Cite this article as: Bawale S & Abdullahi S. (2026). Harnessing Artificial Intelligence (AI) in African Historiography: Augmenting Historical Sources. Zamfara International Journal of Humanities, 4(2), 207-214. www.doi.org/10.36349/zamijoh.2026.v04i02.019
By
Shamsuddeen Bawale1
Sulaiman Abdullahi2
1&2 Arewa House Centre for Historical
Documentation and Research, Ahmadu Bello University, Zaria
Abstract: Artificial Intelligence (AI) has emerged as one of the
most transformative technologies of the twenty-first century, rapidly reshaping
academic fields and research methodologies. In African historiography, where
oral traditions, written document, archaeological materials and historical
linguistic form the core foundations of historical sources. Artificial
intelligence offers unprecedented opportunities for enrichment, preservation
and reinterpretation of human past. African history has long been shaped by
fragmented archives, limited documentation, colonial distortions, and the
vulnerabilities of orally transmitted memory. This study adopts a mixed-methods
approach that combines qualitative archival and oral-historical methods with
computational techniques, reflecting the interdisciplinary nature of
integrating AI into African historiography, draws on secondary sources to
analyze the AI factor in historical research and documentation. AI presents
complementary tools that strengthen these sources: speech recognition models
help preserve and transcribe oral narratives; machine learning corrects and
deciphers ancient manuscripts and Ajami texts; satellite-based AI enhances
archaeological discovery; and computational linguistics reveals hidden historical
relationships embedded in African languages. This paper examines how AI can
serve as a complement rather than a replacement to traditional historical
methods. Through theoretical engagement, methodological analysis, and empirical
examples from across Africa, it demonstrates how AI expands the historical
sources, safeguards endangered knowledge systems, and empowers African scholars
and communities to reclaim historical narratives. The paper acknowledges
limitations and ethical concerns, emphasizing the need for community
participation in historical reconstruction, decolonized digital practices and
human interpretative authority.
Keywords: Artificial Intelligence (AI), Historiography, History,
Sources
Introduction
Theoretical Framework
The theoretical framework synthesizes insights
from historical source theory, Actor-network theory, and Postcolonial archival
critique to explain the role of AI in African historiography. From the
viewpoint of historical source theory, oral traditions, written documents,
archaeological artifacts and historical linguistic embodied unique epistemic
claims. AI is understood as a complementary analytical layer that enhances interpretive
precision by identifying patterns invisible to manual inspection, translating
texts across time and script, and enabling cross-referencing across large set
data.[1] Actor-network
theory enables the cyclical understanding of historical facts, AI systems,
elders, archives, digital tools, and physical sites form a distributed network
of knowledge production, within this network, agency is shared, and the
reliability of a historical argument emerges from collaboration between human
judgment and technological capability.[2] Postcolonial
archival theory further anchors the framework by emphasizing that archives in
Africa have been shaped by unequal power relations, selective documentation and
colonial biases; hence, AI’s role must be ethically grounded in restoring access,
correcting distortions, and ensuring community participation in knowledge
governance.[3] Together,
these theoretical orientations position AI not as a hegemonic computational
authority but as a democratizing and restorative tool that strengthens the
resilience of African historical evidence, enhances transparency, and broadens
the historian’s capacity to engage deeply with the continent’s diverse memory
systems.
African Historical Research
In the 1950s African historiography championed
by renowned scholars such as Kenneth Nnweka DK, Samuel Ajayi Crowder, Abdullahi
Smith among others championed the decolonization of African historiography,
instituted the application of other sources of history such as, oral tradition,
archival records. Prior to revolution in global historical research, Africa has
always relied on a unique and diverse combination of sources that reflect the
continent’s complex social, political, and cultural evolution. These sources
are: oral tradition, written documents, archaeology, and historical linguistic
that form the backbone of African historiography and represent a methodology
deeply distinct from European or Asian historical reconstruction.[4] For
centuries, African societies preserved their histories primarily through spoken
narratives, memory systems, praise poetry, community chronicles, clan
genealogies, ritual performances, and artistic traditions. While written
documentation later became significant through Arabic scholarship, missionary
introduction of literacy, Islamic manuscripts, and colonial administrative
records, the continent’s historical memory remains predominantly rooted in oral
and material traditions.[5]
Archaeology and linguistics have also played
indispensable roles, particularly in reconstructing periods for which no
written documentation exists, such as early African state formation, migration
patterns, technological development, and interactions between cultural groups.
