonte et al., 2009). The temple of ḥr(w) ‘Heru’ at Edfu is an example. 7. The (mid) quarter-cardinal family: According to Belmonte, et al. (2009) “The quarter-cardinal family is defined as a group of temples whose orientation is close to 45° (NE), 135° (SE), 225° (SW) or 315° (NW).” He further argues that these produce “a peak at -39°±¾° (VII—) and its symmetric at some 40° (VII+)” (Belmonte et al., 2009, p. 238). The New Kingdom Million Year Temples of Abydos and Western Thebes are good exemplars of this group. Having considered the above, we obtain a clear indication that astronomical knowledge complemented and informed the architectural practices of Kmt(yw) ‘Black people, people of Kmt’. Solar Alignments As shown above in Figure 4, an exemplification of the marriage between now disparate Sciences and Humanities can be seen in another instance of hierophany. What is significant about hierophany as enshrined in architecture is the fact that its accomplishment is as a result of precise calculations such that Rˁ ‘the sun’ will shine his light into a specifically constructed space on specific dates/times at significant celestial/solar events. That is to say, mathematical

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 227 227

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019) precision serves ultimately humanistic purposes. A further exemplification of this principle can be seen in the example in Figure 5:

Figure 5: Light hierophany at sunrise at the winter solstice (Photo Credit Belmonte et al., 2009, p. 226) Belmonte (2009, p. 231) gives a thorough articulation of the effect achieved in the temple of Sobek-Ra as follows: A splendid light hierophany at sunrise at the winter solstice (December 21st 2006) at the temple of Sobek-Re in Qsar Qarun (Dionysas) in El-Fayum Oasis. Notice the illumination of the inner sanctuary (left), where the statue of the god would have been located, and of the various successive portals. The winged-disc symbol of Horus Behedety blossoms above the first gate at the pale yellow light of the first rays of his physical counterpart, the solar-disk. As the sun rises, its light moves to the north illuminating the small shrine to the right of the sanctuary while the niche to its left, where the mummy (the underworld aspect) of the god was presumably located, always remains in darkness. Notice the parallelism with the illumination phenomenon at Abu Simbel (p. 231) Whereas in Eurasian thought, so-called religious complexes are not seen as being intrinsically related to math and science, in Kmt ‘land of Black people’, the astronomical alignment of spiritual structures were measured and considered from their very inception via the stretching of the cord ceremonies, ensuring accuracy of placement and orientation (Montgomery, 2017). Yet another example of this phenomenon can be seen in the orientation of the colossi of Amenhotep III, which have faced the sunrise at the time of the winter solstice for 3,350+ years in

Page 228 228 Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole: Whole: Evidence Evidence from from Archeoastronomy Archeoastronomy alignment with the temple of millions of years (Belmonte et al., 2009, p. 216).

Figure 6: sunrise at the time of the winter solstice (Photograph courtesy of M. A. Molinero and N. Delgado) in (Belmonte et al., 2009, p. 216) Yet another example of solar alignment can be found in the small temple of Aton at Akhetaton. In Figure 6, we can see clearly that the temple is oriented toward a cleft in the horizon where the royal tomb was excavated, which is the precise spot where Rˁ ‘the sun’ rises in late February and October. This alignment is thought to be the impetus for the name Akhetaton ‘Horizon of Aton’ (Belmonte et al., 2009, p. 232).

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 229 229

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019)) Figure 7: the temple is oriented toward a cleft in the horizon where the royal tomb was excavated, which is the precise spot where the sun rises in late February and October (Photograph courtesy of M. Gabolde) in (Belmonte et al., 2009, p. 232) While one may be inclined to think that temple orientations such as those mentioned above were perhaps random or happenstance, as can be seen from the graph above, out of 330 temples surveyed by Belmonte, et. al (2009, p. 223), there are clear and obvious statistically significant peaks near the ‘equinoctial’ sunrise (due-East) and the winter solstice (WS) sunrise, which is equated to the due-North orientation skewed by 90°. There are also clear peaks at SE and NE showing clear evidence of intentional astronomical/solar orientations of temples as shown in Figure 8.

Figure 8: Chart illustrating 330 Solar and Celestial Temple Orientations (Belmonte, Shaltout, & Fekri, 2009, p. 223)

We will now turn our attention to the orientation of temples vis-à-vis celestial phenomena. In Belmonte et al.’s research, there were definite correlations between temples associated with female nṯrwt ‘female divinities’ and spdt ‘Sopdet/Sirius’. Likewise, there were clear orientations toward equinoxes and solstices for solar-related nṯrw ‘male divinities.’ This is shown in Figure 11, which illustrates the absolute declination histograms 34 temples of nṯrwt ‘female divinities’ and 42 temples of nṯrw ‘male divinities’ of solar character, respectively (Belmonte, et. al 2009: 226).

