Latin texts in medieval Iceland: sources

Given the absence of surviving Latin manuscripts, reconstructing textual circulation in medieval Iceland relies on two sources: booklists and linguistic treatises. The máldagar (ecclesiastical inventories of episcopal possessions) are a crucial resource. Olmer (Reference Olmer1902) pioneered their study, revealing book collections from centres of Latin learning. Monasteries, cathedrals, and ecclesiastical institutions are the places where Latin was very important (Bullitta and Van Deusen Reference Bullitta and Van Deusen2022). Equally fundamental are linguistic treatises, a corpus of grammatical and poetic texts produced between the mid-12th and mid-14th centuries. These divide into two subcategories: poetological and grammatical treatises (Clunies Ross Reference Clunies Ross2005). For all of these works Patzuk-Russell (Reference Patzuk-Russell2021, 213) uses the expression vernacular grammatica. The earliest known poetological work, Háttalykill inn forni (a metrical key), was likely composed c. 1140 by Earl Rögnvaldr of Orkney and Hallr Þórarinsson (Helgason and Holtsmark Reference Helgason and Holtsmark1941). The most significant text, however, is Snorri Sturluson’s Edda (13th century), written to foster King Hákon Hákonarson’s engagement with skaldic poetry. Its tripartite structure, Gylfaginning (Faulkes Reference Faulkes2005), Skáldskaparmál (Faulkes Reference Faulkes1998), and Háttatal (Faulkes Reference Faulkes2007), provided aspiring poets with both a framework for traditional verse and the technical tools for new composition (Nordal Reference Nordal2001).

The grammatical treatises (GTs), preserved in the 14th-century Codex Wormianus, are numbered sequentially on the basis of their order in the manuscript:

1. 1. The First Grammatical Treatise (FGT; c. 1150) addresses orthography and phonetics.

2. 2. The Second Grammatical Treatise (SGT; c. 1270 to 1300) expands on these themes.

3. 3. The Third Grammatical Treatise (TGT; c. 1250) adapts Priscian’s Institutiones grammaticae (Books 1 to 2) and Donatus’s Ars maior (Book 3).

4. 4. The Fourth Grammatical Treatise (FoGT; c. 1330 to 1340) focuses on rhetoric.Footnote ¹

While scholars agree on Latin influence (Raschellà Reference Raschellà1983, Reference Raschellà, Quinn, Heslop and Wills2007), the degree and mechanisms of adaptation – whether direct borrowing, creative reworking, or hybrid innovation – remain contested. Identifying Latin pedagogical methodologies requires close analysis of literary sources. Evidence emerges from Biskupa sögur (Bishops’ sagas; Parker and O’Donoghue Reference Parker and O’Donogue2024) and hagiographic literature (Jensson Reference Jensson and Goullet2017). This linkage is unsurprising, as Latin proficiency was chiefly clerical; thus, ecclesiastical biographies such as Jóns saga helga, Lárentíus saga, and Mírmanns saga document teaching practices (Patzuk-Russell Reference Patzuk-Russell2021, 160, 186 n. 671).

Beyond sagas, glossaries provide another critical category of evidence. The late 12th-century GKS 1812 4to preserves a concise glossary of 190 Latin terms with Old Icelandic equivalents (Scardigli and Raschellà Reference Scardigli, Raschellà and Weber1988, 299). Related material appears in AM 249 I fol., likely from the same original codex, alongside three Latin–Old Icelandic pairs in AM 671 4to (14th century) and six more in Codex Wormianus (AM 242 fol.), later obscured by 15th-century verses (Raschellà Reference Raschellà, Bergmann, Glaser and Moulin-Fankhänel2001). Though not designed exclusively for Latin instruction, pedagogical manuscripts included adaptable learning material. AM 921 III 4to, the sole surviving example, reflects the breadth of medieval scholarly education.