In the twenty-first century, Artificial Intelligence (AI) has emerged as an
increasingly influential tool capable of transforming virtually every aspect of
historical inquiry. While historians must approach AI critically recognizing
that it does not generate historical facts but analyzes existing evidence it
nonetheless offers methods that can dramatically expand the accessibility,
preservation, and interpretation of historical data.[6] In
Africa, where many sources are fragile, dispersed, endangered, or only
partially documented, AI’s potential is especially profound. The integration of
AI into African historiography represents not a break from tradition but a
significant enhancement of the tools available to historians seeking to
understand Africa’s past with greater accuracy and depth.
Methodology
This study adopts a mixed-methods approach that
combines qualitative archival and oral-historical methods with computational
techniques, reflecting the interdisciplinary nature of integrating Artificial
Intelligence into historiography. Primary sources include recorded oral
interviews, transcriptions of Ajami and Arabic manuscripts, colonial
administrative records, missionary correspondence, archaeological site reports,
and corpora of indigenous languages. These were selected purposively to
represent a range of geographical zones, languages, and source types across
West, Central, and East Africa. Fieldwork methods involve semi-structured
interviews with custodians of oral tradition, elders, traditional chroniclers,
and community historians whose testimonies were audio-recorded under ethical
consent protocols.
Harnessing Artificial Intelligence in African
Historiography
The use of AI within historical scholarship has
sometimes been misunderstood. One of the most common conceptual errors is the
assumption that AI, by generating text, becomes a primary source in itself.
This misunderstanding must be corrected because, AI is not a producer of
original historical events; rather, it functions as an instrument that
organizes, analyzes, and correlates human-generated historical materials.
Historical facts still originate from human memory, material artifacts,
archival documents, linguistic structures, ecological contexts, and the lived
experiences of people. AI enters this intellectual space as an analytical
engine capable of processing vast amounts of information far more quickly than
human researchers could manually achieve. This distinction is vital because,
African historiography has always struggled against stereotypes suggesting that
Africa “lacks history” due to limited written records. AI helps dispel this
notion by demonstrating that Africa’s historical evidence is not lacking but
distributed across many forms that require sophisticated tools to interpret
comprehensively.[7]
Artificial Intelligence and Oral Tradition
The historical significance of oral tradition in
Africa cannot be overstated. For centuries, scholars, elders, praise-singers,
and custodians of community memory transmitted knowledge of royal lineages,
migrations, dynastic disputes, religious transformations, and significant
social events. These narratives provided coherence for societies that valued
oral tradition as a source of legitimacy and identity. Yet oral tradition has
often been criticized by earlier European historians as unreliable due to its
fluidity, performativity nature, and potential for embellishment. In reality,
oral tradition is highly structured and reflects complex mnemonic techniques
and narrative frameworks maintained across generations.[8]
AI enhances the reliability and preservation of
oral tradition in several important ways. First, speech-to-text technologies
allow researchers to record large numbers of interviews and convert them into
written form without losing the subtleties of tone, cadence, or vocabulary.