Page 230 230 Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole: Whole: Evidence Evidence from from Archeoastronomy Archeoastronomy

Figure 9: The complex of the temple of Figure 10: “Absolute declination histograms of two Hathor at Dendara (Belmonte et al., 2009, independent series of temples. (a) 34 temples of goddesses p. 235) of all periods. (b) 42 temples of divinities with a solar character. Notice the most significant peak corresponding to Sopdet family for series (a)” (Belmonte et al., 2009, p. 226)

Figure 9, showing the temple complex for ḥwt ḥr ‘Het-Heru/ Hathor’ at Dendara shows the main building, orientated north, in the direction of msḫtyw ‘Meshketyu’ in the direction that it was located in the late Ptolemaic period (Belmonte et al., 2009, p. 234). The small temple dedicated to

Ist ‘Aset/Isis’ is also observed to be aligned to the heliacal rising of spdt ‘Sopdet/Sirius as it is not exactly perpendicular to the main temple (Belmonte et al., 2009, p. 234). The significance of these alignments is explained in detail as follows: The god Sah personified the constellation of Orion - the most distinctive of all the constellations in the night sky. While not part of the 'imperishable' circumpolar stars, the constellation became important in Egyptian mythology especially as it

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 231 231

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019)) rose directly before the adjacent star Sirius (the Egyptian Sothis) – the brightest fixed star which was utilized in the calculation of the Egyptian calendar. The constellation god was thus connected with the star Sothis from an early date and the two came to be viewed as manifestations of Osiris and Isis respectively (Wilkinson, 2003, p. 127).

Figure 11: Three stars in Orion’s belt pointing through Sirius to the location of the Sunrise at the Winter Solstice (King, 2018)

Interestingly, the three (3) stars in the belt of sȝḥ ‘Orion’, which was associated with isir ‘Osiris’ point to spdt ‘Sopdet/Sirius’, which was associated with Ist ‘Aset/Isis’, the wife of isir ‘Osiris’, and this line points to where the ḥrw ‘Horus/The Sun’, associated with the sun, rises at dawn during the winter solstice in the south-eastern sky as shown in Figure 11. Thus, we find that temples to Ist ‘Aset/Isis’ are orientated towards spdt ‘Sopdet/Sirius’, as shown in Figure 10 as this is the star system with which she was associated. Similarly, as also shown in Figure 10, temples associated with solar divinities are oriented toward ḥrw ‘Horus/The Sun’.

Page 232 232 Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole: Whole: Evidence Evidence from from Archeoastronomy Archeoastronomy

Figure 12: Sopdet and Sah on the astronomical ceiling of the tomb of Senenmut in Deir el Bahari (Photo Credit Belmonte et al., 2009, p. 217)

We find this association between Ist ‘Aset/Isis’, spdt ‘Sopdet/ Sirius’ and monuments oriented toward Sirius in other instances as well. As noted by Belmonte et al. (2009), “The goddess Sopdet standing for Sirius, the brightest star of the Egyptian firmament, and the god Sah (southern part of Orion), as beautifully depicted in the astronomical ceiling of the tomb of Senenmut in Deir el Bahari, one of several monuments oriented towards Sirius in the geography of Egypt” (p. 217).

Figure 13: Original “Immaculate Conception” concept of Ist as kite alighting on the phallus of Asr with the spirit form of Hrw to the left at the temple of Seti I. (Tausch, 2011)

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 233 233

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019)) Here, again, we see the link between the so-called sciences (i.e., astronomy) and the so-called humanities in architecture as well as in other depictions of the astronomical phenomena. These connections are highlighted in that the interrelation between isir ‘Osiris’ and ist ‘Aset/Isis’ to produce their son, ḥrw ‘Horus/The Sun’ was depicted as a virgin birth as illustrated in Figure 13. The story of the original “immaculate conception”, which is an allegory referring to the Sun’s Death and Resurrection at the Winter Solstice is found in the world’s oldest spiritual texts, the Pyramid Texts, where it is stated that: “Isis comes to you rejoicing through love of you; your seed issues into her, she being ready as Sothis. Ḥar-Sopd has issued from you in his name of ‘Horus who is in Sothis.’” (Faulkner, 1969, p. 244) Pyramid Alignments Before delving into the alignments between pyramids and celestial and/or terrestrial phenomena, it behooves us to first take into account that the construction of the pyramids was a set of fundamentally Humanistic projects that were at the same time, the world’s first and largest recorded testable science experiments. Articulated more clearly, the goal of the pyramids was to send the ȝḫ ‘ effective spirit’ of the deceased ruler to the imperishable stars by constructing structures in alignment with those stars that never set below the horizon. The idea was that on the premise of “As above, so below,” since stars have an impact on the lives of those on earth, if the ruler in spirit form could go and join these stars that are always seen in the firmament, then the ruler could continue to advocate for and intercede on behalf of those still on earth. The pyramid texts are replete with references to the ruler joining the imperishable stars to these ends. The following texts come from the oldest known of all of these, the pyramid text of nswt bity Wnis, inscribed within approximately 2345-2315 BCE: Humanity will cry out to you once the Imperishable Stars have raised you aloft. (Allen, 2015, p. 34) See me, as you have seen the forms of the progeny who know their spells, the Imperishable Stars, and see (in me) the two in the palace—that is, Horus and Seth. (Allen, 2015, p. 34) The portals act for me, the (Red Crown’s) coil is tied on for me, and I lead the Imperishable Stars. (Allen, 2015, p. 52) I go up on Isis’s thighs, I climb on Nephthys’s thighs. My father Atum takes my arm and allots me to those gods who are wise and experienced, the Imperishable Stars (Allen, 2015, p. 53)

