Language and Linguistics in Medieval Europe
- Deborah HaydenDeborah HaydenMaynooth University
During the period from the fall of the Roman empire in the late 5th century to the beginning of the European Renaissance in the 14th century, the development of linguistic thought in Europe was characterized by the enthusiastic study of grammatical works by Classical and Late Antique authors, as well as by the adaptation of these works to suit a Christian framework. The discipline of grammatica, viewed as the cornerstone of the ideal liberal arts education and as a key to the wider realm of textual culture, was understood to encompass both the systematic principles for speaking and writing correctly and the science of interpreting the poets and other writers. The writings of Donatus and Priscian were among the most popular and well-known works of the grammatical curriculum, and were the subject of numerous commentaries throughout the medieval period. Although Latin persisted as the predominant medium of grammatical discourse, there is also evidence from as early as the 8th century for the enthusiastic study of vernacular languages and for the composition of vernacular-medium grammars, including sources pertaining to Anglo-Saxon, Irish, Old Norse, and Welsh. The study of language in the later medieval period is marked by experimentation with the form and layout of grammatical texts, including the composition of textbooks in verse form. This period also saw a renewed interest in the application of philosophical ideas to grammar, inspired in part by the availability of a wider corpus of Greek sources than had previously been unknown to western European scholars, such as Aristotle’s Physics, Metaphysics, Ethics, and De Anime. A further consequence of the renewed interest in the logical and metaphysical works of Aristotle during the later Middle Ages is the composition of so-called ‘speculative grammars’ written by scholars commonly referred to as the ‘Modistae’, in which the grammatical description of Latin formulated by Priscian and Donatus was integrated with the system of scholastic philosophy that was at its height from the beginning of the 13th to the middle of the 14th century.
The Classical conception of the liberal arts as the defining framework of the educational curriculum, encompassing a trivium of verbal arts (grammar, rhetoric, and dialectic) followed by a quadrivium of scientific disciplines (arithmetic, geometry, music, and astronomy), held considerable sway throughout the medieval period. Late Antique sources, such as the De ordine of the influential Christian theologian and philosopher Augustine of Hippo (354–430), reflect Neoplatonic speculation about the pursuit of knowledge, whereby reason was understood to seek the contemplation of divine things through mastery of the liberal arts (Marenbon, 1994, p. 172). Although the disciplines of grammar and rhetoric formed the basis for the study of language throughout the medieval period, they were generally conceived in broader terms than their modern-day counterparts. Thus the art of grammar was commonly defined as the discipline that treated both the systematic principles for speaking and writing correctly, as well as the science of analyzing and interpreting poets and other writers. Its chief function was to facilitate the study of a canon of texts that were seen as essential for the education of literate citizens, and therefore to promote the authority of written tradition generally (Irvine, 1994, p. 79). The art of rhetoric, on the other hand, entailed mastery of form and style for both oral delivery and a variety of literary genres. The two disciplines were not always clearly distinguished in the medieval period, however, and many grammatical sources treat both linguistic principles and stylistic matters alongside one another.
Our knowledge of linguistic thought in Europe during the medieval period is, as in other areas of scholarship, largely limited to the evidence of surviving manuscripts, and often this does not paint a representative picture of intellectual activity in any given region. The matter is further complicated by the fact that the transmission of Classical and Late Antique grammatical texts was by no means uniform. Literate education was conducted in monastic centers for much of the medieval period, and most individual libraries held no more than half a dozen or so different texts on grammar, with collections varying considerably from one location to another (Law, 2003, p. 112). Following the dissolution of the Roman Empire in the late 5th century, Greek philosophy and literature continued to be subjects of study in the Eastern Empire, where Byzantine scholars wrote explications and commentaries on the Tekhnē grammatikē of Dionysius Thrax and the syntactical treatise by Apollonius Dyscolus. The Western Empire fared less well with regard to the transmission of Greek texts, which were for the most part unknown to scholars until ca. 1100 onward. Much Classical literature was lost following the fall of the Empire, but the spread of Christianity across Europe, and the contribution of ecclesiastical centers to the preservation of earlier educational tradition, resulted in the survival of many important texts and in the maintenance of Latin as the language of the Church and of textual learning (Robins, 1997, p. 80).
Latin grammatical manuals from the Late Antique period that survived in western Europe were typically aimed at either native speakers or advanced foreign students, and did not provide the concise but comprehensive introduction to Latin inflection required by pupils, who not only had little contact with native speakers of Latin, but in many cases also lacked literacy skills in their own vernacular. In her pioneering work on the extant medieval Latin grammars, Vivien Law (1994, pp. 88–92, 2003, pp. 65 & 83) classified these sources into two principal but overlapping categories. Works belonging to the so-called Schulgrammatik genre offer a systematically structured introduction to basic concepts, such as the parts of speech and their properties, and are characterized by a rigorously hierarchical and systematic structure and by a tendency to foreground semantic categories (since it is assumed that the forms of Latin would be known to native speakers). In contrast, grammars belonging to the so-called Regulae genre provide long lists of examples of morphological phenomena, and were apparently intended more for reference than for systematic study, being aimed at advanced learners who had already mastered the basic forms of the language. Examples of this genre include the works of the Byzantine scholars Phocas and Eutyches, who wrote detailed accounts of nominal declension and verbal conjugation, respectively. Several writers produced grammars that combined aspects of the Schulgrammatik and Regulae types, including Consentius, Diomedes, and Priscian (for editions of the works by these and other authors, see Keil, Hertz, Mommsen, & Hagen, 2009). Faced with the problem of how to teach grammar to nonnative speakers of Latin, many early medieval scholars composed didactic works that show evidence of pedagogical experimentation on the basis of these sources, including commentaries, supplementation from multiple sources, or collections of paradigms organized in different ways: for example, nouns categorized according to gender, termination, or language of origin (Law, 1994, pp. 90–91).
Among the most influential grammatical texts in the medieval European world were the works of Aelius Donatus (ca. 350 AD). His Ars minor, a short treatise on the parts of speech written in question-and-answer form, remained the standard Latin primer throughout the Middle Ages. Donatus’ more advanced Ars maior was also a popular and widely known text, which was divided into three parts: the first book dealt with sound (uox), letter/speech-sound (littera), syllables, metrical feet, accents, and punctuation; the second, with the parts of speech; and the third, with stylistic matters, such as barbarisms, solecisms, metrical faults, metaplasms, schemes, and tropes (for an edition and detailed discussion of this text, see Holtz, 1981). The Ars maior provided the standard structure for any treatment of grammar until it was replaced by the fourfold division into ortographia, prosodia, ethmologia, and diasintastica (‘syntax’) current in the later Middle Ages (Copeland & Sluiter, 2009, p. 82). The widespread currency of Donatus’ work is demonstrated by the use of his name in some vernacular languages as a general reference to a grammar or elementary book of instruction, for example, dwned in Middle Welsh, itself a borrowing from Middle English donet (Matonis, 1981, p. 122; Russell, 2016, p. 139). Donatus’ work was also the subject of numerous commentaries in the Late Antique and medieval periods by authors like Servius, Sergius, Cledonius, Pompeius, and Julian of Toledo, who provided explanations of technical terms, elucidations of difficult concepts in Donatus’ text, and many additional examples (these texts are also edited by Keil et al., 2009).