Educational system

The transfer of knowledge between generations constitutes a characteristic of oral cultural systems (Ong Reference Ong2012). This process assumes complexity in literate societies, where cultural continuity becomes dependent upon the mastery of literacy skills. The medieval European context presented challenges, as Latin functioned as an acquired scholarly language across the continent. Various regions adopting Latin Christianity developed distinct approaches to meet these linguistic requirements (Reynolds Reference Reynolds1996). One approach entailed the translation of Bible into vernacular languages, exemplified by the case of Iceland, where segments of sacred texts were rendered into the local tongue at an early stage (Bampi et al. Reference Bampi, Buzzoni and Khalaf2015). This process of translation demanded linguistic competence in Latin and a rigorous adherence to semantic and theological precision. Ensuring an accurate correspondence between the original Latin text and its vernacular counterpart was paramount, as only through such meticulous fidelity could the divine message retain its doctrinal integrity across linguistic boundaries (Leclercq Reference Leclercq1982).

In Iceland, as in contemporaneous Europe, Latin literacy served as the fundamental prerequisite for preserving the doctrinal orthodoxy of Christian teachings (Quinn Reference Quinn and Clunies Ross2000). This necessity engendered the development of an educational system centred on the institutions of the new faith (Bouressa Reference Bouressa2022). The earliest documented evidence of such formal instruction dates to the 12th century, coinciding with the establishment of ecclesiastical schools throughout Iceland. Among these, the cathedral schools at Skálholt and Hólar functioned as the principal centres for clerical training, employing professional instructors remunerated for their instruction in grammar (Cormack Reference Cormack and Bernharðsson2016). Although sporadic instances of uncompensated teaching are attested, these appear to be exceptional deviations from the norm. The existence of monastic schools remains less definitively established, though fragmentary evidence hints at their probable operation (Gunnlaugsson Reference Gunnlaugsson and Bernharðsson2016). Current scholarship provides the most comprehensive insights into the education of the clergy (Barrow Reference Barrow2015), which primarily targeted young ecclesiastical students and emphasised Latin proficiency as an indispensable element of liturgical formation (Vitz Reference Vitz, Begley and Kotersky2005). Iceland’s educational system seems to adhere to a tripartite curricular structure comprising basic, intermediate, and advanced tiers, each demanding distinct financial investment. All pupils commenced their studies by acquiring proficiency in the Latin alphabet, a pedagogical priority evidenced by the pronounced emphasis on orthographic precision in Iceland’s grammatical treatises. This is particularly exemplified by the FGT and SGT, which systematically address orthographic conventions.

The elementary curriculum was explicitly tailored to fulfil the core competencies required for priestly functions. A 1504 record from Skriða monastery, preserved in Diplomatarium Islandicum (DI 1857–1872; VII, pp. 714–715), is a primary source attesting to this program. The document details a prior’s contractual obligation to educate a novice until his ordination, delineating a syllabus organised around four pillars: lés (reading), sǫngr (chant), skrift (writing), and rím (computus). Subsequent stages of instruction focused on developing Latin reading comprehension, drawing primarily on Donatus’s Ars minor and Ars maior, as well as Priscian’s Institutiones grammaticae. The circulation of these grammatical works throughout Scandinavia is well-attested (Poli Reference Poli2008), even though Donatus’s texts do not appear in the limited number of extant inventories. Nevertheless, Donatus’s reception in Iceland can be established indirectly through his presence in the TGT, whose opening section constitutes a reworking of the Ars maior.

As ecclesiastical institutions, these schools predominantly employed sacred texts for pedagogical purposes, with the Psalter occupying a position of prominence in Iceland, mirroring its central role in broader European liturgical education. Traditionally attributed to King David, the Psalms were subjected to rigorous study and memorisation. They were also performed musically as an essential devotional practice. Musical instruction and plainsong constituted an indispensable element of the elementary curriculum. This core training was augmented by instruction in arithmetic and writing skills, the latter taught through wax tablets following established Anglo-Saxon pedagogical models. Numerical training served the function of facilitating Easter computus calculations. The majority of students progressed no further than this elementary stage, which provided inadequate preparation for active Latin composition and speech. These competencies remained exceptional. It bears emphasis that Iceland’s decentralised educational infrastructure permitted regional variations in curricular implementation. Nevertheless, the fundamental pedagogical objectives maintained consistency across geographical divisions. For parish priests discharging routine liturgical obligations, functional literacy in sacred texts and competence in performing offices proved sufficient. For the minority possessing both financial resources and scholarly ambition, the educational system afforded intermediate-level instruction encompassing advanced Latin studies, secular disciplines such as jurisprudence, and, in select cases, opportunities for continental study. Complete Latin mastery represented a significant intellectual achievement, which conferred both the capacity for active Latin discourse and the ability to engage with sophisticated theological and legal exegesis (Patzuk-Russell Reference Patzuk-Russell2021, 177–190).