Page 234 234 Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole: Whole: Evidence Evidence from from Archeoastronomy Archeoastronomy In addition to temples aligned with solar and celestial phenomena, massive mrw ‘pyramids’ are also demonstrated to have been constructed with precise scientific and mathematical precision. This is shown in Figure 14. Quoting Browder (1992), One of the most precisely constructed passages in the pyramid is the Descending Passage […] Two shafts were cut through 200 feet of solid masonry to the outer surface of the pyramid. These shafts, which face in a northerly and southerly direction, have long been regarded as "air shafts" that ventilated the King's Chamber. New evidence has been presented which shows the so-called air shafts are inclined, within one degree of accuracy, to the northern celestial pole, and to the three stars of Orion's Belt to the south. The main star viewed through the northern shaft is called Alpha Draconis. Around it turned the Circumpolars (often railed the "Indestructibles" because they never disappeared below the horizon). It was for this reason that these stars symbolized immortality. A soul which had chosen this path in life was said to ascend into the imperishable dominion of the northern heavens (Browder, 1992, pp. 108-109). These alignments have led several scholars to posit dates for the construction of the great pyramid of Giza based on when it has been calculated that the so-called air shafts would have been pointing to these stars as shown below: Table 1: Dating of the Great Pyramid based on Celestial Alignments

Author Date using shafts

Virginia Trimble (1964) 2600 BCE

Alexander Badawy (1965) 2600 BCE

Robert Bauval & Gilbert (1994) 2450 BCE +/- 25

Mary Bruck (1995) 2450 BCE +/- 60

Kate Spence (2000) 2478 BCE +/- 5

Figure 14: Great pyramid celestial alignments (McKenty, 2000)

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 235 235

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019)) The calculations made as presented in Table 1 and illustrated in Figure 14 give insight into the relationship between areas contemporarily known to be within the area of the so-called Humanities and those of the so-called Sciences. Interestingly, recent research has shown that not only were the pyramids constructed with reference to celestial phenomena, they were also aligned with each other and other terrestrial phenomena. In other words, archaeoastronomy is further complemented by what we may term archaeogeography. This can be seen clearly in the pyramidal alignment pointing to iwnw ‘Heliopolis’ as shown in Figure 16 and Figure 17.

Figure 15: Intra-Alignment of Great Pyramids of Giza (Magli, 2011, p. 6)

Figure 16: Alignment of Great Pyramids with reference to the txn ‘obelisk’ of iwnw ‘Heliopolis’ (Magli, 2011, p. 6)

Page 236 236 Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole: Whole: Evidence Evidence from from Archeoastronomy Archeoastronomy Even more remarkably, these alignments are not only restricted to the pyramids and iwnw ‘Heliopolis’, but extend far afield to monuments throughout the area, which are oriented to point toward iwnw ‘Heliopolis’, the ancient spiritual capital and location of the temple of Rˁ.

Figure 17: The geographical area of interest: H=Heliopolis, K= Cairo Citadel, A=Abu Roash, G=Giza, Z=Zawiet el Arian, B=Abu Gorab, S=Abu Sir, Q=Saqqara main field, M=Memphis (Magli, 2009, p. 4)

Figure 18: The pyramids, the Sun Temples and the Heliopolitan ‘diagonals’: (1) Abu Roash diagonal, (2) Giza diagonal, (3) Abusir diagonal. The dotted straight line denotes invisibility. The stepped line denotes the maximal Nile flood (Magli, 2009, p. 7)

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 237 237

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019)) Making the point even more vividly is the fact that there is no direct line of sight between Abu Sir and Heliopolis, meaning that the calculations had to be reckoned mathematically. Conclusion We argue that our acceptance of a compartmentalized false dichotomy between the so-called Humanities and the Sciences constitutes pale white Eurasian- induced conceptual incarceration which impedes a holistic Afrikan view of complementary aspects of a necessary whole. For Afrikans=Black people, astronomy and associated mathematical calculations influenced architecture, literature, etc. Further, the so-called Humanities (spirituality, philosophy, literature, etc.) influenced how astronomy was understood. As such, a recognition of the interrelations and interdependence of the Humanities and the Sciences is not a future ideal to be attained, but rather a past reality which, once known, can inform our trajectories and actions. The significance of this knowledge is that as unapologetically committed Afrikan=Black researchers, our commitment is that towards the re-membering of Afrikan people throughout space, time, energy and matter. This endeavour requires the concomitant re-membering of Afrikan=Black knowledge production systems and their application in our reality. This will be a great contribution to the total liberation of Afrikans wherever we may be. In such a Sankɔfa ‘Go back and fetch it’ movement, the so-called Sciences and the so- called Humanities should no longer necessarily be viewed as disparate phenomena, but rather be seen once again as complementary aspects of an Afrikan=Black whole for the restoration of Mȝˁt on Earth as she exists in the Celestial Sphere.