Alongside the works of Donatus, the writings of Priscian exerted a considerable influence on grammatical thinking from the early medieval period onward. A teacher at Constantinople in the early 6th century, Priscian was chiefly known in the medieval period for three distinct works on grammar aimed at native speakers of Greek. The first, known as the Institutio de nomine et pronomine et verbo, was a beginner’s textbook dealing principally with morphology that swiftly taught students how to decline and conjugate, and was a popular textbook among early medieval teachers in non-Latin-speaking areas (Jeudy, 1972; Passalaqua, 1993). The somewhat longer Partitiones is a similarly introductory text consisting of a minute analysis, in question-and-answer form, of the first line of each book of Virgil’s Aeneid; this work was not used extensively in the early Middle Ages, but was rediscovered during the Carolingian Renaissance, when it helped inspire the new pedagogical genre of the ‘parsing grammar’ (Law, 2003, pp. 87 & 148). Priscian’s most extensive and widely known work, however, was the Institutiones grammaticae, a monumental treatise in eighteen books that deals with everything from speech-sounds and orthography to Latin inflectional and derivational morphology, the parts of speech, and syntax, the last of which seems to have never before been treated systematically in the Latin grammatical tradition (on Priscian’s syntactical description, see Luhtala, 2013, p. 347). The Institutiones is characterized by an attempt to improve on traditional Latin grammar by applying Greek theoretical ideas, and it is a valuable repository of excerpts from Roman literary sources that are invoked in the text to serve as illustrative examples of grammatical phenomena (Copeland & Sluiter, 2009, p. 168). Like the Partitiones, the Institutiones grammaticae was rediscovered in the early 9th century, when it was studied in conjunction with Aristotle’s writings on dialectic (Law, 2003, p. 91).
Another important conduit for Classical and Late Antique grammatical knowledge in the Middle Ages are summaries found in works of an encyclopedic nature, such as Book III of the De nuptiis Mercurii et Philologiae (On the marriage of Mercury and Philology), by the 5th-century African scholar Martianus Capella. The De nuptiis is an allegorical work that provides a set of introductions to each of the seven liberal arts, and it survives in some 250 medieval manuscripts (Contreni, 2014, p. 91); it enjoyed particular popularity in the later 9th and 10th centuries (Marenbon, 1994, p. 173). Knowledge of grammar and the other arts was also transmitted in the medieval period via the Institutiones of the 6th-century Roman senator Cassiodorus, whose chapter on rhetoric highlights the importance of oratorical training in the Late Antique period. It cites earlier sources, such as Cicero’s De inventione and De oratore, alongside Quintilian’s Institutio oratoria, and is often found by itself in medieval manuscripts as a kind of preface or supplement to other widely known rhetorical manuals (Copeland & Sluiter, 2009, p. 212). The most influential of the encyclopedic sources was the Etymologiae of Isidore of Seville (ca. 560–636), which treated all aspects of knowledge within a Christian framework. Isidore’s discussion of the liberal arts begins with a comprehensive account of grammar that combines a technical approach to the discipline with explanations of the origins of ideas. His encyclopedia also played a central role in popularizing medieval etymological method as a discursive practice that uses different verbal or extraverbal criteria to explain the origins and significations of words (Amsler, 1989). Etymology in the medieval sense of the term could involve a number of different processes: for example, aspects of word-formation, such as analysis of a compound or derivative, or explanation of a word in terms of onomatopoeia or sound-symbolism or by association with one or more similar-sounding words that were felt to shed light on the meaning of the word in question. An important aim of this approach was to imbue words with a spiritual or allegorical dimension by way of interpretation (Gneuss, 1990, p. 22).
Both encyclopedias and grammars were also important sources for condensed accounts of rhetoric, in addition to the three main Roman sources used to study this subject in the Middle Ages: namely Cicero’s De inventione, the anonymous Rhetorica ad Herennium, and Quintilian’s Institutio oratoria. Donatus’ section on tropes and figures in Book III of his Ars maior, moreover, served as a model for many medieval discussions of stylistic devices as they applied to both oral and written usage.
2. The Early Middle Ages (ca. 500–800)
Following the spread of Christianity throughout Europe from the 5th century onward, the chief purpose of grammatical study was increasingly perceived as facilitating an enhanced understanding of the Latin Scriptures. Latin was not only the principal medium of grammatical manuals inherited from late-imperial teachers of the 4th and 5th centuries, but also the language of the early medieval Church, which was responsible for the preservation and production of written texts and the cultivation of literate education. Consequently, mastery of Latin was a necessity for any scholar working in an ecclesiastical milieu during this period. The exegetical techniques learned in the study of grammatica, which encompassed the four divisions of reading (lectio), interpretation (enarratio), correction (emendatio), and criticism (iudicium), were also those applied to the Bible (Irvine, 1994, pp. 4–5).
While the role of early medieval ecclesiastical centers in preserving the literature of Greece and Rome is inestimable, many scholars of this period nevertheless sought to justify the study of earlier grammatical texts in which technical points were illustrated by examples from pagan authors, such as Virgil or Cicero. The didactic works of the Anglo-Saxon monk Bede, active in the early 8th century at Wearmouth-Jarrow in what was then the Kingdom of Northumbria, serve as an apt example of the interest that medieval European scholars had in integrating Latin grammatical pedagogy with Christian doctrine. In addition to a treatise on orthography, Bede composed the influential teaching-text De arte metrica, a discussion of the composition of Latin verse that drew on authors like Donatus and Servius, but used examples from Christian poets in addition to Virgil. His De schematibus et tropis, an explanation of figures and tropes, likewise illustrated each of Donatus’ examples with excerpts from the Bible or Christian writers (Law, 2003, p. 116; on Bede’s grammatical works, see Martin, 1984, and King, 1979). Similarly, the Benedictine monk Smaragdus of Saint-Mihiel (ca. 760–ca. 840) replaced pagan examples with excerpts from Biblical sources in his commentary on Donatus, and attempted to link grammatical concepts with scriptural ones—comparing, for example, the eight parts of speech to the eight passengers of the Ark or the eight Beatitudes (Law, 1994, pp. 100–101; on Smaragdus’ grammar, see also Holtz & Lambert, 1986; Löfstedt, Holtz, & Kibre, 1986).
