The Language of the Pretani

The Language of the Pretani: An Analysis of P-Celtic Dialects in Northern Britain (c. 700 BCE - 200 CE)

Section 1: The Celtic Linguistic Landscape of Pre-Roman Britain

1.1 From Proto-Celtic to Insular Celtic: Situating the Pretani Language

The language spoken by the Pretani tribes of northern Britain did not emerge in isolation. It was the northernmost extension of a complex and evolving linguistic family that had spread across Europe over the preceding millennium. Understanding its origins requires tracing its lineage back to Proto-Celtic, the reconstructed common ancestor of all Celtic languages. This ancestral tongue is generally associated with the Urnfield and Hallstatt archaeological cultures of Central Europe and is thought to have been spoken between approximately 1300 and 800 BCE.¹ From this heartland, its dialects spread, eventually giving rise to two major geographical, and to some extent linguistic, groupings: the Continental Celtic languages (such as Gaulish, Celtiberian, and Lepontic) and the Insular Celtic languages of Britain and Ireland.²

The Insular Celtic languages, which include the direct ancestors of the Pretani language, form a distinct subgroup characterised by a series of shared linguistic innovations that are largely absent from their continental relatives. These features, which must have developed during a period of common evolution after the languages arrived in the British Isles, point to a prolonged separation from the continent. Among the most significant of these shared traits are the development of inflected prepositions (where a preposition and a pronoun merge into a single word), a rigid Verb-Subject-Object (VSO) sentence structure, and a complex system of initial consonant mutations known as lenition.⁶

However, the precise relationship between the branches of the Celtic family tree remains a subject of intense scholarly debate, a debate that directly impacts the classification of the Pretani language. Two primary models exist. The "Insular Celtic" hypothesis posits that the languages of Britain (Brythonic) and Ireland (Goidelic) form a single, coherent branch, defined by the shared grammatical innovations mentioned above. This model suggests a long period of shared development within the Isles, separate from the continent.⁷ In contrast, the "Gallo-Brittonic" or "P-Celtic" hypothesis groups the Brythonic languages of Britain together with Gaulish, based on a crucial shared phonological innovation: the shift of the Proto-Celtic sound

kʷ to p. This model suggests a closer linguistic bond across the English Channel than across the Irish Sea.⁹

This academic uncertainty presents a fundamental methodological challenge. The defining features of the Insular Celtic hypothesis are primarily grammatical and syntactic. As the Pretani left no written records from this period, it is impossible to directly observe their sentence structure or the nature of their prepositions. Conversely, the defining feature of the Gallo-Brittonic hypothesis is phonological, a sound change that is readily apparent in the tribal and place names recorded by classical observers. Consequently, while the language of the Pretani belongs geographically to the Insular sphere, its most visible and verifiable linguistic markers align it with the P-Celtic/Gallo-Brittonic model. The language of northern Britain must therefore be understood as an Insular Celtic language that participated in the key P-Celtic phonological shift, a position that complicates any simple binary classification and highlights the layered nature of its linguistic history.

1.2 The P-Celtic/Q-Celtic Division: The Defining Phonological Isogloss

The most fundamental and widely accepted division within the Celtic language family is phonological, hinging on the treatment of the Proto-Indo-European labiovelar consonant kʷ. This sound, a 'k' pronounced with simultaneous lip-rounding, evolved in two distinct ways, creating a clear isogloss that separates the Celtic world into two great families: P-Celtic and Q-Celtic.¹⁰

In the Q-Celtic languages, which include Goidelic (the ancestor of Irish, Scottish Gaelic, and Manx) and the continental language Celtiberian, the original kʷ sound was preserved, eventually simplifying to a hard /k/ sound. This is why the Ogam script, developed for an early form of Irish, had a distinct letter for this sound, represented by 'Q'.¹³ In the P-Celtic languages, which include the Brythonic tongues (the ancestors of Welsh, Cornish, and Breton) and Gaulish, the

kʷ sound underwent a definitive shift to /p/.³

This divergence provides a powerful diagnostic tool. The word for "four," reconstructed in Proto-Celtic as *kʷetwores, became cethair in Old Irish (Q-Celtic) but pedwar in Welsh (P-Celtic). Similarly, the word for "son," *kʷenkʷe in Proto-Celtic, became cóic in Old Irish but pimp in Welsh and pempe in Gaulish.¹⁰ The presence of 'p' where 'c' or 'q' would be expected is the indelible fingerprint of a P-Celtic language. As will be demonstrated, the onomastic evidence from northern Britain places the language of the Pretani squarely within this P-Celtic group.