While Iceland’s medieval educational system differed from the conventional European model organised around the Trivium and Quadrivium (the seven liberal arts), it shared the fundamental emphasis on grammatical training as the foundation of learning. Latin maintained substantial cultural capital for ecclesiastical figures across the social spectrum, from elite prelates to parish clergy. This valuation is particularly well-documented in Jóns saga baptista, where multiple passages explicitly foreground grammatical training as an essential component of cathedral school education (Cormack Reference Cormack1994). The saga further ascribes soteriological significance to grammatical knowledge, framing it as consequential for individual salvation. This elevated status of Latin persisted as a consistent feature of Icelandic intellectual life throughout the Middle Ages (Sigurdsson Reference Sigurdson2016).

This functional orientation also shaped the contexts in which Latin learning took place, prompting scholars to reconsider the assumption that education was confined exclusively to ecclesiastical institutions. In this light, current scholarship has posited the existence of non-ecclesiastical educational practices occurring within domestic environments or through fosterage networks (Hansen Reference Hansen and Lewis-Simpson2008). In the absence of direct documentary evidence, the hypothesis is nevertheless substantiated by circumstantial indicators, foremost among them the paradigmatic example of Snorri Sturluson (1179 to 1241), a lay scholar of remarkable erudition (Clunies Ross Reference Clunies Ross1987). This education extended to members of his family, in particular two of his nephews, Óláfr hvítaskáld Þórðarson (1210/1212 to 1259; Goetting Reference Goetting, Gade and Marold2017) and Sturla Þórðarson (1214 to 1284; Gade Reference Gade, Gade and Marold2017), both authors of several works in prose and verse. An examination of the works produced by members of Snorri Sturluson’s family suggests the presence of elite secular pedagogical structures, within which the curriculum may have comprised the following domains:

• Poetic composition and analysis.

• Jurisprudential studies (particularly relating to legal traditions).

• Historiographical and genealogical transmission.

• Runic literacy and epigraphic practice.

While the precise extent of secular education’s contribution to Iceland’s textual culture remains challenging to quantify, its role in shaping medieval Icelandic intellectual life seems decisive. The boundaries between religious and secular pedagogy were often permeable, as demonstrated by the formative education of Saint Þorlákr, whose simultaneous engagement with clerical training and secular scholarship epitomises this institutional fluidity (Patzuk-Russell Reference Patzuk-Russell2021, 90–91).

A final dimension of this educational system warrants consideration, as evidence from the sagas suggests that women could also participate, often receiving instruction in the hannyrðir, that is, handicrafts and specialised textile work. In some cases, however, women’s education appears to have encompassed book learning from an early age and proficiency in multiple languages. This is illustrated by Gratiana, daughter of the emperor of Grikkland in the Dámusta saga. The most notable case is that of Ingunn, a woman documented as possessing expertise in both grammatical and theological domains. According to Jóns saga helga, Ingunn pursued an educational trajectory that necessarily diverged from the standardised clerical training available to male aspirants. She taught at Hólar without receiving remuneration. As noted by Patzuk-Russell (Reference Patzuk-Russell2021, 44–46), the education of aristocratic women can be interpreted as partially overlapping with clerical instruction. Beyond this, the sources provide little additional information, offering no insight into the accessibility of such education for women.

Within Iceland’s educational framework, Latin operated in constant dialectical relationship with the vernacular across all pedagogical levels, manifesting in both liturgical practice and instructional settings. Unlike contemporary academic approaches to language study, medieval Latin acquisition was never conceived as an autonomous scholarly pursuit. Rather, it functioned as a pragmatic tool for accessing specialised knowledge, most notably within the intertwined spheres of theology and jurisprudence.