Page 238 238 Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole: Whole: Evidence Evidence from from Archeoastronomy Archeoastronomy References Allen, J. P. (2015). The Ancient Egyptian Pyramid Texts. Atlanta: SBL Press. Ani, M. (1994). Yurugu: An African Centred Critique of European Cultural Thought and Behavior. Trenton, NJ: Africa World Books. Badawy, A. (1965). Ancient Egyptian architectural design: a study of the harmonic system (Vol. 4). Oakland, CA: University of California Press. Bauval, R., & Gilbert, A. (1994). The Orion mystery: unlocking the secrets of the pyramids. New York: Three Rivers Press. Beatty, M. (1997). Maat: The cultural and intellectual allegiance of a concept. In J. H. Carruthers & L. C. Harris (Eds.), The preliminary challenge: World history project (pp. 211-244). Los Angeles: Association for the Study of Classical African Civilizations Kemetic Institute. Belmonte, J. A., & Shaltout, M. (2010). Keeping Ma’at: An astronomical approach to the orientation of the temples in ancient Egypt. Advances in Space Research, 46(4), 532-539. Belmonte, J. A., Shaltout, M., & Fekri, M. (2009). Astronomy, landscape and symbolism: a study of the orientation of ancient Egyptian temples. In search of cosmic order: Selected essays on Egyptian archaeoastronomy, 1, 213. Brophy, T. G., & Rosen, P. A. (2005). Satellite imagery measures of the astronomically aligned megaliths at Nabta Playa. Mediterranean archaeology and archaeometry, 5(1), 15-24. Browder, A. T. (1992). Nile Valley Contributions to Civilization: Exploding the Myth (Vol. 1). Washington, DC: Institute of Karmic Guidance. Bruck, M. T. (1995). Can the Great Pyramid be astronomically dated? Journal of the British Astronomical Association, 105, 161-164. Carruthers, J. H. (1989). Essays in ancient Egyptian studies. Los Angeles, CA: University of Sankore Press. Carruthers, J. H. (1999). Intellectual Warfare. Chicago: Third World Press. Chandler, W. B. (2000). Ancient Future: The Teachings and Prophetic Wisdom of the Seven Hermetic Laws of Ancient Egypt. Baltimore, MD: Black Classic Press. Diop, C. A., & Cook, M. (1989). The African Origin of Civilization: Myth Or Reality. Chicago: Chicago Review Press. Etymonline. (2018). Science. Retrieved from https://www.etymonline.com/word/ science

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 239 239

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019)) Faulkner, R. O. (1969). The Pyramid Texts. Oxford: Oxford University Press. Finch, C. (1998). The star of deep beginnings: The genesis of African science and technology. Decatur, GA: Khenti, Incorporated. Graham, K. (2016, 21 January 2019). Evidence of earliest massacre found dates back 10,000 years. Retrieved from http://www.digitaljournal.com/ science/evidence-of-earliest-massacre-found-dates-back-10-000-years/ article/455457#ixzz67UOA6uJY Griaule, M. (1934 (1970)). Conversations with Ogotemmeli: An introduction to Dogon religious ideas. Oxford: Oxford University Press. Harrell, J. A., & Brown, V. M. (1992a). The oldest surviving topographical map from ancient Egypt:(Turin Papyri 1879, 1899, and 1969). Journal of the American Research Center in Egypt, 29, 81-105. Harrell, J. A., & Brown, V. M. (1992b). The World's oldest surviving geological map: the 1150 BC Turin Papyrus from Egypt. The Journal of Geology, 100(1), 3-18. Hilliard, A. G. (1998). SBA: The reawakening of the African mind: Makare Pub Co. Kambon, O. (2017). Akan Ananse Stories, Yorùbá Ìjàpá Tales, and the Dikenga Theory: Worldview and Structure. Contemporary Journal of African Studies, 4(2), 1-36. King, B. (2018). The Winter Triangle Meets The Preposterous Monoceros. Retrieved from https://astrobob.areavoices.com/2018/02/11/the-winter-triangle-meets- the-preposterous-monoceros/ Kreamer, C. M. (2012). African cosmos: stellar arts. New York: Monacelli Press. Lynch, B., & Robbins, L. (1978). Namoratunga: The first archeoastronomical evidence in sub-Saharan Africa. Science, 200(4343), 766-768. Magli, G. (2009). Topography, astronomy and dynastic history in the alignments of the pyramid fields of the Old Kingdom. arXiv preprint arXiv:0903.1416. Magli, G. (2011). Reconstruction of ancient conceptual landscapes in the Nile Valley. arXiv preprint arXiv:1104.1785. Martin, D. (2008). Maat and Order in African Cosmology: A Conceptual Tool for Understanding Indigenous Knowledge. Journal of Black Studies, 38(6), 951- 967. doi:10.2307/40035033 McK Malville, J., Schild, R., Wendorf, F., & Brenmer, R. (2007). Astronomy of Nabta Playa. African Skies, 11, 2.