Biblical narratives concerning the origins of language also formed an important part of medieval grammatical discourse. Some sources link the derivation of words to Adam’s role as the first to name all of God’s newly created animals, or explain linguistic diversity by citing the narrative of the Tower of Babel, according to which God dispersed all the languages of men as a punishment for human arrogance (for an example of the latter, see the discussion of the medieval Irish grammar Auraicept na nÉces in Section 5.4). Theological commentary also provided ample opportunity for medieval scholars to reflect on the meaning of signs. In his treatise De Trinitate, for example, Augustine contemplated the distinction between the verbum as ‘word-as-meaning’ and the vox as ‘word-as-form’, and compared these to the distinction made in the Gospel of John between the divine Word and the description of John the Baptist as a ‘voice crying in the wilderness’ (Law, 2003, pp. 101–108).
The work of the 7th-century grammarian known as Virgilius Maro Grammaticus, a figure whose possible Irish origins and unorthodox works have been a subject of considerable debate since the first half of the 19th century, demonstrate the extent to which the study of grammar was entrenched in the liberal arts curriculum of the early medieval period. Virgilius wrote two treatises claiming to deal with Latin grammar, the Epitomae and Epistolae, which contain a mixture of conventional grammatical material, seemingly modeled on the work of Donatus, as well as more tantalizing commentary in the realm of pseudophilosophy, religion, biographical and historical anecdotes, etymologies, and word games, among other elements, much of which cannot be traced to existing sources (Herren, 1979, pp. 27–29). Yet, despite the uncertainty surrounding the author and the texts on which he drew, it is clear the Virgilius’ works had considerable impact on Latin and vernacular grammatical and exegetical works written in Ireland and in Irish centers on the Continent during the 8th century. The most recent assessments of his writings attempts to place them in the context of an established tradition of ‘wisdom literature’ practiced by a group of Irish Latin scholars (Bracken, 2002; Law, 1995).
3. The Central Middle Ages (ca. 800–1100)
The study of language was a subject of revived interest during the flowering of scholarship, art, architecture, and literature that characterized the Carolingian Renaissance in the 8th and 9th centuries, during the reigns of the Frankish kings Pippin the Short, Charlemagne, Louis the Pious, and Charles the Bald. This was in large part due to the educational and church reforms instigated by Charlemagne, who reigned from 768 until his death in 814. Charlemagne was responsible for setting up numerous new schools and scriptoria and for overseeing the reproduction of many Classical texts. He also sought to standardize the writing of Latin and the script used in copying books, a strategy that had the specifically didactic purpose of defining the ideal of a textually learned clergy, able to instruct the people from the central texts of Christian tradition (Irvine, 1994, p. 305). To this end, Charlemagne assembled a number of leading scholars at his court, including Peter of Pisa, Theodulf of Orléans, the Frankish scholar Einhard, and Irish teachers, such as Dúngal and Clement.
The activities of these and later Carolingian scholars encompassed the analysis of many Late Antique grammatical texts, such as the Ars maior of Donatus, on which several Latin commentaries were composed during the 8th and 9th centuries by figures like Remigius, Murethach (Muridac), Sedulius Scottus, and Smaragdus (for a detailed discussion of these texts and their diffusion, see Holtz, 1981). The 9th century also saw the development of the ‘parsing grammar’ as a tool for elementary linguistic pedagogy. Drawing on the question-and-answer format popularized in a grammatical milieu by the work of Donatus, these tracts, many of which have been transmitted anonymously, follow the example of Priscian’s Partitiones in that they aim at a detailed discussion of specific words in a text as a way of teaching students the principles of grammatical analysis (Law, 2003, p. 148). Also extant from the 9th century are two treatises concerned with word order, which demonstrate that the Carolingians’ concern with construing sentences went beyond the comparatively elementary analysis of syntax offered in Priscian’s Institutiones by taking into account issues like compound sentences and embedded constructions (Luhtala, 2013, pp. 347–348).
Among the most important scholars of the Carolingian period was Alcuin of York, who became head of Charlemagne’s palace school in the last decades of the 8th century and played a key role in developing the methodological models and authoritative textual practices that would be used throughout the British Isles and the Continent during the early 8th century (Irvine, 1994, pp. 315–316). In a more explicit way than the works of Bede and Smaragdus, Alcuin’s writings affirm the role of grammar as the cornerstone of a thoroughly Christianized education. For example, his Disputatio de vera philosophia envisions the acquisition of the liberal arts as a means of achieving wisdom (sapientia), construed as insight into the correct reading of the Bible (Copeland & Sluiter, 2009, p. 273). Among Alcuin’s principal didactic works, which also include treatments of rhetoric, dialectic, and orthography, is the De grammatica, a treatise that draws on both the content and question-and-answer format of Donatus’ Ars minor, supplemented with material from Priscian. The dialogue form of the text opposes two teenage boys and nonnative students of Latin, one a Frank and the other a Saxon; the former poses questions to the latter, who occasionally refers for advice to a master (Copeland & Sluiter, 2009, pp. 273–274; the text is edited by Migne, 1844–1865, Vol. 101, cols. 849–902).
Alcuin’s influence on the study of grammar during the Carolingian period and afterward can also be seen in other areas. For example, he compiled a collection of excerpts on Priscian’s Institutiones grammaticae that was the first of its kind, and may have been responsible for popularizing this work in scholarly circles on the Continent. Priscian’s Institutiones was used by only a few select scholars during the 7th and 8th centuries, but around 800 there was a sudden growth of interest in the work, and numerous manuscript copies of it were made; the earliest of these emanate from monasteries close to Tours, where Alcuin was based at the turn of the 9th century (Law, 2003, pp. 145–146). Priscian’s more expansive treatment of grammatical metalanguage was an attractive feature for scholars working with his text, and much learned activity in Carolingian circles between the 9th and 12th centuries focused on analyzing and digesting Priscian’s doctrine, as is evidenced by the numerous glossed or annotated copies of the text that are still extant (Passalacqua, 1978). A testament to the importance of this glossatory activity is the fact that one of the most substantial corpora of evidence for the form of the Old Irish language between ca. 700 and ca. 900 AD survives in a single 9th-century copy of the Institutiones grammaticae that contains numerous glosses in the vernacular (Stokes & Strachan, 1975). Although commentary on Priscian’s text generally took the form of glossing prior to the 11th century, evidence of efforts on the part of early medieval scholars to comment on the Institutiones in a more systematic way have also survived, such as a fragmentary commentary by the 9th-century Irish scholar Sedulius Scottus (Löfstedt, 1977) and a complete work by the prominent 9th-century philosopher John Scottus Eriugena (Luhtala, 2000).