This is particularly important because many custodians of oral memory are
elderly, and the rapid disappearance of indigenous languages threatens the
survival of centuries-old historical knowledge. AI also assists by detecting
thematic patterns within multiple oral narratives, identifying points of
convergence and divergence, and highlighting inconsistencies that may require
further investigation. This analytical capacity allows historians to examine
oral tradition with a level of systematic rigor that was previously
unattainable.[9]
Furthermore, AI-driven translation models are
increasingly capable of interpreting African languages such as Hausa, Yoruba,
Igbo, Fulfulde, Kiswahili, and others, thereby making oral histories accessible
to global audiences without losing their cultural and linguistic integrity.
Through these methods, AI does not replace oral tradition but preserves and
strengthens it, ensuring that the living memory of African societies becomes
part of the permanent historical record.
Artificial Intelligence and Written Document
Written documents have long been essential for
constructing African history, although their availability varies widely across
regions and periods. Arabic manuscripts from the Sahelian emirates, missionary
records from the nineteenth century, colonial administrative reports,
indigenous chronicles, legal documents, early newspapers, Ajami manuscripts,
and personal letters represent an invaluable body of evidence. However, many of
these documents remain inaccessible due to physical deterioration, the difficulty
of interpreting old scripts, restricted archival access, or the sheer volume of
material scattered across global repositories. AI addresses these obstacles
primarily through optical character recognition (OCR) technologies, which
convert handwritten or printed texts into digital formats. Particularly
important is the development of OCR capable of recognizing non-Latin scripts
such as Arabic, Ajami, and early missionary orthographies, which historically
posed challenges for digitization.[10]
Digitized archives enable historians to search
for keywords, names, locations, and themes across thousands of pages in
seconds, a task that would have previously required months of manual archival
work. AI also assists in classifying documents by date, subject matter, or
geographical region, making it possible to link disparate historical events or
actors that appear across different colonial archives. Moreover, AI’s capacity
for text mining allows researchers to identify patterns or themes that might otherwise
be overlooked, such as repetitive references to certain diseases,
administrative conflicts, trade routes, or Christian missionary medical
practices. For African historiography, these techniques provide new insights
into political, economic, and social transformations that occurred during
periods of rapid change, including colonial conquest, resistance movements,
Christian missionary expansion, and the reforms introduced by Islamic
scholarship. In this way, AI enhances written documentation by making it more
accessible, interpretable, and interconnected.[11]
Artificial Intelligence and Archeology
Archaeology is another domain where AI has
dramatically expanded the scope of African historical reconstruction. The
continent is home to thousands of archaeological sites, including ancient
cities, iron-working centers, trade settlements, burial grounds, agricultural
systems, and ritual landscapes. Yet many of these sites remain undiscovered or
insufficiently studied due to limited funding, vast geographic distances, and
dense vegetation in forested regions. AI-driven satellite analysis and remote
sensing technologies help archaeologists identify potential sites by analyzing
soil texture, vegetation patterns, topography, and anomalies in the landscape
that might indicate buried structures. This allows archaeologists to locate
ancient settlements without physically excavating every area.[12]
AI also supports archaeological interpretation
by classifying artifacts such as pottery, beads, bones, and tools,
distinguishing stylistic or chronological variations that may indicate cultural
interactions or migration patterns. In addition, 3D reconstruction technologies
can digitally rebuild damaged artifacts or reassemble fragmented items that
would otherwise remain incomplete. This capacity is invaluable for
understanding the symbolic and artistic dimensions of African material culture.