The precise chronology of this linguistic split is difficult to establish with certainty, and scholarly estimates vary widely. Some evidence from Continental Celtic languages like Lepontic and Celtiberian suggests the divergence was already underway before the 7th century BCE.¹³ Other models, based on glottochronology, have proposed dates as late as the mid-1st millennium AD, though these are now largely discounted by historical linguists.¹⁵ The current consensus, based on the likely dating of Proto-Celtic itself to the late Bronze Age, places the P-Celtic/Q-Celtic split somewhere in the early Iron Age, perhaps around 900 - 700 BCE, allowing sufficient time for the innovations to spread and solidify before the languages were first attested.¹

1.3 The Development of Common Brythonic (c. 400 BCE - c. 50 CE)

Within the P-Celtic family, the language spoken in Britain evolved into a distinct branch known as Common Brythonic (or British). This was the direct parent language of Welsh, Cornish, Breton (carried to the continent by British migrants), and Cumbric, and it represents the immediate linguistic ancestor of the dialect spoken by the Pretani.¹⁰ Between roughly 400 BCE and the mid-1st century CE, Common Brythonic underwent a series of shared phonological and morphological changes that distinguished it from its closest P-Celtic relative, Gaulish.

One of the most profound of these changes was the process of lenition, or consonant softening. In specific phonetic environments, particularly between vowels or between a vowel and a resonant consonant (like 'l' or 'r'), consonants began to be pronounced with less obstruction. Voiceless stops (p, t, k) became voiced (b, d, g), and voiced stops (b, d, g) became fricatives (v, ð, ɣ).¹⁷ While this process began in the Common Brythonic period, its effects became fully phonemicised later, forming the basis of the complex mutation systems of the modern Brythonic languages.

Alongside lenition, a series of vowel changes, known as affection (or umlaut), began to alter the sound of the language. Vowels began to change their quality in anticipation of the sound of a vowel in the following syllable. For example, under i-affection, a vowel would become fronted and raised if the next syllable contained an /i/ or /j/ sound. Conversely, under a-affection, high vowels like /i/ and /u/ were lowered to /e/ and /o/ respectively if the following syllable contained an /a/.¹⁷

Finally, this period saw the beginning of the most dramatic change to affect the language: the gradual weakening and eventual loss of final syllables, a process known as apocope. This had profound consequences, as the case endings of nouns and the personal endings of verbs, which were carried in these final syllables, were lost. This transformed Brythonic from a synthetic language, which expressed grammatical relationships through inflection (like Latin), into a more analytic language that relied on word order and prepositions, a structure that defines its modern descendants.²⁰ While the process of apocope was not completed until the 5th-6th centuries CE, its origins lie in the Common Brythonic period. These shared innovations created a unified linguistic entity across Britain, the direct baseline from which the northern dialect of the Pretani can be analysed.

Section 2: Pritenic: The Language of the Northern Tribes (c. 600 BCE - 200 CE)

The specific northern P-Celtic speech horizon associated with the Roman-era Pretani and related northern tribes is often discussed under the term Pritenic. The term was coined by Kenneth Jackson in 1955 to designate the hypothetical Roman-era ancestor of the later Pictish language, which he believed began to diverge from southern Brittonic around the 1st century CE. The evidence for this language is indirect, derived from the names of tribes, people, and places recorded by external classical observers; this onomastic evidence is fragmentary and often corrupted, but it remains the principal window into the phonology and lexicon of the northern tribes during the Iron Age. Pritenic is best treated as the earlier linguistic frame against which later Pictish evidence can be compared, not as proof that early medieval Pictish political identity already existed in the Pretani period.

2.1 Onomastic Evidence from Classical Sources: The Linguistic Fossil Record

The primary sources for Pritenic are the geographical and historical works of Greek and Roman authors, most notably the Geographia of Claudius Ptolemy (c. 150 CE) and the De vita et moribus Iulii Agricolae of Tacitus (c. 98 CE).³ These texts provide a list of tribal names that constitute our core linguistic data set. Among the peoples located in the territory of the Pretani were the

Caledones, Vacomagi, Taexali, Venicones, Epidii, Creones, Carnonacae, Caereni, Cornavii, Lugi, Smertae, and the Boresti.²²

Etymological analysis of these names, using reconstructed Proto-Celtic and Common Brythonic forms, confirms their P-Celtic character and offers insights into how these tribes perceived themselves or were perceived by their neighbours.

·         Epidii: Located in Argyll and Kintyre, their name is transparently derived from the P-Celtic root *epos ('horse'), cognate with Welsh ebol ('foal'). This name, meaning "the horse people," is a clear P-Celtic marker, showing the characteristic shift from Proto-Celtic *ekwos.¹¹ It may suggest a connection to a horse deity or a culture in which horsemanship was central.