Page 240 240 Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019)

Kambon, B.Ọ. O.B. & Asare, Y. M./M./HumanitiesHumanities andand SciencesSciences asas ComplementaryComplementary AspectsAspects ofof anan Afrikan=BlackAfrikan=Black Whole: Whole: Evidence Evidence from from Archeoastronomy Archeoastronomy McKenty, R. (2000). Are The Pyramids Of Giza Aligned With The Sun And The Stars? Yes. Retrieved from http://www.merriam-webster.com/dictionary/ complementary McMahon, D. (1992). The Turin Papyrus Map The Oldest Known Map with Geological Significance. Earth Sciences History, 11(1), 9-12. Merriam-Webster. (2018a). Complementary. Retrieved from http://www.merriam- webster.com/dictionary/complementary Merriam-Webster. (2018b). Humanities. Retrieved from http://www.merriam- webster.com/dictionary/humanity Montgomery, A. B. (2017). An Examination of Ra Emerging from Nu. Words Spoken by the Imy-Wenewity. The African Star-Watchers Who Invented “Time!”. Detroit: lulu.com / the Medew Netcher Study Group, Inc. Museum, B. (Cartographer). (2019). Map of cemetery 117 at Jebel Sahaba [ ]. Retrieved from https://blog.britishmuseum.org/violence-and-climate- change-in-prehistoric-egypt-and-sudan/ Obenga, T. (2004). African Philosophy: The Pharaonic period, 2780-330 B.C. Popenguine, Senegal: Per Ankh. Spence, K. (2000). Ancient Egyptian chronology and the astronomical orientation of pyramids. Nature, 408(6810), 320-324. Stevenson, A. (2010). Oxford Dictionary of English. Oxford: OUP Oxford. Tausch, O. (2011). Abydos Tempelrelief Sethos I (pp. Deutsch: Relief der Erweckung des Gottes Osiris durch die Göttin Isis im Totentempel Sethos I. in Abydos, Ägypten Čeština: Eset v podobě ptáka kopuluje se zesnulým Usirem. Na levé straně je Hor, přestože v té době ještě nebyl narozen, a na straně druhé Eset v lidské podobě. Chrám Sethiho I., Abydos).Web: Wikimedia Commons. Trimble, V. (1964). Astronomical Investigation Concerning the So-Called Air- Shafts of Cheops' Pyramid. Mitteilungen der Institut Fur Orientforschung, 10(2/3), 183-187. Van Sertima, I. (1983). Blacks in science: Ancient and modern. New Brunswick, NJ: Transaction Publishers. Van Sertima, I. (1994). Egypt: Child of Africa. New Brunswick, NJ: TransactionPublishers. Waite, M. (2009). Oxford Thesaurus of English. Oxford: OUP Oxford.

Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Page 241 241

Kambon, B.Ọ. O.B. & Asare, Y. M./M./LegonLegon JournalJournal ofof thethe HumanitiesHumanities Vol.Vol. 30.2 30.2 (2019 (2019)) Wilkinson, R. H. (2003). The Complete Gods and Goddesses of Ancient Egypt. London: Thames and Hudson. Womack, Y. (2013). Afrofuturism: The world of Black sci-fi and fantasy culture. Chicago: Chicago Review Press.

Page 242 242 Legon Journal of the Humanities Vol. 30.230.2 (2019)(2019) Kuwabong, D./ Philosophical Foundations of the African Humanities through Postcolonial Perspectives BOOK REVEW

Philosophical Foundations of the African Humanities through Postcolonial Perspectives Edited by Helen Yitah and Helen Lauer, Brill Rodopi, 2019, pp. I-XVI; 1-297. ISSN 0924-1426 / ISBN 978-90-04-37759-2 (hardback) / ISBN 978-90-04-39294-6 (e-book) Dannabang Kuwabong Professor English Department University of Puerto Rico, San Juan. Email: [email protected]

Philosophical Foundations of the African Humanities through Postcolonial Perspectives, edited by Helen Yitah and Helen Lauer of the University of Ghana is one more excellent scholarly collection of essays from well-established and internationally acclaimed scholars in their respective fields. The book follows in the footsteps of the impressive 2012 publication, Reclaiming the Human Sciences and Humanities through African Perspectives, vols I & II, edited by Helen Lauer and Kofi Anyidoho. This collection, in addition to the 2012 collections, makes me believe that at long last, our universities are creating new scholarly foundations that will enable scholars in Ghana to take control of their own scholarship, rather than continuing to depend on materials that are published outside the reach of Ghanaian academics and students. So let me offer my thanks to the editors upfront for the great work this book is, and how important it will be for those Ghanaians outside who want to read works produced by scholars in Ghana. It demonstrates that the vision of the editors and authors of these essays to device ways in which to argue for the decolonization of the humanities in African universities is bearing fruit. I am particularly happy that the essays are transgenerational and transdisciplinary positions on better ways of addressing the study of the humanities in African universities. The interlacing of multiple areas and perspectives on decolonizing and recuperating African voices in the humanities, be it in challenging reading habits in literature, literary theory, the problems of neocolonialism, structural weakness in research funding and knowledge dissemination, the deceptions of neo- liberal capitalism, research methodologies, cultural disarticulation, language studies,