One of the features of Priscian’s Institutiones that seems to have appealed to Alcuin was the grammarian’s more philosophical approach to language, and this no doubt played a central role in the renewed interest in the study of dialectic and its application to grammar that characterizes Alcuin’s contribution to Carolingian scholarship. Although most works of Aristotle’s Organon were not rediscovered until the middle of the 12th century, Alcuin and his associates are known to have studied such texts as a Latin paraphrase of Aristotle’s Categories known as the Categoriae decem, Boethius’ Latin translation of Porphyry’s Isagoge (an introduction to the Categories), Aristotle’s De interpretatione, and Boethius’ commentaries on these works (Law, 2003, p. 147). Not only did these texts inspire an interest amongst Carolingian scholars in logic as a subject of study in itself, but they also encouraged them to experiment with the techniques of logical disputation as a general tool of intellectual inquiry across various disciplines. In particular, the more linguistically oriented nature of Aristotle’s earlier works on logic invited comparison with grammatical doctrine (Luhtala, 2000, p. 117). Many grammatical commentaries of this period explore problems of definition with regard to various linguistic concepts, such as the parts of speech, or seek parallels between linguistic and real-world phenomena (Law, 2003, pp. 150–155).
4. The Later Middle Ages (ca. 1100–1500)
The beginning of the 13th century saw the development of some of the first universities in cities like Paris, Oxford, Cambridge, Lisbon, Toulouse, and Naples. Grammar formed a core part of the arts curriculum in these establishments and was based on a set of prescribed texts, such as the Institutiones grammaticae of Priscian and the logical works of Aristotle. From the early 12th century, scholars in the medieval West had also become aware of numerous Greek philosophical sources that had previously been unknown to them, such as Aristotle’s Physics, Metaphysics, Ethics, and De Anime. Although they were familiar with some of Aristotle’s writings on logic through the translations of Boethius, Aristotle’s physical and ethical works had only circulated in the Greek East and the Islamic world during the early medieval period. By the middle of the 12th century, these works had spread to the West through the medium of Latin translations from both the original Greek texts and from Arabic versions, inspiring grammarians to ponder more theoretical dimensions of language.
Donatus’ Artes remained the basic introduction to grammar in the later medieval period, and both these manuals and Priscian’s Institutiones continued to be the subjects of study and commentary. Their dominance was challenged, however, by the popularity of works like the Doctrinale of the French scholar and teacher Alexander of Villa-Dei, completed in 1199. This text, which appeared in some 300 printed editions, illustrates a growing trend during the later medieval period for putting grammatical information, particularly of a lexical and morphological nature, into verse form for the purpose of facilitating memorization (Law, 2003, pp. 180–181; on this and related works, see also Grondeux, 2000). The Doctrinale deals with aspects of grammar like noun declension, syntax, prosody, accents, and figures of speech, and was probably aimed at an intermediate-level audience already familiar with the work of Donatus. The Graecismus of Evrard of Bethune, written in 1212 and taking its title from its opening section on Greek terms, also enjoyed considerable popularity in the later medieval period; it takes the form of a versified commentary on Donatus’ Ars maior in 4440 hexameter lines, and was added to the curriculum of the University of Paris in 1366 alongside the Doctrinale (Murphy, 1974, p. 151).
One of the earliest pedagogical grammars to incorporate a treatment of syntax is Hugh of St Victor’s De grammatica (ca. 1120), which introduced a number of innovations to the definition of the sentence found in the work of Priscian. The Doctrinale of Alexander of Villa-Dei and the Summa of Pietro da Iolella were among the first teaching grammars to employ the subject–predicate distinction, which was otherwise not assimilated into mainstream grammatical tradition until the 18th century (Luhtala, 2013, p. 352).
Many pedagogical works on language written during the later medieval period, in both prose and verse form, synthesize grammatical and rhetorical elements with the aim of teaching the precepts of poetics (the ars metrica or ars poetria), letter writing (the ars dictaminis), and preaching (the ars rithmica) for use in the Latin-based school curricula. Several such works were produced in the late 12th and early 13th century, particularly by scholars in northern France, England, and Germany. These include the Ars versificatoria of Matthew of Vendôme (ca. 1175), a prose work interspersed with numerous verse examples that treats aspects of verse composition in a fairly general way, and was evidently aimed at intermediate learners (Murphy, 1974, pp. 163–168). The slightly later Poetria nova of Geoffrey of Vinsauf, a 2,000-line poem in hexameter verse written between 1208 and 1213—shortly before its prose counterpart, the Documentum de modo et arte dictandi et versificandi—deals mainly with rhetorical topics, such as amplification, stylistic ornament, metrical faults, and oral delivery, and remained influential even until the end of the 15th century (Murphy, 1971, p. xxii; for an edition and discussion of the text, see ibid., pp. 27–108).
One of the consequences of the renewed interest in the logical and metaphysical works of Aristotle during the later Middle Ages is the composition of so-called ‘speculative grammars’ written by scholars commonly referred to as the ‘Modistae.’ These theoretical treatises, which existed side by side with more practical Latin teaching manuals, such as Alexander of Villedieu’s Doctrinale, represent the integration of the grammatical description of Latin formulated by Priscian and Donatus with the system of scholastic philosophy that was at its height from the beginning of the 13th to the middle of the 14th century (Robins, 1997, p. 88). Its proponents were a relatively small group of scholars based mainly in the faculty of arts at the University of Paris (Law, 2003, p. 174). The Modistae sought to integrate Priscian’s description of the Latin language with an investigation of the theory that underlies it; in so doing, they attempted to link the categories of grammar with those of logic, epistemology, and metaphysics, and aimed to identify the characteristics of a universal grammar that could be applied to all languages. In exploring the ways in which language could be linked to external reality, they also identified distinct ‘modes’ of meaning in language. Thus, the modi essendi (‘modes of being’), were understood to indicate the properties existent in an object of understanding, while the modi intelligendi (‘modes of understanding’) were the properties of the concept that is apprehended by the mind, and the modi significandi (‘modes of signifying’) were the properties of the sign that is used to express a given concept. Speculative grammars took a number of forms, including treatises on the eight parts of speech and commentaries on the Institutiones grammaticae of Priscian. An early example of the latter is the Summa super Priscianum (‘Compendium on Priscian’) of the 12th-century scholar Petrus Helias. The most complete textbook of speculative grammar is the De modis significandi seu grammatica speculativa of Thomas of Erfurt (ca. 1300), a work formerly attributed to John Duns Scotus, which was widely copied and commented upon in the Middle Ages (Bursill-Hall, 1972).