·         Caledones: This powerful confederation, which gave its name to the entire region (Caledonia), likely derives its name from the Brythonic word *caled ('hard', 'harsh'), a reference to either the rugged, mountainous terrain they inhabited or their perceived hardy and resilient nature.²⁷

·         Vacomagi: Their name is likely a compound of *wāko- (a variant of wāgro-) meaning 'curved' or 'wandering', and *magos meaning 'field' or 'plain'. Thus, they were the "people of the curved fields," perhaps referring to the fertile plains of Strathmore where they were located.³¹

·         Venicones: Situated in modern-day Fife and Tayside, their name has been reconstructed as *weni-kones, meaning "kindred hounds" or "hunting hounds," from *weni- ('kindred', 'family') and the element *kunes ('hounds').³³ This name is of particular linguistic importance for the debate surrounding the evolution of the Proto-Celtic stem *

kuno- in the northern dialect.

This onomastic evidence, however, must be approached with significant critical caution. The names we possess are not pure Pritenic forms but are distorted echoes that have passed through a long and complex chain of transmission. A name spoken by a Pretani individual was likely first heard and recorded in Latin by a Roman soldier or official, then compiled in Greek by the geographer Marinus of Tyre, before being incorporated into Ptolemy's final Greek text. This text was then subject to centuries of scribal copying, introducing further errors and variations.²² The famous 90-degree eastward rotation of Scotland on Ptolemy's map is a stark visual reminder of the potential for geographical distortion; it is almost certain that similar linguistic distortions occurred.²² Scribal variants are common,

Taexali also appears as Taizali, and Nabaiou as Nabarou, demonstrating the instability of the forms that have reached us.²² Therefore, every etymology is a hypothesis built upon a potentially corrupted foundation. We are analysing a disturbed fossil bed, and this inherent uncertainty is a central reality of the field.

2.2 Phonological Characteristics of Pritenic: A Dialect in Divergence?

The scholarly debate over Pritenic has centred on whether it was merely a regional dialect of Common Brythonic or a distinct language that had begun to diverge at a much earlier stage. The arguments for an early and significant split have traditionally rested on a small number of supposed phonological peculiarities observed in the onomastic evidence. However, recent critical re-evaluation of this evidence, most notably by Guto Rhys, has cast considerable doubt on these claims, suggesting that most of these features can be explained more parsimoniously through other linguistic mechanisms.²²

·         The abor vs. aber Question: A key argument for Pritenic distinctiveness was the presence of the place-name element abor- (attesting an o-grade vowel) in northern names like Aporcrosan (Applecross), contrasting with the standard Brythonic aber- (e-grade vowel), meaning 'river mouth' or 'confluence'. This was interpreted as an archaism preserved in the north. However, analysis shows that the earliest attestations of these names come from Gaelic-speaking contexts. The shift from /e/ to /o/ after a labial consonant like /b/ is a known phonetic development within Gaelic. It is therefore more likely that the word was borrowed from Pritenic as aber and later evolved to abor within Gaelic, rather than being an original Pritenic feature.²²

·         *The Treatment of kuno- ('hound'): The appearance of *Con- instead of *Cun- in names such as Mailcon (a later Pictish king) and the tribal name Venicones was proposed as a unique Pritenic sound change where /u/ was lowered to /o/. However, the Proto-Celtic stem *kuno- exhibited ablaut, a system of internal vowel gradation, meaning that forms with both /u/ and /o/ existed in different grammatical cases across the Celtic world. The presence of *Con- can be explained as the generalisation of one of these original ablaut variants or as a simple scribal confusion between the similar-sounding vowels, rather than a unique phonological law of Pritenic.²²

·         Retention of Initial s-: It has been argued that Pritenic retained the Proto-Celtic initial s-, whereas in Common Brythonic it weakened to h-. This argument was based on names like Carn Smairt. However, closer analysis reveals that the etymologies supporting this claim are often weak or that the names in question are better interpreted as being of Gaelic origin, where initial s- was always preserved.²² There is no unambiguous evidence from the Pretani period showing a name that should have

h- but instead has s-.

·         Treatment of Consonant Clusters xs and xt: A potential link to Gaulish was proposed based on the theory that Pritenic, like Gaulish and Goidelic, simplified the consonant cluster xs to s. This is based almost entirely on the river name Loxa (equated with the modern Lossie) and the personal name Lossio Veda from the Colchester inscription. The evidence is highly problematic; the identification of Loxa is uncertain, and the name Lossio could have alternative origins. There is no broader evidence to support this as a general sound law in Pritenic.²²

The cumulative effect of this critical re-evaluation is to dismantle the case for Pritenic as a language that had significantly diverged from its southern counterpart during the pre-Roman and Roman Iron Age. The anomalies that once seemed to point to a separate linguistic identity are, upon closer inspection, better explained by later language contact, normal Celtic morphological variation, or the simple ambiguity of a severely limited dataset. The most robust conclusion is that the language of the Pretani was a northern regional variant of Common Brythonic, existing on a dialect continuum with the languages spoken further south, rather than a distinct, early-branching language.