AI further enhances chronological reconstruction by analyzing radiocarbon
dating results and stratigraphic information in ways that reduce error margins,
thereby improving the precision of archaeological timelines. More efficient
site identification, artifact classification, and digital reconstruction
collectively broaden the archaeologist’s ability to interpret Africa’s past
with greater clarity and accuracy.[13]
Artificial Intelligence and Historical Linguistics
Historical Linguistics has always been a
cornerstone of African historiography, especially for ancient periods that lack
written documentation. African languages contain embedded historical evidence
about migration, cultural contact, trade, environmental knowledge, and
political organization. Comparative historical linguistics helps researchers
trace language families, reconstruct proto-languages, and understand how groups
spread across the continent. AI amplifies these linguistic tools by analyzing
large datasets of lexical items, phonological structures, and syntactic
patterns from hundreds of African languages, including those with limited
documentation. This makes it possible to identify previously unnoticed
relationships between languages or dialects.[14]
AI also contributes significantly to deciphering
old scripts such as Ajami manuscripts written in modified Arabic characters to
represent African languages. These manuscripts are among Africa’s richest
written traditions, but many remain unread due to the difficulty of
interpreting their orthography. AI models trained on Ajami texts can assist
researchers in transcribing, translating, and interpreting these materials
accurately. Additionally, semantic analysis helps track how the meanings of
important political, economic, and religious terms have changed over time,
revealing cultural transformations within African societies. Such linguistic
insights provide deeper understanding of state formation, religious change,
social hierarchy, and interactions between communities.[15]
Impact of Artificial Intelligence on African Historiography
The interaction between AI and African
historiography therefore reveals a broader historical truth: Africa has always
had history, but challenges in transmission, preservation, and accessibility
limited its visibility within global scholarship. AI provides tools that help
historians overcome these limitations by preserving endangered knowledge,
expanding the analytical scope of existing sources, and connecting disparate
forms of evidence into cohesive historical narratives. Through these methods,
AI enhances Africa’s capacity to control its own narratives and contribute more
strongly to global historical knowledge.[16]
Nevertheless, AI in African historiography must
be approached with caution and critical awareness. The primary responsibility
of the historian remains interpretation, verification, and contextualization.
AI may assist with pattern recognition and data processing, but it cannot
replace the historian’s judgment, cultural understanding, or methodological
rigor. The limitations of AI include occasional errors in translation,
misinterpretation of context, biases embedded within training data, and the
possibility of over-reliance on machine-generated insights without sufficient
human verification. For African historiography to benefit fully from AI,
historians must develop frameworks that integrate AI tools responsibly,
ensuring that technology enhances rather than distorts the historical record.[17]
The impact of AI’s contributions to oral,
written, archaeological, and linguistic sources demonstrates that technology
serves as a powerful complement rather than a replacement for traditional
historical methods. AI does not determine historical truth; it facilitates
access, preserves endangered sources, organizes large datasets, and enhances
analytical rigor. Historians must still interpret evidence, verify results, and
apply contextual knowledge, ensuring that AI serves within a framework of
scholarly responsibility.
Conclusion
Artificial Intelligence represents a profound
methodological advancement in African historiography, offering tools that
enhance the preservation, accessibility, and interpretation of the continent’s
diverse historical sources. By strengthening oral tradition through
transcription and translation, making written documents more accessible through
digitization and text analysis, advancing archaeological discovery through
remote sensing and artifact classification, and enriching linguistic research
through computational modeling, AI expands the evidential base of African
historical research. Nevertheless, AI does not replace the historian, it
functions as a complement that assists in revealing deeper insights and
preserving fragile knowledge. As Africa continues to engage with digital
technologies, AI holds the promise of safeguarding cultural memory,
democratizing access to historical evidence, and enabling a richer, more
accurate reconstruction of the continent’s past. The integration of AI into
African historiography therefore marks a significant step forward in ensuring
that Africa’s history is not only preserved but interpreted with the precision,
dignity, and depth it deserve.
References
1. Aarup, M., Arentoft, M. M., Parrod, Y., Stader,
J., and Stokes, I. (1994). Optimum-AIV: A knowledge-based planning and
scheduling system for spacecraft AIV. In Fox, M. and Zweben, M. (Eds.), Knowledge
Based Scheduling. Morgan Kaufmann.
2. Abney, S.
(2007). Semi supervised Learning for Computational Linguistics. CRC
Press.
3. Abramson, B. and Yung, M. (1989). Divide and
Conquer Under Global Constraints: A Solution to the N-queens Problem. J.
Parallel and Distributed Computing, 6(3), 649–662.