Legon Journal of the Humanities Vol. 30.2 (2019) Page 243 Kuwabong, D./Legon Journal of the Humanities Vol. 30.2 (2019) performance arts, spirituality, art, healing practices, music, dance, and theory inter alia makes reading the book exciting. One is never bored. Moreover, though this is a collection of essays by academic heavy weights, such as Gayatri Chakravorty Spivak, J. H. Kwabena Nketia, Kwasi Wiredu, M. E. Kropp Dakubu, Helen Lauer, Gordon S. K. Adika, John Collins, Alexis B. Tengan, Francis Nii-Yartey, Olúfẹ́mi Táíwò, Abena Oduro, James Gibbs, and Bernth Lindfors, the essays are easy to read and digest. Their readability does not mean the absence of serious scholarship nor rigorous research. Even the last essays, which are celebratory remembrances of some of the contributors, are so well written they read like short academic biographies, and contribute to the overall excellence of the book. What then is Philosophical Foundations of the African Humanities through Postcolonial Perspectives all about? What is its contribution to the growing collection of books that I believe are consolidating African based contributions to the humanities and housed in African universities? The introductory essay, “Introduction: Mediating a Hapless Postcolonialism through Experiential Critique,” by Helen Yitah, brilliantly brings together the main arguments, strategic debates, and liberatory maneuvers posited by the authors to define what African scholars need to do to achieve an Africanized and African-centred humanities programs. A better introduction could not have been written. Yitah undertakes incisive reflections of these complex essays and translates them into readable summaries that make the reading of the essays better appreciated. Additionally, I find the arrangement of the essays very apt. Making the transcribed theoretical conversation between Gayatri Chakravorty Spivak and doctoral students titled “Philosophical Foundations of the Humanities” as the opening chapter, is ingenious. The dialogue with the students at least tangentially touches on almost all the issues raised by the subsequent essays. Her conversation takes on the title of the book and questions whether philosophy can ever play the role of being the foundation of the humanities. She rejects position such as onto-phenomenological positions which humanities scholars are not competent to handle. Likewise, her position argues against any pretense to the idea of universality in the humanities. Here she raises the specters of gendered, racialized, and ethnocentric learning praxes that can turn into “unexamined culturalism” and the dangers of cultural nationalism through which marginal voices and identities are subsumed under the illusion of universality and linguistic neutrality in order to silence difference. She calls for academics in postcolonial or neocolonial academic institutions not to be sucked in by false dichotomies nor remain passive recipients of knowledge transmission, but to engage in emerging/emergent resistance to counter the psychological mechanism used by neo-imperialism.

Page 244 Legon Journal of the Humanities Vol. 30.2 (2019) Kuwabong, D./ Philosophical Foundations of the African Humanities through Postcolonial Perspectives Awoonor’s piece, “The Humanities and the Postcolonial Ghanaian Experience: The Jubilee Year” probably written a few years earlier at Ghana’s golden jubilee year after the establishment of the Institute of African Studies, by Kwame Nkrumah, touches on pioneer Ghanaian political thinkers such as Casely- Hayford and Kobena Sekyi. He contrasts the two thinkers and gives a brief review of their creative work. In this review, he seems to list J. Casely-Hayford among the cultural nationalists, who advocated indirect rule, and Sekyi his opposite. Awoonor also rejects the concept of the universality of knowledge or of knowing, and challenges the institutions in Ghana to open up to new frontiers of learning. His position is based on the failures of the Institute of African Studies set up by Kwame Nkrumah in 1965 to live up to its mandate. Helen Lauer in “Beyond the Labor Market: Reinforcing the Epistemic Advantage of African Universities in the Global Knowledge Society,” picks up where both Spivak and Awoonor leave off, to articulate the need for the equitable diversification of knowledge, especially in the age of digitization and artificial intelligence, compounded by neo-liberal economics that controls project funding which often leaves out scholars in African institutions. Thus, a point she raises which is pertinent to this book’s central focus is how to break away from the paralysis caused by what she calls the mechanistic capitalism that acts as a gatekeeper to knowledge. Arguing for a multidirectional flow of knowledge which the richer nations in the West are unwilling to practice, which turns African academics to be mere research assistants or raw material reserves to well-funded researchers from the academies in the West, she challenges local researchers also to be more inclusive in their research areas in Ghana. One factor she claims as hampering research in Ghana also is the type of culturalism that Spivak speaks against, which hierarchizes research locations and communities while silencing others in Ghana. She gives examples of healing cultures among the Dagaaba and Ga for example that are never investigated by senior Ghanaian researchers as they perpetuate what Spivak labels as unexamined cultural nationalism. Lauer’s call for a rejection of Western notions of science especially in the study of African-based knowledge systems engaged in healing practices, is taken up later by Alexis B. Tengan in “Scientific Decolonization and Language Use in the Study of African Medicine, Religion, and Art.” Tengan, like Awoonor, Nketia, and Wiredu, laments the distorted views of everything African pertaining to religion, healing, etc. Though he spends a lot of energy tracing the history of the conversion of the Dagara to Christianity, he also maintains that this conversion has not eroded the worldview of the Dagara, and hence their strong pre-Christian cosmological worldview enables them, as Wiredu in his piece testifies about the Akan, to approbate certain Western Judeo-Christian ideologies and enculturated