2.3 The Question of a "Gaulish Connection"

The hypothesis that Pritenic may have been more closely related to Continental Gaulish than to southern Brittonic has been a recurring theme in linguistic debate. This idea was predicated on the identification of shared features, particularly archaisms, that appeared to have been lost in the more innovative Brythonic of the south. If substantiated, this would suggest a different migratory history or a more prolonged period of contact between northern Britain and the continent.

The primary evidence cited in favour of a special Gallo-Pritenic link has been phonological. As discussed, the proposed simplification of the consonant cluster xs to s in Pritenic would mirror the situation in Gaulish and Goidelic, contrasting with the development to ch (/x/) in Brittonic. However, as demonstrated, the evidence for this change in Pritenic is exceptionally thin and rests on the ambiguous etymology of the names Loxa and Lossio.²²

Another point of comparison has been the place-name element pett, meaning a 'piece' or 'portion' of land, which is extremely common in later Pictish place-names. It was argued that this term was a direct borrowing from a Gaulish word pettia, with the same specific meaning of a land unit, and that its absence in Welsh or Cornish place-names pointed to a unique Gallo-Pritenic lexical link. This argument, however, is flawed. The Gaulish word pettia simply meant 'piece' in a general sense, and its specific application to land was a later development in Gallo-Romance. Furthermore, the word is not attested in Gaulish place-names at all. The most economical explanation is that pett is a native Brythonic word, cognate with Welsh peth ('thing', 'piece'), which in the specific socio-economic context of Pictland came to acquire the specialised meaning of a land unit.²²

When the full range of onomastic evidence is considered, the overwhelming majority of Pretani tribal and place-names can be explained satisfactorily through a Common Brythonic etymological framework, without any need to invoke a special connection to Gaulish. While Gaulish remains an invaluable comparative resource for understanding the structure and vocabulary of a contemporary P-Celtic language, the evidence to support a unique Gallo-Pritenic linguistic branch is unsubstantiated. The language of the Pretani remains, on current evidence, fundamentally a dialect of British P-Celtic.

Section 3: Spoken and Written Language in Pretani Society

3.1 Orality, Memory, and the Absence of a Native Script

During the period from 600 BCE to 200 CE, the Pretani tribes of northern Britain inhabited a world of sophisticated orality. Despite their complex social structures, monumental architecture, and intricate artistic traditions, there is no evidence that they developed or adopted a system of writing for their language.³⁹ The earliest writing systems to appear in Britain were the runic alphabet of the Anglo-Saxons and the Latin alphabet introduced by the Romans, both of which post-date the period under consideration.⁴¹

The absence of a native script should not be interpreted as a cultural deficit. Rather, it reflects a society where knowledge, law, genealogy, and sacred lore were composed, preserved, and transmitted through highly structured oral traditions. Comparative evidence from later, literate Celtic societies in Ireland and Wales reveals the immense cultural prestige and social power vested in a learned class of specialists, druids, bards (bardoi), and seers (vates), who acted as living repositories of ancestral knowledge.⁴⁵ These individuals would have undergone years of rigorous training to memorise vast quantities of material in verse, a form that aids recall and ensures fidelity of transmission across generations.

The archaeological record of the Pretani corroborates this picture of a non-literate yet symbolically rich culture. The effort invested in monumental stone architecture such as brochs and duns, and the complex, abstract symbolism of their decorative metalwork, points to a society that communicated identity, status, and cosmological beliefs through powerful visual and spatial means rather than through text.²² In this context, the spoken word, structured in poetry and narrative, would have been the primary medium for recording history and affirming social order.

3.2 Comparative Literacy: Why Not the Pretani?

The lack of writing among the Pretani stands in contrast to the situation among some of their Continental Celtic contemporaries. In Cisalpine Gaul (northern Italy) and Gallia Narbonensis (southern France), Celtic-speaking peoples adopted and adapted scripts from their literate Mediterranean neighbours. From the 6th century BCE, Lepontic was written in a variant of the Etruscan alphabet, and from the 3rd century BCE, Gaulish was written using the Greek alphabet, primarily for religious dedications, funerary inscriptions, calendars (such as the Coligny Calendar), and potters' accounts.⁴ Later, under Roman influence, the Latin alphabet was widely adopted.