4. Achlioptas, D. (2009). Random Satisfiability. In
Biere, A., Heule, M., van Maaren, H., and Walsh, T. (Eds.), Handbook of
Satisfiability. IOS Press.
5. Achlioptas, D., Beame, P., and Molloy, M.
(2004). Exponential Bounds for DPLL Below the Satisfiability threshold.
In SODA-04.
6. Binmore,
K. (1982). Essays on Foundations of Game Theory. Pitman.
7. Bishop,
C. M. (1995). Neural Networks for Pattern Recognition. Oxford
University Press.
8. Bishop,
C. M. (2007). Pattern Recognition and Machine Learning.
Springer-Verlag.
9. Bisson,
T. (1990). They’re Made Out of Meat. Omni Magazine.
10. Bistarelli, S., Montanari, U., and Rossi, F.
(1997). Semiring-Based Constraint Satisfaction and Optimization. JACM, 44(2),
201–236.
11. Bitner, J. R. and Reingold, E. M. (1975).
Backtrack Programming Techniques. CACM, 18(11),
651–656.
12. Bizer, C., Auer, S., Kobilarov, G., Lehmann, J.,
and Cyganiak, R. (2007). DBPedia – querying Wikipedia like a database. In Developers
Track Presentation at the 16th International Conference on World
Wide Web.
13. Blazewicz, J., Ecker, K., Pesch, E., Schmidt,
G., and Weglarz, J. (2007). Handbook on Scheduling: Models
and Methods for Advanced Planning (International Handbooks on
Information Systems). Springer-Verlag New York, Inc.
14. Blei, D. M., Ng, A. Y., and Jordan, M. I.
(2001). Latent Dirichlet Allocation. In Neural Information Processing
Systems, Vol. 14.
15. Blinder, A. S. (1983). Issues in the
Coordination of Monetary and Fiscal Policies. In Monetary Policy Issues
in the 1980s. Federal Reserve Bank, Kansas City, Missouri.
16. Bliss, C. I. (1934). The Method of
Prohibits. Science, 79(2037), 38–39.
17. Block, H. D., Knight, B., and Rosenblatt, F.
(1962). Analysis of a Four-Layer Series-Coupled perceptron. Rev. Modern
Physics, 34(1), 275–282.
18. Blum, A. L. (1996). On-line Algorithms in
Machine Learning. In Proc.Workshop on On-Line Algorithms, Dagstuhl,
pp. 306–325.
19. Blum, A. L. and Furst, M. (1995). Fast Planning
Through Planning Graph Analysis. In IJCAI-95, pp.1636–1642.
20. Blum, A. L. and Furst, M. (1997). Fast Planning
Through Planning Graph Analysis. AIJ, 90(1–2), 281–300
21. Blum, A. L. and
22. Mitchell, T. M. (1998). Combining Labeled and
Unlabeled Data with Co-Training. In COLT-98, pp. 92–100.
23. Blumer, A., Ehrenfeucht, A., Haussler, D.,
andWarmuth, M. (1989). Learnability and the Vapnik- Chervonenkis
dimension. JACM, 36(4), 929–965.
24. Bobrow, D. G. (1967). Natural Language Input for
a Computer Problem Solving System. In Minsky, M. L. (Ed.), Semantic
Information Processing, pp. 133– 215. MIT Press.
25. Bobrow, D. G.,Kaplan, R.,Kay,M.,Norman, D. A.,
Thompson, H., and Winograd, T. (1977). GUS, a frame driven dialog system. AIJ, 8,
155–173.
26. Boden, M.
A. (1977). Artificial Intelligence and Natural Man. Basic Books.
27. Grow with Google (https://grow.google/intl/uk/courses-and-tools)
IBM AI Training (https://www.ibm.com/training/ailearning).