5. Vernacular Grammars
While the survival of Latin grammars provides evidence for the dominance that the language had over the realm of textual culture and international communication throughout medieval Europe, another important dimension of linguistic thought during the medieval period is the increasing engagement of scholars with the study of vernacular languages, many of which emerged in written form following the spread of literacy to various regions. A range of sociolinguistic factors were typically the driving force behind the creation of vernacular grammars, such as a perceived need to help students with limited knowledge of Latin, or to compensate for what was seen to be falling standards in the realm of poetic composition (Law, 2003, pp. 192–193). Grammatical descriptions of languages like Italian, Spanish, and Portuguese only appeared toward the beginning of the humanist era, for example, inspired in part by the gradual abandonment of Latin as an official language of communication and a concomitant need to understand an increasingly large and diverse literary corpus in the vernacular (Swiggers & Vanvolsem, 1987, p. 158). Other languages, such as Irish, Welsh, and Old Norse, possessed a vernacular literature from a much earlier stage of the medieval period, and it is thus unsurprising to find contemporary textual sources that explore the structure and features of those languages. Such sources often take the form of contrastive analysis between Latin and the vernacular through the medium of vernacular glosses on Latin grammars and other texts, or through glossary collections. In some cases, however, we find grammars of both Latin and the vernacular written through the medium of the vernacular. While vernacular-medium grammars of the medieval period often owe a considerable debt to the format and content of Latin textbooks, they also frequently show remarkable insight on the part of their authors into the distinctive features of the particular language with which they are concerned. The following section surveys some of the chief surviving sources for study of the European vernaculars during the medieval period prior to the advent of the printing press in the first half of the 15th century.
Latin had been introduced into England in the course of the 7th century and had been studied with the aid of Latin-medium Insular elementary grammars, such as the rather short-lived treatises of the Mercian scholar Tatwine and his Wessex contemporary Boniface (Law, 1987b, 2003, p. 131), and subsequently through the works of scholars like Bede and Alcuin (on which see Sections 2 and 3, respectively). Although there is limited evidence for the application of grammatical knowledge to the vernacular during the 7th and 8th centuries, the glossing of Tatwine’s grammar and other Latin texts with the aid of Old English words suggests that early linguistic instruction in England may have been conducted in English (Gneuss, 1990, p. 10; Law, 1977). The compilation of Latin-Old English glosses and glossaries also goes as far back as the 7th-century school of Theodore and Hadrian at Canterbury. Such texts were either arranged by subject for use in teaching, drawing in large part on the Etymologiae of Isidore, or in alphabetical order for reference purposes, in which case they typically consisted of glosses excerpted from glossed texts (Gneuss, 1990, p. 19).
In the last quarter of the 9th century, King Alfred (849–899) encouraged the production of translations of key Christian texts into the vernacular, leading to the emergence of a relatively standardized form of Old English known as Late West Saxon (Law, 2003, p. 193). These efforts to foster Latin learning were further strengthened by the educational and institutional changes promoted by the leaders of the 10th-century Benedictine reform movement. Toward the end of this period, the Anglo-Saxon scholar Aelfric (ca. 950–after 1010), abbot of Eynsham in Oxfordshire, wrote several works that shed light on the use of the vernacular language in a didactic setting. These include a Latin-Old English glossary, an Old English adaptation of the Excerptiones de Prisciano that provides a vernacular description of Latin at a fairly elementary level, and a Colloquium (Latin conversation book) based on the works of Priscian and Donatus, which was intended as a practical manual addressed to schoolboys and was subsequently supplemented with Old English glosses (for an edition and discussion of these works, see Law, 1987a, and Zupitza, 1880). Evidence for the teaching of grammar to students with no native knowledge of Latin during this period is also found in the form of parsing grammars, such as the Beatus quid est, which synthesizes a number of different approaches to grammar in an attempt to create a coherent and useful teaching-text; like Aelfric’s Excerptiones, it demonstrates how the pedagogical strategies of the classroom finally emerged from oral practice into manuscript form (Bayless, 1993).
An Insular Celtic language closely related to Cornish and Welsh, Breton was established in what is now modern Brittany in the early Middle Ages. Most of the extant evidence for Old Breton (the form of the language between the 5th and 11th centuries) survives in the form of glosses, including several on a copy of Priscian’s Institutiones grammaticae in Paris, BN MS Lat. 10290, where they are intermixed with Irish and Welsh glosses (Lambert, 1982; Lemoine, 1986). About 20 glosses in Old Breton are found in a 9th-century copy of Smaragdus’ commentary on the grammar of Donatus (Holtz & Lambert, 1986), and a fragment of the Ars de uerbo by Eutyches in Oxford, Bodleian MS Auct. F. 4. 32 (‘St Dunstan’s Classbook’) contains Old Breton and Welsh glosses that date from the same period. In many cases, these glosses were probably added by scribes working at scriptoria outside of Brittany. After the 11th century, word lists and grammars were also produced in Middle Breton (Le Duc, 1974–1975, 1979, 1980).
5.3 Eastern European Languages
The study of language in Eastern Europe during the medieval period, where many languages already had long-established written traditions, was centered primarily on the translation of the Scriptures into the vernacular and on the production of theological writings and commentaries. The earliest vernacular grammars of the Greek East include a Syriac reworking of the Tekhnē attributed to Dionysius Thrax and a commentary on the same work by a 7th-century Armenian scholar (Law, 2003, pp. 123–124). An anonymous 9th-century text in Old Church Slavonic on the eight parts of speech survives in several Serbian and Russian manuscripts of the 14th to 17th centuries (Jagić, 1968; Worth, 1983).
One of the earliest grammars of a Western European vernacular is the Irish text known as Auraicept na nÉces (‘The Scholars’ Primer’), a compilation of notes on fairly basic grammatical and prosodic concepts, such as letters, syllables, accents, and stylistic faults (Ahlqvist, 1983; Calder, 1995). The text appears to have been intended as a primer on these subjects for the education of the fili, a term conventionally translated as ‘poet’, but in fact designating an individual whose expertise ranged across the disciplines of poetry, law, and history. The core of the Auraicept was probably first written down in the 8th century, but all extant copies include a substantial quantity of later commentary that shows the influence of the etymological method popularized by Isidore of Seville (on which see the discussion in Section 1). The vernacular annotated text has affinities with several Latin commentaries on Donatus’ Ars maior written on the Continent during the 9th century by Irish scholars like Murethach and Sedulius Scottus (Poppe, 2002), but its transmission appears to have been confined to Ireland in the medieval period. One of the most notable characteristics of the Auraicept’s commentary is its markedly defensive stance regarding the status of Irish vis-à-vis the sacred languages of Latin, Greek, and Hebrew: it includes a lengthy pseudohistorical prologue in which it is claimed that Irish was not merely a product of the chaos caused by the dispersal of the languages at Babel, but was in fact constructed by an Irish grammarian from the best parts of all of those languages (Russell, 2005).
In the early 13th century, a further series of grammatical treatises was composed in Irish for the use of poets trained in the Bardic schools of Ireland and the Gaelic parts of Scotland, which existed until around the 17th century (for editions of these tracts, see Bergin, 1916–1955; Mac Cárthaigh, 2014; McKenna, 1979). They include detailed treatments of topics like rhyme, the compounding of words, the government of prepositions, adjectival derivation, nominal declension, abstract nouns, verbs, and metrical faults, and reflect a largely standardized, literary language for itinerant scholars (Ó Cuív, 1973, 1980). In addition to the considerable quantity of grammatical information and technical terminology that the Bardic tracts can offer the linguistic historian, an important feature of these treatises is the sizeable collection of metrical examples, culled from the works of celebrated poets and in many cases not attested elsewhere, that are adduced in them to support various grammatical rules and to illustrate paradigms.