The divergence in the adoption of literacy can be explained by socio-economic and geographical factors. The Celts of southern and Cisalpine Gaul were in direct and sustained commercial and political contact with the highly literate civilisations of the Mediterranean, the Greeks, Etruscans, and Romans. This intensive interaction created a practical impetus for writing in contexts of trade, where records were needed; law, where contracts required formalisation; and civic administration, where public declarations were made.

The Pretani, by contrast, were situated on the far northwestern periphery of Europe. Their primary interactions were with other non-literate Iron Age communities within Britain and Ireland. They lacked the direct, sustained pressure from a neighboring literate state that would have created a functional necessity for writing. In the socio-political environment of Iron Age northern Britain, oral contracts, sworn oaths, and memorised genealogies were sufficient and culturally appropriate mechanisms for regulating society. Literacy, therefore, did not develop indigenously because the specific external pressures that catalysed its adoption on the continent were absent in the north.

Section 4: Comparative Analysis and Broader Context

4.1 Pritenic and Gaulish: A Comparative View

A comparative analysis of Pritenic and Gaulish provides a valuable perspective on the diversity within the P-Celtic language family during the Iron Age. As a contemporary and relatively well-attested P-Celtic language, Gaulish serves as a useful, though imperfect, analogue for understanding the potential structure and vocabulary of the unwritten language of the Pretani.⁴⁹

Both languages shared the defining P-Celtic phonological shift of kʷ > p and possessed a lexicon with a vast number of shared cognates derived from Proto-Celtic, such as *magos ('field, plain'), *dūnon ('fort'), *briga ('hill'), and *rix ('king').⁵⁰ However, the hypothesis of a special "Gallo-Brittonic" branch, which would posit a closer relationship between Gaulish and Brythonic (including Pritenic) than either had with Goidelic, remains contentious. While they share some innovations, the Insular Celtic languages (Brythonic and Goidelic) also share a suite of profound grammatical and syntactic features, such as VSO word order and initial consonant mutation, which are not attested in Gaulish and suggest a long period of shared, isolated development in the British Isles.⁵

The specific arguments for linking Pritenic more closely to Gaulish than to southern Brittonic have not withstood critical scrutiny. As previously discussed, proposed shared phonological innovations like the simplification of xs > s are based on highly ambiguous evidence.²² Furthermore, a detailed examination of the Pretani onomastic corpus reveals that the vast majority of names can be explained through a Common Brythonic etymological framework without needing to posit a special connection to Gaulish. The linguistic landscape is best understood not as a set of discrete, isolated branches, but as a dialect continuum. The languages of Britain and Gaul would have been part of a broad P-Celtic continuum, with mutual intelligibility likely decreasing with geographical distance. The language of the Pretani was the northernmost dialect of the British part of this continuum, distinct but not fundamentally separate from the dialects spoken to its south.

4.2 Linguistic Continuity and the Archaeological & Genetic Record

The linguistic evidence for an indigenous, evolving P-Celtic language in northern Britain finds powerful corroboration in the archaeological and genetic records. For decades, the dominant model for the arrival of Celtic languages in Britain was one of invasion and population replacement, with waves of "Celtic" peoples arriving from the continent during the Iron Age. Modern scientific evidence has rendered this model largely obsolete.

The archaeological record of northern Britain, particularly in the Atlantic regions of Scotland, demonstrates a profound cultural continuity from the late Bronze Age through the entire Iron Age. The defining architectural features of the Pretani period, such as the monumental Atlantic roundhouses and brochs, are now understood not as imported architectural forms but as a dynamic, indigenous development rooted in earlier local traditions.²² There is no archaeological horizon that indicates a large-scale migration or population replacement coinciding with the estimated arrival of the Celtic language.

Genomes of early medieval Pictish individuals found that they fall "firmly within the Iron Age gene pool in Britain." The analysis revealed a strong regional biological affinity and demonstrated "substantial genetic continuity in Orkney for the last ~2,000 years".²² This supports population continuity between Iron Age and early medieval northern Britain, and argues against a simple model of a new wave of continental migrants replacing the local population, but it should not collapse the Pretani and Picts into the same historical period.

The convergence of these three independent lines of evidence, linguistic, archaeological, and genetic, points toward a new paradigm. The spread of the P-Celtic language into northern Britain was not the result of a "Celtic invasion." Instead, it must be understood as a process of cultural diffusion and language shift that took place over many centuries, likely beginning in the late Bronze Age. The language probably spread northward from southern Britain, adopted by indigenous communities through mechanisms such as elite dominance, trade networks, and social prestige. The people of the north became Celtic-speakers, but they remained, by and large, the same people. This model of linguistic expansion without significant population replacement provides a coherent and evidence-based framework for understanding the origins of the Pretani and their language.