Latin pedagogical texts in medieval Iceland: inventories and implications

The manuscript AM 732b 4to preserves the only known examples of macaronic Latin poetry from medieval Iceland (Patzuk-Russell Reference Patzuk-Russell2021, 172 n. 615). This singular survival underscores the gap in understanding the role of Latin in medieval Icelandic culture. As previously noted, the absence of Latin manuscripts and original Latin compositions presents challenges. However, máldagar (ecclesiastical inventories) offer valuable insights into which Latin texts were available during this period (Jensson and Marne Reference Jensson and Marne2014). None of the surviving booklists contain works exclusively dedicated to education. The first booklist comes from the episcopal seat at Hólar. A 1396 inventory (DI III, 613) reveals a collection divided between devotional/liturgical texts and educational materials. The latter are divided into two categories (Clunies Ross and Wellendorf Reference Clunies Ross and Wellendorf2014, xl–xli): Iure (legal texts) and Skólabækr (school books), including Doctrinalia ij (two copies), Brito á tveim bókum, Huguicio, and Græcismus. The presence of two copies of the Doctrinale, Alexander of Villa Dei’s versified grammar (1199), and Eberhard of Béthune’s Græcismus (1212) demonstrates the adoption of pedagogical innovations from the continental Europe. Brito likely could be either William Brito’s Expositiones vocabulorum biblie or the Summa Britonis (1248 to 1267), a biblical lexicon. Huguicio refers to Hugutio of Pisa’s Derivationes (late 12th century). The Hólar inventory of skólabækr incorporates grammatical treatises (Doctrinale and Græcismus) as well as lexical resources (Derivationes and Expositiones) within the designated ‘school books’.

A second booklist, originating from the episcopal seat at Viðey and dated to 1397 (DI IV, pp. 110–111), similarly employs the designation ‘school books’ (Item í skólabókum). This inventory includes the Doctrinale and Græcismus, followed by Aurora, a biblical compendium attributed to Peter Riga, and by Alexander Magnus, a glossed version of Tobias, the ubiquitous Disticha Catonis (here in glossed form), nine additional poetic works, a collection of Marian miracles, a Life of Saint Margaret, and two breviaries (Clunies Ross and Wellendorf Reference Clunies Ross and Wellendorf2014, xl).

The 1461 inventory from Möðruvellir’s Augustinian monastery (DI V, p. 288) adopts a different organisational principle, categorising volumes as ‘Latin books’ (í latinubókum) and ‘Old Norse texts’, followed by unclassified liturgical works. The Latin section features the Grecissimus (Eberhard of Béthune’s aforementioned Græcismus) and the Minus volume containing the syntactic books of Priscian’s Institutiones grammaticae (5th to 6th century CE). Also present are Isidore of Seville’s Synonyma and De summo bono (6th to 7th century CE), alongside the Vita Thobiae, commonly associated with Matthew of Vendôme’s Tobias (c. 1174 to 1206) and employed as a reading primer. Some overlap exists between the books classified here as Latin and those classified above as ‘school books’. While it is difficult to assert with certainty that these texts served a pedagogical purpose, such a function is plausible, given that they all originate from an episcopal school intended for the training of ecclesiastical personnel.

A comparison of the three episcopal inventories offers several observations. The canonical Latin auctores, Ovid and Virgil in poetry and Cicero and Caesar in prose, are absent, whereas Latin didactic poetry, such as the Doctrinale and Græcismus, is consistently represented. At this stage, a strong caveat is warranted: While these data invite interpretation, the implications proposed here remain plausible hypotheses rather than conclusions firmly supported by evidence. From the extant sources, a curricular preference for medieval Latin texts appears evident; however, losses in transmission prevent any definitive assertion.

Teaching Latin grammar

The study of Latin grammar in Iceland is substantiated by multiple evidentiary strands. The first is direct textual witnesses, including a conjugated paradigm of the verb amo accompanied by Old Icelandic equivalents and Latin–Icelandic glosses (Raschellà Reference Raschellà, Bergmann, Glaser and Moulin-Fankhänel2001; Scardigli and Raschellà Reference Scardigli, Raschellà and Weber1988), along with grammatical fragments preserved in manuscript AM 921 III 4to. The second one is indirect testimony from narrative sources, particularly Lárentíus saga and Jóns saga Helga, the former chronicling the educational trajectory of a 14th-century Icelandic schoolmaster whose formative training concentrated on grammatical studies and the latter recounting Þóroddr Gamlason’s exceptional acquisition of grammatical knowledge through passive exposure to priestly instruction (Biskupa sögur I, vol. 2, p. 206). Manuscript AM 238 XXIII fol. (15th century) explicitly stipulates that priests were required to demonstrate competence in distinguishing Latin grammatical genders, underscoring the linguistic expectations embedded in clerical formation.