Several grammars of French were written in the French language from the mid-13th century, with the aim of providing a fairly elementary level of instruction to native speakers of the vernacular. These are mainly abridged translations and adaptations of Donatus’ Ars minor, but offer some of the first evidence for the use of French grammatical terminology. More advanced textbooks of French survive from the 15th century onward that are drawn from less obvious sources and deal with various problems of (morpho-)syntax (for editions and discussion of 11 of these early vernacular grammars, see Stadtler, 1988).
The first materials for teaching French to English speakers also date from the mid-13th century, some 200 years after the Norman Conquest (Kibbee, 1991, p. 1). In the period immediately following the Conquest, the vast majority of official business was conducted in Latin, and there appears to have been little need for manuals of French grammar. The first pedagogical texts to appear were mainly orthographical treatises or vocabularies based on everyday needs, and had a clearly practical purpose. In 1409 John Barton produced the Donait françois, one of the earliest attempts to describe French in the Latin model, which was written in dialogue in the format of Donatus’ Ars minor. The work was intended for use in the training of clerks, and it is notable for Barton’s use of trade vocabulary as examples (Kibbee, 1991, p. 191; for an edition, see Swiggers, 1985).
Although systematic grammars of the German language in the main only appeared from the 15th century onwards, linguistic reflection on the vernacular is evidenced from a much earlier period in the form of glosses on a variety of Classical texts. Approximately half of the extant glosses in Old High German (the form of the language up to the 11th century) pertain to the Bible and its exegesis, while a third take the form of dictionaries and glossaries. A small number of Old High German glosses are found on grammatical works, such as Priscian’s Institutiones or the works of Donatus, Eutyches, and Phocas, as well as on later grammar books by authors like Alcuin and the Irishman Clement (Siewert, 1997, pp. 141–142). A version of Donatus’ Ars minor written in Middle High German ca. 1400 also survives (Müller, 1969, pp. 1–7).
Several grammatical descriptions of Occitan (or Provençal), the Romance language spoken along the Mediterranean coast and in Catalonia, were composed from the middle of the 13th century (for an overview, see Schlieben-Lange, 1991, and also Swiggers, 1992). This was partly in response to a perceived need to maintain standards for the highly regarded courtly poetry of the troubadours, which was at its height in the 11th and 12th centuries and spread to Italy following a period of political turmoil in Provence in the early 1200s (Law, 2003, pp. 201–202). For example, a grammarian by the name of Uc Faidit wrote the Donatz proensals in Italy around 1240 for nonnative speakers of the language; his work largely follows the tradition of Donatus and Priscian but occasionally notes structural discrepancies between Latin and Occitan, such as the fact that there were fewer morphological differences in the case system of the latter, which distinguished between only the nominative and accusative singular in some nouns (Marshall, 1969; Robins, 1997, pp. 86 & 107). In the 14th century, the so-called Leys d’amors (‘Law of Love’) was written by the Toulouse lawyer Guilhem Molinier and several colleagues, appearing in several versions between 1332 and 1356 (Gatien-Arnoult, 1841–1843). This treatise offered both a grammar of the language and instruction in the composition of Troubadour-style poetry, and aimed to assist judges of the annual poetry prize established by the Consistoire du Gai Savoir (Law, 2003, pp. 202–203).
5.8 Old Norse
The first grammatical work in Old Icelandic is a 12th-century treatise on orthography known as the First Grammatical Treatise (Benediktsson, 1972; Haugen, 1972). The anonymous author of this text attempts to devise an alphabet appropriate to his native language, using Latin letters where possible but dropping them as necessary and adding a number of new vowel symbols to represent sounds in the vernacular. It also cites minimal pairs as a way of establishing the inventory of distinctive phonemes in the Old Icelandic language. The treatise is preserved in a single manuscript witness, the Codex Wormianus, alongside other grammatical material, such as a work corresponding to the parts of Donatus’ Ars maior that deal with letters and figures of speech. The latter incorporates material on Old Icelandic runes and replaces Donatus’ examples from the Latin poets with ones taken from Icelandic skaldic verse (Law, 2003, pp. 199–200). The Codex Wormianus also includes a copy of the so-called ‘Second Grammatical Treatise’ (Raschellà, 1982), which represents an attempt to establish an orthographic norm for phonological changes that had been affecting the Icelandic language; the tract is notable for its comparison of human speech with sounds produced by a musical instrument.
The Welsh collection of grammatical and prosodic material known as the Gramadegau Penceirddiaid (‘Grammars of the chief bards’) dates from the first half of the 14th century. The author of the earliest version of these grammars, composed in the early 1320s, has been identified as Einion the Priest. A revision of his work, apparently completed by one Dafydd Ddu from Hiraddug in what is now Flintshire, was written soon thereafter at the Cistercian monastery of Valle Crucis in northeast Wales (Hayden & Russell, 2016, pp. 8–9). The various tracts that make up the collection treat subjects like the letters of the alphabet, syllables, diphthongs, the parts of speech, and syntax in Welsh in a manner that is often markedly reliant on Latin sources, sometimes even at the expense of the vernacular. For example, the alphabet presented in the grammars does not reflect either the orthographic perplexities or phonetic characteristics of Middle Welsh; three genders are discussed, despite the fact that by the Middle Welsh period only two were to be found in the vernacular; and virtually all accidence of the highly inflected Welsh verb is ignored (Matonis, 1981). However, the grammars also contain detailed discussions of meter, versification, and poetic principles that pertain specifically to Welsh and bear little correspondence with the Latin tradition. Moreover, they were transmitted alongside a collection of triads concerned with the classification of rules of grammar and poetry and with the ethics of the poetic profession (Russell, 2016, pp. 163–164). The recently identified text designated the Gramadeg Gwysanau, written in northeast Wales during the 14th century, is quite different from the other surviving grammars; it discusses matters like composition, transmission of poetry (orally and in written form), and orthography in a lively matter, and offers advice to pupil poets (Parry Owen, 2010, 2016).
- Kelly, L. G. (2002). The mirror of grammar. Theology, philosophy and the Modistae. Studies in the History of the Language Sciences, 101. Amsterdam/Philadelphia: John Benjamins.
- Ahlqvist, A. (Ed.). (1983). The early Irish linguist. An edition of the canonical part of Auraicept na néces. Commentationes Humanarum Litterarum, 73. Helsinki: Societas Scientiarum Fennica.
- Amsler, M. (1989). Etymology and grammatical discourse in late antiquity and the early Middle Ages. Studies in the History of the Language Sciences, 44. Amsterdam/Philadelphia: John Benjamins.
- Bayless, M. (1993). Beatus quid est and the study of grammar in late Anglo-Saxon England. In V. Law (Ed.), History of linguistic thought in the early Middle Ages (pp. 67–110). Studies in the History of the Language Sciences, 71. Amsterdam/Philadelphia: John Benjamins.