Section 5: Conclusion: Defining the Language of the Pretani

The language spoken by the Pretani tribes of northern Britain between approximately 600 BCE and 200 CE can be defined with a significant degree of confidence, despite the complete absence of contemporary written texts. A synthesis of the available linguistic, archaeological, and genetic evidence allows for a nuanced and robust conclusion that overturns older models of invasion and replacement.

First and foremost, the language of the Pretani was unequivocally a P-Celtic language. This is demonstrated by the onomastic evidence preserved in classical sources, which consistently shows the diagnostic phonological shift of the Proto-Celtic labiovelar kʷ to p. This feature firmly aligns the language with the Brythonic and Gaulish branches of the Celtic family and distinguishes it sharply from the Q-Celtic Goidelic languages of Ireland.

Second, within the P-Celtic family, the Pretani language is best classified as a northern dialect of Common Brythonic. While the scholarly term "Pritenic" is a useful and established label for this northern variant, it is misleading if it is taken to imply a separate language that had undergone a deep and early divergence from its southern counterpart. The specific phonological features once cited as evidence for such an early split have been shown to be weak, ambiguous, or more plausibly explained by later linguistic phenomena, such as the influence of Gaelic on borrowed place-names. The language of the Pretani existed on a dialect continuum with the Brythonic spoken in the rest of Britain.

Third, the Pretani inhabited a pre-literate society with a sophisticated oral culture. The absence of a native writing system was not a sign of cultural simplicity but a reflection of a society that prioritised oral tradition for the transmission of law, genealogy, and sacred knowledge. Their complex symbolic art and monumental architecture served as powerful, non-literate forms of communication.

Finally, the presence of this P-Celtic language in northern Britain is best understood as the result of long-term, indigenous processes of cultural diffusion and language shift, not mass migration. The archaeological and genetic evidence demonstrates a profound continuity of population from the Bronze Age through the Iron Age. The P-Celtic language likely spread northward gradually, adopted by local communities over centuries. The Pretani, therefore, were not Celtic invaders; they were the indigenous peoples of northern Britain who became Celtic-speakers, creating a unique cultural and linguistic identity on the northernmost frontier of the Celtic world.

Bibliography

Adams, J. N. Bilingualism and the Latin Language. Cambridge University Press, 2003.

Adams, J. N. The Regional Diversification of Latin, 200 BC-AD 600. Cambridge University Press, 2007.

Anderson, A. O. Early Sources of Scottish History: A.D. 500 to 1286. Oliver and Boyd, 1922.

Anderson, M. O. 'The Lists of the Kings'. Scottish Historical Review, vol. 28, no. 106, 1949, pp. 108 - 18.

Armit, Ian. Towers in the North: The Brochs of Scotland. B. T. Batsford, 1990.

Armit, Ian. The Archaeology of Skye and the Western Isles. Edinburgh University Press, 1996.

Armit, Ian. Celtic Scotland. B. T. Batsford/Historic Scotland, 1997.

Ballin Smith, B. editor. Howe: Four Millennia of Orkney Prehistory. Society of Antiquaries of Scotland, 1994.

Barclay, G. J. and O. A. Owen. 'Historic Scotland's Backlog Programme 1992-95: The Results'. Antiquity, vol. 69, no. 266, 1995, pp. 1003 - 8.

Barrow, G. W. S. 'The Uses of Place-Names and Scottish History, Pointers and Pitfalls'. In The Uses of Place-Names, edited by S. Taylor, Scottish Cultural Press, 1998, pp. 54 - 74.

Bartrum, P. C. A Welsh Classical Dictionary: People in History and Legend up to about A.D. 1000. National Library of Wales, 1993.

Breeze, Andrew. 'Philology on Tacitus's Grampian Hill and Trucculan Harbor'. Proceedings of the Society of Antiquaries of Scotland, vol. 132, 2002, pp. 305 - 11.

Broun, Dauvit. 'The Seven Kingdoms in De situ Albanie: A Record of Pictish Political Geography or Imaginary Map of Ancient Scotland?'. Innes Review, vol. 56, no. 1, 2005, pp. 1 - 45.

Calise, J. M. P. Pictish Sourcebook: Documents of Medieval Legend and Dark Age History. Greenwood Press, 2002.

Carver, M. O. H. 'An Iona of the East: The Early-Medieval Monastery at Portmahomack, Tarbat Ness'. Medieval Archaeology, vol. 48, 2004, pp. 1 - 30.

Carver, M. O. H. and C. Spall. 'Excavating a Parchmenerie: Archaeological Correlates of Making Parchment at the Pictish Monastery at Portmahomack, Easter Ross'. Proceedings of the Society of Antiquaries of Scotland, vol. 134, 2004, pp. 183 - 200.