The precise pedagogical instruments employed remain more elusive to reconstruct. Extant booklists conspicuously omit copies of Donatus’s works, while Priscian is represented exclusively through the final two books addressing syntax. Nevertheless, knowledge of both grammarians is demonstrably preserved through vernacular adaptations, most the TGT, which incorporates extensive translated excerpts from both authorities. This suggests that grammatical instruction could be effectively delivered with limited textual resources, potentially relying on memorised exemplars. Latin medieval pedagogical treatises undoubtedly constituted a fundamental component of instruction, particularly the versified Doctrinale and Græcismus, whose Icelandic circulation is verified both through direct manuscript evidence and the derivative adaptations in the FoGT (Longo Reference Longo2024). But they presented hermeneutic challenges. Their highly condensed format offered grammatical terminology with minimal explication and selective exemplification. Although the advantages of their verse structure facilitated retention, this characteristic impeded conceptual clarity. These texts functioned primarily as didactic aids rather than autonomous reference works, serving as bases for vernacular explication in classroom settings. Privileged access to personal collections constituted an exceptional circumstance, as documented at Mǫðruvellir.

Techniques

For the development of oral proficiency, historical evidence confirms the use of reading aloud as a foundational technique, frequently implemented through the repetition. This method followed a graduated progression: first, the repetition of isolated sounds; second, the formation and reiteration of syllables; and finally, the repeated articulation of complete words. The primary text employed for this introductory training was the Pater Noster. The mnemonic retention and performative delivery of textual materials persisted as cornerstone didactic priorities. Within this framework, poetic composition occupied a position of exceptional pedagogical importance, as substantiated by multiple evidentiary strands. The 1397 Viðey register enumerates nine distinct poetic codices among its documented scholastic holdings. Iceland’s curricular practices mirrored broader European tendencies in systematically including versified grammatical treatises (Ommundsen Reference Ommundsen and Eriksen2016). Manuscript AM 792 4to preserves Latin grammatical and prosodic marginalia (Frank Reference Frank1909, 145, n. 1), demonstrating conclusively that Latin versification retained at least a residual presence within the instructional paradigm (Patzuk-Russell Reference Patzuk-Russell2021, 182).

To reinforce oral production skills, the technique of singing was extensively used. The Psalms, memorised and performed through song, served a dual purpose: They were integral to the liturgical obligations of clerics, who were required to chant in Latin during mass, while also functioning as a pedagogically effective tool for language acquisition. By embedding repetition, a critical component of language learning, within a melodic framework, singing transformed rote practice into an engaging exercise. In essence, it provided a form of linguistic pattern drilling, reinforcing syntactic and phonological structures through rhythmic and melodic repetition. The reading-aloud technique was also applied at more advanced levels, where it was practiced using liturgical books. Within this instructional context, where syllabification, recitation, singing, and precise Latin pronunciation converged, the SGT was likely composed, given its explicit focus on the interplay between orthography and musical notation.

A vivid portrayal of the educational milieu at Hólar is provided in Jóns saga helga, which offers details about Ingunn’s pedagogical activities, shedding light on further teaching techniques employed in Icelandic classrooms. One of them involved her oral correction of Latin texts as they were being read. This suggests that in-class Latin readings were followed by a phase of comprehension and interpretation, achieved through vernacular translation. While evidence remains limited, it is plausible that translation from Latin into Icelandic played a role in medieval Icelandic education.