- Benediktsson, H. (Ed.). (1972). The first grammatical treatise: Introduction, text, notes, translation, vocabulary, facsimiles. University of Iceland Publications in Linguistics, 1. Reykjavik: Institute of Nordic Linguistics.
- Bergin, O. (Ed.). (1916–1955). Irish grammatical tracts. Supplements to Ériu 8 (1916), 9 (1921–1923), 10 (1926–1928), 14 (1946), 17 (1955).
- Bracken, D. (2002). Virgilius Grammaticus and the earliest Hiberno-Latin literature. In M. Richter & J.-M. Picard (Eds.), Ogma: Essays in Celtic studies in honour of Próinséas Ní Chatháin (pp. 251–261). Dublin: Four Courts Press.
- Bursill-Hall, G. L. (Ed. and Trans.). (1972). Thomas of Erfurt: Grammatica speculativa. The Classics of Linguistics, 1. London: Longmans.
- Calder, G. (Ed.). (1995). Auraicept na n-éces. The scholars’ primer. Dublin: Four Courts Press. (Original work published 1917.)
- Contreni, J. J. (2014). Learning for God: Education in the Carolingian age. Journal of Medieval Latin, 24, 89–129.
- Copeland, R., & Sluiter, I. (Eds.). (2009). Medieval grammar & rhetoric. Language arts and literary theory, AD 300–1475. Oxford: Oxford University Press.
- Gatien-Arnoult, A. F. (Ed.). (1841–1843). Las Flors del Gay Saber, estier dichas Las Leys d’Amors (Vols. 1–3). Toulouse: Paya.
- Gneuss, H. (1990). The study of language in Anglo-Saxon England. Bulletin of the John Rylands University Library of Manchester, 72, 3–32.
- Grondeux, A. (2000). La grammatica positiva dans le Bas Moyen-Âge. In S. Auroux, E. F. K. Koerner, H.-J. Niederhe, & K. Versteegh (Eds.), History of the Language Sciences/Geschichte der Sprachwissenschaften/Histoire des sciences du langage: An international handbook on the evolution of the study of language from the beginnings to the present (pp. 598–610). Handbooks of Linguistics and Communication Science, 18/3. Berlin and New York: Walter de Gruyter.
- Haugen, E. (Ed.). (1972). First grammatical treatise: The earliest Germanic phonology. An edition, translation, and commentary (2d ed.). London: Longman.
- Hayden, D., & Russell, P. (2016). Editors’ introduction. In D. Hayden & P. Russell (Eds.), Grammatica, gramadach and gramadeg. Vernacular grammar and grammarians in medieval Ireland and Wales (pp. 1–10). Studies in the History of the Language Sciences, 125. Amsterdam/Philadelphia: John Benjamins.
- Herren, M. (1979). Some new light on the life of Virgilius Maro Grammaticus. Proceedings of the Royal Irish Academy, 79(C), 27–71.
- Holtz, L. (1981). Donat et la tradition de l’enseignement grammatical: Étude sur l’Ars Donati et sa diffusion (IXe – IXe siècle) et édition critique. Paris: Centre national de la recherche scientifique.
- Holtz, L., & Lambert, P.-Y. (1986). La tradition ancienne du Liber in partibus Donati de Smaragde de Saint-Mihiel [avec une note additionnelle de Pierre-Yves Lambert: Les gloses en vieux-breton à la grammaire de Smaragde]. Revue d’Histoire des Textes, 16, 171–211.
- Irvine, M. (1994). The making of textual culture. ‘Grammatica’ and literary theory 350–1100. Cambridge Studies in Medieval Literature, 19. Cambridge, U.K.: Cambridge University Press.
- Jagić, V. (Ed.). (1968). Codex Slovenicus rerum grammaticarum. Slavische Propyläen, 25. Munich: Fink. (Original work published 1896.)
- Jeudy, C. (1972). L’Institutio de nomine, pronomine et verbo de Priscian. Manuscrits et commentaires médiévaux. Revue d’Histoire des Textes, 2, 73–144.
- Keil, H., Hertz, M., Mommsen, T., & Hagen, H. (Eds.). (2009). Grammatici Latini (Vols. 1–8). Cambridge, U.K.: Cambridge University Press. (Original work published 1855–1880.)
- Kibbee, D. A. (1991). For to speke Frenche trewely: The French language in England, 1000–1600: Its status, description and instruction. Studies in the History of the Language Sciences, 60. Amsterdam/John Benjamins.
- King, M. H. (1979). Grammatica mystica, a study of Bede’s grammatical curriculum. In M. H. King & W. M. Stevens (Eds.), Saints, scholars and heroes: Studies in medieval culture in honour of Charles W. Jones (Vol. I, pp. 145–159). Collegeville, MN: Monastic Manuscript Library.
- Lambert, P.-Y. (1982). Les gloses du manuscript BN lat. 10290. Études Celtiques, 19, 173–213.
- Law, V. (1977). The Latin and Old English glosses on the ars Tatuini. Anglo-Saxon England, 6, 77–89.
- Law, V. (1987a). Anglo-Saxon England: Ælfric’s Excerptiones de arte grammatica Anglice. Histoire Épistémologie Langage, 9, 47–71.
- Law, V. (1987b). The Insular Latin grammarians. Studies in Celtic History, 3. Woodbridge, U.K.: Boydell Press.
- Law, V. (1994). The study of grammar. In R. McKitterick (Ed.), Carolingian culture: Emulation and innovation (pp. 88–110). Cambridge, U.K.: Cambridge University Press.
- Law, V. (1995). Wisdom, authority and grammar in the seventh century: Decoding Virgilius Maro Grammaticus. Cambridge, U.K.: Cambridge University Press.
- Law, V. (2003). The history of linguistics in Europe. From Plato to 1600. Cambridge, U.K.: Cambridge University Press.
- Le Duc, G. (1974–1975). Le Donoet, grammaire latine en moyen-breton. Études Celtiques, 14, 525–565.
- Le Duc, G. (1979). Le Donoet, grammaire latine en moyen-breton. Études Celtiques, 16, 237–259.
- Le Duc, G. (1980). Les gloses en moyen-breton du Liber uocabulorum. Études Celtiques, 17, 229–255.
- Lemoine, L. (1986). Les methods d’enseignement dans la Bretagne du haut Moyen Age d’après les manuscrits Bretons: L’exemple du Paris, B. N., Lat. 10290. In Landévennec et le monachisme Breton dans le haut Moyen Age (pp. 45–63). Landévennec: Association Landévennec.
- Löfstedt, B. (Ed.). (1977). Sedulius Scottus. Grammatici Hibernici Carolini aevi IV: In Donati artem minorem. In Priscianum. In Eutychum. Corpus Christianorum Continuatio Mediaevalis, 40C. Turnhout: Brepols.