Chadwick, H. M. Early Scotland: The Picts, the Scots & the Welsh of Southern Scotland. Cambridge University Press, 1949.

Charles-Edwards, T. M. Wales and the Britons, 350-1064. Oxford University Press, 2013.

Clancy, Thomas Owen. 'The Real St Ninian'. The Innes Review, vol. 52, no. 1, 2001, pp. 1 - 28.

Clancy, Thomas Owen. 'Philosopher-King: Nechtan mac Der-Ilei'. The Scottish Historical Review, vol. 83, no. 2, 2004, pp. 125 - 49.

Colgrave, B. and R. A. B. Mynors, editors. Bede's Ecclesiastical History of the English People. Clarendon Press, 1991.

Collingwood, R. G. and R. P. Wright. The Roman Inscriptions of Britain, I, Inscriptions on Stone. Oxford University Press, 1965.

Cunliffe, Barry. Iron Age Communities in Britain: An Account of England, Scotland and Wales from the Seventh Century BC until the Roman Conquest. 4th ed. Routledge, 2005.

Delamarre, Xavier. Dictionnaire de la langue gauloise: Une approche linguistique du vieux-celtique continental. 2nd ed. Éditions Errance, 2003.

Delamarre, Xavier. Noms de personnes celtiques dans l'épigraphie classique. Éditions Errance, 2007.

Diack, Francis C. 'Place-Names of Pictland'. Revue Celtique, vol. 38, 1920, pp. 109 - 32.

Evans, Nicholas. 'The King in the North: The Pictish Realms of Fortriu and Ce'. The Heroic Age, no. 15, 2012.

Feachem, Richard. A Guide to Prehistoric Scotland. B. T. Batsford, 1963.

Forsyth, Katherine. Language in Pictland: The Case Against 'Non-Indo-European' Pictish. De Keltische Draak, 1997.

Forsyth, Katherine. 'Literacy in Pictland'. In Literacy in Medieval Celtic Societies, edited by Huw Pryce, Cambridge University Press, 1998, pp. 39 - 61.

Forsyth, Katherine. 'Evidence of a Lost Pictish Source in the Historia Regum Anglorum of Symeon of Durham'. In Kings, Clerics and Chronicles in Scotland 500 - 1297, edited by Simon Taylor, Four Courts Press, 2000, pp. 19 - 34.

Forsyth, Katherine. 'Pictish Language and Documents'. In Celtic Culture: A Historical Encyclopedia, edited by John T. Koch, ABC-CLIO, 2006, pp. 1447 - 51.

Fraser, James E. From Caledonia to Pictland: Scotland to 795. Edinburgh University Press, 2009.

Graham-Campbell, James, and Anna Ritchie. The Picts: A History. Birlinn, 2000.

Green, Miranda. The Gods of the Celts. Alan Sutton, 1986.

Green, Miranda. Symbol and Image in Celtic Religious Art. Routledge, 1992.

Hallén, Y. 'The Use of Bone and Antler at Foshigarry and Bac Mhic Connain, Two Iron Age Sites on North Uist'. Proceedings of the Society of Antiquaries of Scotland, vol. 124, 1994, pp. 189 - 231.

Hamilton, J. R. C. Excavations at Jarlshof, Shetland. HMSO, 1956.

Hamilton, J. R. C. Excavations at Clickhimin, Shetland. HMSO, 1968.

Harding, Dennis W. The Iron Age in Northern Britain: Britons and Romans, Natives and Settlers. Routledge, 2004.

Haselgrove, C. et al. Understanding the British Iron Age: An Agenda for Action. Wessex Archaeology, 2001.

Hedges, J. W. Bu, Gurness and the Brochs of Orkney. British Archaeological Reports, 1987.

Hood, A. B. E. editor and translator. St. Patrick: His Writings and Muirchu's Life. Phillimore, 1978.

Hunter, Fraser. Beyond the Edge of the Empire: Caledonians, Picts and Romans. Groam House Museum, 2009.

Hunter, J. R. et al. 'Recent Investigations at Pool, Sanday, Orkney'. Proceedings of the Society of Antiquaries of Scotland, vol. 123, 1993.

Isaac, Graham R. 'The Antonine Itinerary Land Routes: Place-Names of Ancient Wales'. Archaeologia Cambrensis, vol. 154, 2005, pp. 181 - 4.

Jackson, Kenneth H. Language and History in Early Britain. Edinburgh University Press, 1953.

Jackson, Kenneth H. 'The Pictish Language'. In The Problem of the Picts, edited by F. T. Wainwright, Thomas Nelson and Sons, 1955, pp. 129 - 66.