[1] T. Falola, African Historiography:
Essay in Honour of Jacob Ade Ajayi. Longman Group United Kingdom
Limited. England. 1993. Pp.156-157, 182
[2] Bruno Latour, “On Actor-network Theory: A
Few Clarification plus More Than a Few Complication”, in Special Issue of the
Nanish Philosophy Journal. Vol. 25, No. 3. P.47-64. 1990
[3] Marlene Manoff, “Theories of the Archive
from Across the Discipline”. Baltimore, Hopkins University Press. Vol. 4, No.
1. P.9-25. 2004
[4] Falola, African Historiography:
Essay in Honor of Jacob Ade Ajayi, Pp. 203-213, See also M.
Mamdani, Citizen and Subject: Contemporary Africa and the Legacy of
Late Colonialism, (Princeton, 1996). Carr, E. H. What is
History? (Great Britain, Penguin Books Ltd, 1961), P. 8, 18 90, 80,
87, 100,109, see
[5] Richard J. Evans, Prologue: What is History? –
Now in (edt) David
Cannadine, What is History Now?( London , Palgrave Macmillan Ltd
2002), P.25
[6] M. Briam, (1973) Archeology: A Brief
Introduction, Boston. Brown and Company. P.26 See also A. Marwick, The
Nature of History. P.141
[7] Falola, African Historiography:
Essay in Honor of Jacob Ade Ajayi, Pp. 203-213, See also M.
Mamdani, Citizen and Subject: Contemporary Africa and the Legacy of
Late Colonialism, (Princeton, 1996).
[8] A. L. Ker, “The Role of Oral Tradition in
the Reconstruction of Nigerian History” in M. O. Odey (edt), History
Research and Methodology in Africa, ( Makurdi, Aboki Publishers, 2007),
P.27
[9] Julia Sampson, Introduction to
Artificial Intelligence (AI) Technology: Guide for Travel and Tourism Leaders.
London, Longman. Pp.4-5. 2024 see also A. L. Ker, “The Role of Oral Tradition
in the Reconstruction of Nigerian History” in M. O. Odey (edt), History
Research and Methodology in Africa, ( Makurdi, Aboki Publishers, 2007),
P.27
[10] Gwarzo, Muhammad Sani. 1999. Hausa Ajami
Manuscripts and Their Role in the Spread of Islamic Culture in Northern
Nigeria. Kano: BUK Press. Carr, What is History? P.11
[11] Stuart Russel, and Peter Nuvigo, Artificial
Intelligence: A Modern Approach. (Third Edition). New Jessy, Pearson
Educational Books INC. P.1-17
[12] M. Briam, (1973) Archeology: A Brief
Introduction, Boston. Brown and Company. P.26 See also A. Marwick, The
Nature of History. P.141 see also T. Show, (1980). “Prehistory” in O.
Ikeme, Groundwork of Nigerian History. Ibadan. Heinemann
Educational Book. P.42
[13] Julia Sampson, Introduction to
Artificial Intelligence (AI) Technology: Guide for Travel and Tourism Leaders.
London, Longman. Pp.4-5. 2024
[14] J. H. (1963). Greenberg, The
Languages in Africa. Bloomington. Indiana University Press. Pp.43, 45 and
47 See also C. Ehret, (1975). Linguistics as a Tool for History,
Evanston. City Publishing Company. Pp.118-130 see also H. Greenberg,
(1963). The Languages in Africa. Bloomington. Indiana University
Press. Pp.43, 45
[15] Stuart Russel, and Peter Nuvigo, Artificial
Intelligence: A Modern Approach. (Third Edition). New Jessy, Pearson
Educational Books INC. P.35
[16] Stuart Russel, and Peter Nuvigo, Artificial
Intelligence: A Modern Approach. (Third Edition). New Jessy, Pearson
Educational Books INC. P.29.
[17] Richard J. Evans, Prologue: What is History? –
Now in (edt) David
Cannadine, What is History Now?( London , Palgrave Macmillan Ltd
2002), P.25
0 Comments