Legon Journal of the Humanities Vol. 30.2 (2019) Page 245 Kuwabong, D./Legon Journal of the Humanities Vol. 30.2 (2019) them into their society. Tengan’s focus, however, is not about the spirituality of the Dagara but about how their spirituality guides them in the direction of crafting strategies toward holistic healing practices. He details briefly how the Dagara house has sacred spots for different divinities, and where religious artifacts are kept. Wiredu’s piece, “The Humanities and the Idea of National Identity,” takes Akan spirituality studies as an approach to craft, defines and defends what he believes is the Akan philosophy of human identity and existence. His position is analogous to Tengan’s as they both reveal and debate the complicated nature of understanding the African human person through Euro-Judeo-Christianity and scientifics alone, while abrogating and negating African concepts of their universe, human relations, morality, God, and how these differentiate and influence their politics and cultures. Wiredu’s and Tengan’s positions, in conjunction with Nii-Yartey’s “Contemporary Ghanaian Dance: A Basis for Scholarly Investigation of the Human Condition,” and Collins’ “Ghanaian Neo-traditional Performance and ‘Development’: Multiple Interfaces between Rural and Urban, Traditional and Modern,” cover the methodological paradigms suggested by Nketia in “The Formulation of Research in the Humanities: Perspectives from the Creative Arts,” as they approach their studies through anthropology, philosophy, healing science, and cultural performances. These others present their studies as cognitive representations of what is available, and can be recuperated, though not as pristine cultural artifacts, but as evolving, developing, and changing modes of behavior while remaining true to their cultural worlds. Nketia’s contribution showcases individuals whose roles in appropriating, extending, and contemporizing their Ghanaian music traditions to make them relevant to modern Ghana, while inserting them in global cultural poetics. Indubitably, Nketia exhorts contemporary Ghanaian humanities researchers to engage critical and cognitive experiential research methodologies that are transdisciplinary, to make their work relevant, as against the current system of seeing culture as abstract objects for research. Collins’ essay proves that even if academics continue to exacerbate the problem of what Lauer labels epistemic injustice which stifles the multi-directional exchange of ideas, non-academic Ghanaian communities seem to be living these ideologies of shared experiences which humanities scholars need to recognize and tap into, rather than the present system in which researchers cocoon themselves in their disciplines, create hegemonies of knowledge, and paralyze new voices that seem to transgress these borders of knowledge. Interestingly, in “Kofi Awoonor: The Essays of a Humanist,” Olúfẹ́mi Táíwò reveals how African humanities scholars have often failed to critically engage the work of thinkers such as Awoonor, even if such an enterprise is necessary.

Page 246 Legon Journal of the Humanities Vol. 30.2 (2019) Kuwabong, D./ Philosophical Foundations of the African Humanities through Postcolonial Perspectives His position is that Awoonor as an essayist, philosopher, literary critic, cultural activist, poet and novelist has demonstrated his plural-disciplinarity of ideas as a Ghanaian/African(ist) humanist which this volume advocates and showcases. Táíwò then proceeds to give the reader a sampling of the extraordinary contribution of Awoonor to the humanities in Africa. He details some of the great thinkers of the twentieth century starting with Kwame Nkrumah through José Marti, Léopold Sédar Senghor, W. E. B DuBois, George Padmore, and other African descendants in diaspora, for whose differing ideologies and positions Awoonor was able to find linkages and compromises toward crafting consensual positions for the liberation of African identities locked up in the control rooms of Western epistemologies and politics of separation. Táíwò elevates Awoonor’s philosophy, which covers other writers whose literary works often hold similar views about their communities as his own, such as Allen Ginsberg, Norman Mailer and Nikki Giovanni. In all his work, Táíwò argues that Awoonor was in search of a rhetoric of peace, selfhood, and cohabitation of all peoples, not revenge poetics. Abena Oduro’s “Sell, Borrow, Work or Migrate? Exploring the Choice of Coping Strategies in Ghana,” explores qualitative research methodology to explore types of coping strategies employed by Ghanaians in times of financial difficulties where there is little or no state intervention to bail out the individual. Oduro’s research is important, first, for its coverage area of six districts in six regions across Ghana, and second, for its arrival at the qualitative evaluation of data through quantitative selection process of respondents. The strategy of leaving open the potential of replacing target respondents in each location created a flexibility in methodology that often seemed difficult for researchers who resist modifications to their research methods. Reviewing the types of shocks that generated individual coping mechanisms, Oduro’s essay reveals the spirit and culture of resilience and independence exhibited by the respondents who represent a cross-section of Ghanaians. The evidence from the research contests economic anthropologists who often present the ordinary Ghanaian as lacking in industry and self-help projects, the role of the extended family and sociocultural groups, and visiting conjugal relationships, in the life of the individual, as hinted by Wiredu and Tengan. Nonetheless, Oduro’s speculation that the absence of a stronger governmental spousal and child support system, the limited educational qualification of most of the respondents except one, and the threats of natural and ecological disasters, all tend to weaken these coping mechanisms. Adika in “Credibility and Accountability in Academic Discourse: Increasing the Awareness of Ghanaian Graduate Students,” tackles an issue close to my heart as teacher of graduate students at my university. The problem, he laments, is