Jackson, Kenneth H. The Gaelic Notes in the Book of Deer. Cambridge University Press, 1972.

James, Alan G. 'A Cumbric Diaspora?'. In Cumbric and Pictish, edited by K. Forsyth, et al. University of Glasgow, 2020.

Koch, John T. 'The Loss of Final Syllables and Loss of Declension in Brittonic'. Bulletin of the Board of Celtic Studies, vol. 30, 1983, pp. 201 - 33.

Koch, John T. Celtic Culture: A Historical Encyclopedia. ABC-CLIO, 2006.

Lambert, Pierre-Yves. La langue gauloise: description linguistique, commentaire d'inscriptions choisies. Éditions Errance, 2003.

MacGregor, Morna. Early Celtic Art in North Britain: A Study of Decorative Metalwork from the Third Century B.C. to the Third Century A.D.. Leicester University Press, 1976.

MacKillop, James. Dictionary of Celtic Mythology. Oxford University Press, 1998.

MacSween, A. 'Later Iron Age Pottery from the Western Isles'. In Neolithic and Later Prehistoric Pottery from the Islands of the Atlantic Fringe, edited by A. Gibson and A. Sheridan, Oxbow Books, forthcoming.

Matasović, Ranko. Etymological Dictionary of Proto-Celtic. Brill, 2009.

McCone, Kim. Towards a Relative Chronology of Ancient and Medieval Celtic Sound Change. Department of Old Irish, St. Patrick's College, 1996.

McManus, Damian. A Guide to Ogam. An Sagart, 1991.

McNeill, F. Marian. The Silver Bough: A Four Volume Study of the National and Local Festivals of Scotland. William MacLellan, 1957 - 1968.

Megaw, Ruth, and Vincent Megaw. Celtic Art: From Its Beginnings to the Book of Kells. Thames & Hudson, 2001.

Moffat, Alistair. Before Scotland: The Story of Scotland Before History. Thames & Hudson, 2005.

Moffat, Alistair. The Faded Map: The Lost Kingdoms of Scotland. Birlinn, 2010.

O'Rahilly, T. F. Early Irish History and Mythology. Dublin Institute for Advanced Studies, 1946.

Okasha, Elisabeth. 'The Non-Ogham Inscriptions of Pictland'. Cambridge Medieval Celtic Studies, vol. 9, 1985, pp. 43 - 69.

Padel, O. J. Cornish Place-Name Elements. English Place-Name Society, 1985.

Pedersen, Holger. Vergleichende Grammatik der keltischen Sprachen. Vandenhoeck & Ruprecht, 1909.

Rhys, Guto. 'Approaching the Pictish language: historiography, early evidence and the question of Pritenic'. PhD thesis, University of Glasgow, 2015.

Rivet, A. L. F. and Colin Smith. The Place-Names of Roman Britain. B. T. Batsford, 1979.

Schrijver, Peter. Studies in British Celtic Historical Phonology. Rodopi, 1995.

Sims-Williams, Patrick. 'Dating the Transition to Neo-Brittonic: Phonology and History, 400 - 600'. In Britain 400 - 600: Language and History, edited by A. Bammesberger and A. Wollmann, Carl Winter, 1990, pp. 217 - 61.

Sims-Williams, Patrick. The Celtic Inscriptions of Britain: Phonology and Chronology, c. 400 - 1200. Blackwell, 2003.

Taylor, Simon. 'Pictish Place-Names Revisited'. In Pictish Progress: New Studies on Northern Britain in the Early Middle Ages, edited by S. T. Driscoll, J. Geddes, and M. A. Hall, Brill, 2011, pp. 67 - 118.

Taylor, Simon, with Gilbert Márkus. The Place-Names of Fife. 5 vols. Shaun Tyas, 2006 - 2012.

Thomson, R. L. Pwyll Pendeuic Dyuet. Dublin Institute for Advanced Studies, 1957.

Traill, W. 'Results of Excavations at the Broch of Burrian, North Ronaldsay, Orkney, during the Summers of 1870 and 1871'. Archaeologia Scotica, vol. 5, 1890, pp. 341 - 64.

Watson, W. J. The History of the Celtic Place-Names of Scotland. William Blackwood & Sons, 1926.

Woolf, Alex. 'The "Moray Question" and the Kingship of Alba in the Tenth and Eleventh Centuries'. The Scottish Historical Review, vol. 79, no. 2, 2000, pp. 145 - 64.

Woolf, Alex. From Pictland to Alba, 789 - 1070. Edinburgh University Press, 2007.

Yorke, Barbara. The Conversion of Britain: Religion, Politics and Society in Britain c.600 - 800. Pearson Longman, 2006.