Glossing technique involves the systematic pairing of Latin lexical items with their Icelandic vernacular equivalents. It is necessary to distinguish the interlinear gloss from the translation. Glossing shares certain functional similarities with translation, particularly in providing first-language (L1; Icelandic) equivalents for second-language (L2; Latin) terms, but it exhibits distinct characteristics. Interlinear glossing operates at the lexical or phrasal level, rarely extending beyond discrete textual units. Its primary purpose was to elucidate obscure, technical, or uncommon vocabulary, with frequently encountered terms typically remaining unglossed. The widespread practice of glossing throughout medieval Europe is well-attested in the scholarly record (Raschellà Reference Raschellà, Bergmann, Glaser and Moulin-Fankhänel2001). The Icelandic context suggests extensive use of glossing, as indicated by the recurrent appearance of Brito (a prominent glossary) in manuscript inventories.

The pedagogical importance of glossing demands serious consideration. Scardigli and Raschellà (Reference Scardigli, Raschellà and Weber1988, 311) delineate three potential pedagogical applications for glosses. They may have functioned as exercises in Icelandic orðspeki (verbal erudition), potentially connected to the þulur (catalogues of poetic terminology) frequently preserved alongside Snorri’s Edda. They could represent attempts to facilitate Latin as a spoken language within certain religious communities. They might constitute annotations by Icelandic travellers who encountered daily Latin usage abroad and recorded these for instructional purposes. Additionally, Raschellà (Reference Raschellà, Bergmann, Glaser and Moulin-Fankhänel2001, 589) proposes that glosses may have served as structured ‘vocabulary exercises’ produced by either instructors or students in the classrooms. These may have been the contexts in which Latin and Icelandic interacted most strongly. The result of this intersection was most evident.

The Icelandic metalanguage present in Icelandic GTs (91 terms in total) can be divided into the following types (Raschellà Reference Raschellà, Di Sciacca, Giliberto, Rizzo and Teresi2018, 619–622): 39 loanwords (Greco-Latin terms, either preserved unchanged or adapted in some way to Icelandic); 17 Greco-Latin terms cited in their original form and subsequently rendered freely into Icelandic; 19 calques (lexical and semantic), and 16 Icelandic terms (not derived from a Latin source). It is reasonable to assume that the metalanguage present in GTs is the result of interlinear glossing of Latin grammatical texts used in classrooms. Although direct evidence is lacking, the numerical consistency of Icelandic grammatical terms of Latin origin suggests that this may have been the case. Of the grammatical terms attested in the treatises (91 words), 82 % (75 words) are of Latin derivation. They seem to prove the interference between the two languages and in some cases even phenomena of code-switching (in particular, the group 2, Greco-Latin terms cited in their original form and subsequently rendered freely into Icelandic). On this basis, it is possible to hypothesise that Latin–Icelandic glossaries functioned as standard classroom resources. These tools were employed not solely for grammatical instruction but across multiple disciplines (Patzuk-Russell Reference Patzuk-Russell2021, 204–209). Glossing emerges as a distinctive technique for Latin acquisition in Iceland.

Regarding the development of writing skill, the instructional sequence began with practicing the Latin alphabet on wax tablets or similar surfaces. Current evidence does not permit definitive conclusions about whether the curriculum included exercises aimed at developing actual Latin composition abilities. We can say the same about the assessment of acquired competencies. The extant evidentiary corpus yields only fragmentary insights regarding assessment protocols, and we cannot establish with certainty the specific methodologies deployed to gauge either the attainment of instructional objectives or the precise gradations of linguistic proficiency achieved. Nevertheless, documentary evidence attests to the implementation of corrective measures for deficiencies in textual recitation. The Lárentíus saga provides illuminating testimony, documenting that candidates designated for liturgical readings during solemn masses submitted to preliminary evaluations conducted by the scholastic authority on the preceding evening, with disciplinary interventions imposed for substandard performance (Biskupa sögur III, p. 373).

The sources provide little information regarding the advanced stages of pedagogy. However, by correlating the evidence from Icelandic sources with knowledge drawn from other geographical contexts, we can cautiously infer that students were expected to attain a certain level of oral fluency in Latin through direct interaction with proficient speakers. Latin functioned as the indispensable lingua franca for clerics across Europe, including major centres such as Paris, Rome, and various German principalities (Mcdougall Reference Mcdougall1986), and served as the primary medium of ecclesiastical communication in interactions with non-Norse populations.