Legon Journal of the Humanities Vol. 30.2 (2019) Page 247 Kuwabong, D./Legon Journal of the Humanities Vol. 30.2 (2019) equally shared by professors in the humanities at the University of Puerto Rico. His critique of structural and pedagogical weaknesses in the teaching of academic writing to graduate students is well argued. As a professor at the Language Center of the University of Ghana, his frustration with the inability of graduate students to effectively communicate their research data in their essays, theses, and dissertations, indicates an area in academia that calls for redress. Adika takes issue with multiple factors that exacerbate the problem: students’ lackadaisical attitude to academic writing, the dictatorial attitudes of professors who do not encourage what Spivak in her piece calls collaborative learning, overcrowded classrooms that undermine one on one tutoring, thus creating a situation in which professors tend to focus more on grammatical correctness and narrative structures, rather than on the internal dynamics that make academic writing a discipline of its own. Implicitly, Adika considers that universities in Ghana need to develop better tools to tackle the problem of academic writing among graduate students to make them globally competitive. Dako’s “About the English Language in Ghana Today and about Ghanaian English and Languaging in Ghana,” takes up the challenge of teaching English in Ghana offered by postcolonial debates on colonial language usages in former colonies such as Ghana. Her study is not so much a lamentation as it is an argument for recognition of English in Ghana as different and Ghanaian. Thus she identifies Ghanaian English (GhaE) not as a foreign language anymore, but as a second language with its own characteristics, something she showcases in her 2004 book, Ghanaianism: A Glossary. She discusses several variables from code-switching, pronunciation, accent, pidginization, and more recently locally acquired foreign accents that have added to the repertoire of accents in the Ghanaian use of English. Her essay is one of the most technical in the field and challenges those who insist on a purist approach to English language teaching, learning and usage in Ghana. But if Dako’s position reveals the complications of English in Ghana, Dakubu’s “Polylectal Description: Reflections on Experience in Ghana,” extends the language debate to cover four Ghanaian languages. Reflecting on her own struggles with facilitating the learning of four related but distinct languages such as Ga, Dangme, Dagaare and Farefari, and having to deal with issues of monodialectal, multidialectal and idiolectal translation and transcription differences and the politicization of language policies in Ghana. Her essay is a critical piece that puts in center stage, the problem of language policy in Ghanaian schools, in which dialects are selected for political reasons which resonate with cultural and numerical hegemony. Though she does not proffer any mono solution, her arguments bend toward crafting language policies such that even as they try to standardize Ghanaian languages in writing,

Page 248 Legon Journal of the Humanities Vol. 30.2 (2019) Kuwabong, D./ Philosophical Foundations of the African Humanities through Postcolonial Perspectives must also be democratic enough to allow for multidialectal pedagogies. Though not a linguist, I as a Dagao understand her despair as a linguist dealing with Dagaare, a language that I often refer to as the creole language of modern Ghana. Thus, Dakubu’s prescription for a multidialectal mechanism toward attempting some standardization of these languages challenges scholars in the humanities in Africa not to try to design a monolingual approach to the study of the humanities, to avoid hierarchization of disciplines, and to try to establish negotiating terrains that are conscious of other voices and positions that, as Wiredu puts it, are cognitive, normative, and decisional consensuses. The appendix contains personal reflections of praise by James Gibbs for Francis Nii-Yartey, Bernth Lindfors for Kofi Awoonor, and Helen Lauer for Mary Esther Kropp Dakubu, who all passed away before the book could be published. I wish to dedicate my review to J. K. Nketia who also passed away while the manuscript was at the press, and hence a dedication review could not be written for him to showcase his great and immeasurable contribution to the study of African music. Finally, I would like to commend the essayists and editors for a job well done. The essays reflect what Nii-Yartey in a subsection of his piece calls a dance of unity in diversity toward creating humanities programs as sites of identity, behavioral formation, interchange of ideas, democratic politics and emergent cultural, academic, and intellectual activism, the dismantling of uneducated culturalism and cultural nationalism, etc. Indeed, the essays show that in the humanities, it is possible to engage in democratic learning in which there is epistemic justice and equity in knowledge interchange. Indeed, Philosophical Foundations of the African Humanities through Postcolonial Perspectives should be required reading for graduate students in all departments of African Studies, and in libraries focused on research in African studies. The essays give hope that all is not lost and that there is a future for the humanities in African academies. The transnational composition of the scholars also testifies to how humanistic scholarship should be done. The collection reminds me of a weeklong workshop I attended in Quarzazate, Morocco in 2018 organized by linguists from the University of Cologne, Germany titled “Hospitable Linguistics.” Philosophical Foundations of the African Humanities through Postcolonial Perspectives is a representation of hospitable humanities. A must read.

Legon Journal of the Humanities Vol. 30.2 (2019) Page 249