- Löfstedt, B., Holtz, L., & Kibre, A. (Eds.). (1986). Smaragdus, Liber in partibus Donati. Corpus Christianorum, Continuatio Mediaevalis, 68. Turnhout: Brepols.
- Luhtala, A. (2000). Early medieval commentary on Priscian’s Institutiones grammaticae. Cahiers de l’Institut du Moyen Âge Grec et Latin, 71, 115–188.
- Luhtala, A. (2013). Pedagogical grammars before the eighteenth century. In K. Allan (Ed.), The Oxford handbook of the history of linguistics (pp. 341–358). Oxford: Oxford University Press.
- Mac Cárthaigh, E. (2014). The Art of bardic poetry. A new edition of Irish grammatical tracts I. Dublin: Dublin Institute for Advanced Studies.
- Marenbon, J. (1994). Carolingian thought. In R. McKitterick (Ed.), Carolingian culture: Emulation and innovation (pp. 171–192). Cambridge, U.K.: Cambridge University Press.
- Marshall, J. H. (Ed.). (1969). The Donatz proensals of Uc Faidit. London: Oxford University Press.
- Martin, L. T. (1984). Bede as a linguistic scholar. American Benedictine Review, 35, 204–217.
- Matonis, A. T. E. (1981). The Welsh bardic grammars and western grammatical tradition. Modern Philology, 79, 121–145.
- McKenna, L. (Ed.). (1979). Bardic syntactical tracts. Dublin: Dublin Institute for Advanced Studies. (Original work published 1944.)
- Migne, J.-P. (1844–1865). Patrologiae cursus completus. Series latina (Vols. 1–217). Paris: Garnier Fraters.
- Müller, J. (Ed.). (1969). Quellenschriften und Geschichte des deutschsprachlichen Unterrichtes bis zur Mitte des 16. Jahrhunderts. Hildesheim: Olms. (Original work published 1882.)
- Murphy, J. J. (Ed.). (1971). Three medieval rhetorical arts. Berkeley: University of California Press.
- Murphy, J. J. (1974). Rhetoric in the Middle Ages. A history of rhetorical theory from Saint Augustine to the Renaissance. Berkeley: University of California Press.
- Ó Cuív, B. (1973). The linguistic training of the medieval Irish poet. Celtica, 10, 114–140.
- Ó Cuív, B. (1980). A medieval exercise in language planning: Classical early modern Irish. In E. F. K. Koerner (Ed.), Progress in linguistic historiography: Papers from the international conference on the history of the language sciences, Ottawa, August 28–31, 1978 (pp. 23–34). Studies in the History of the Language Sciences, 20. Amsterdam: John Benjamins.
- Parry Owen, A. (2010). Gramadeg Gwysanau (Archify Sir y Fflint, D/DG 2082). Llên Cymru, 33, 1–31.
- Parry Owen, A. (2016). Gramadeg Gwysanau. A fragment of a fourteenth-century Welsh bardic grammar. In D. Hayden & P. Russell (Eds.), Grammatica, gramadach and gramadeg. Vernacular grammar and grammarians in medieval Ireland and Wales (pp. 181–200). Studies in the History of the Language Sciences, 125. Amsterdam/Philadelphia: John Benjamins.
- Passalacqua, M. (1978). I codici di Prisciano. Sussidi eruditi, 29. Rome: Edizioni di Storia e Letteratura.
- Passalaqua, M. (1993). Priscian’s Institutio de nomine et pronomine et verbo in the ninth century. In V. Law (Ed.), History of linguistic thought in the early Middle Ages (pp. 193–204). Studies in the History of the Language Sciences, 71. Amsterdam/Philadelphia: John Benjamins.
- Poppe, E. (2002). The Latin quotations in Auraicept na nÉces: Microtexts and their transmission. In P. Ní Chatháin & M. Richter (Eds.), Ireland and Europe in the early middle ages: Texts and transmission/Irland und Europa im früheren Mittelalter: Texte und Überlieferung (pp. 296–312). Dublin: Four Courts.
- Raschellà, F. D. (Ed.). (1982). The so-called Second Grammatical Treatise: An orthographic pattern of late thirteenth-century Icelandic. Edition, translation, and commentary. Filologia Germanica: Testi i Studi, 2. Florence: Felice le Monnier.
- Robins, R. H. (1997). A short history of linguistics (4th ed.). London & New York: Longman.
- Russell, P. (2005). “What was best of every language”: The early history of the Irish language. In D. Ó Cróinín (Ed.), A new history of Ireland: Prehistoric and early Ireland (pp. 405–450). Oxford: Oxford University Press.
- Russell, P. (2016). Teaching between the lines: Grammar and grammatica in the classroom in early medieval Wales. In D. Hayden & P. Russell (Eds.), Grammatica, gramadach and gramadeg. Vernacular grammar and grammarians in medieval Ireland and Wales (pp. 133–148). Studies in the History of the Language Sciences, 125. Amsterdam/Philadelphia: John Benjamins.
- Schlieben-Lange, B. (1991). Okzitanisch: Grammtikographie und Lexikographie. In G. Holtus, M. Metzeltin, & C. Schmitt (Eds.), Lexikon der Romanistichen Linguistik, 5.2. Tübingen, Germany: Niemeyer.
- Siewert, K. (1997). Vernacular glosses and classical authors. In N. Mann & B. M. Olsen (Eds.), Medieval and Renaissance scholarship. Proceedings of the second European Science Foundation workshop on the classical tradition in the middle ages and the Renaissance (pp. 137–152). New York: Brill.
- Stadtler, T. (Ed.). (1988). Zu den Anfängen der französischen Grammatiksprache. Beihefte zur Zeitschrift für romanische Philologie, 223. Tübingen, Germany: Niemeyer.
- Stokes, W., & Strachan, J. (Eds.). (1975). Thesaurus Palaeohibernicus: A collection of Old-Irish glosses scholia prose and verse. Vol. II: Non-Biblical glosses and scholia. Old Irish prose. Names of persons and places. Inscriptions. Verse. Indexes. Oxford: Oxford University Press. (Original work published 1903.)
- Swiggers, P. (Ed.). (1985). Le Donait François: la plus ancienne grammaire du français. Édition avec introduction. Revue des Langues Romanes, 89, 235–251.
- Swiggers, P. (1992). Les plus anciennes grammaires occitanes: Tradition, variation et insertion culturelle. In G. Gouiran (Ed.), Contacts de langues, de civilisations et intertextualité. IIIème congrès international de l’Association Internationale d’Études Occitanes, Montpellier, 20–26 Septembre 1990 (Vol. I, pp. 131–148). Montpellier: Centre d’Études Occitanes.
- Swiggers, P., & Vanvolsem, S. (1987). Les premières grammaires vernaculaires de l’italien, de l’espagnol et du portugais. Histoire Épistémologie Langage, 9(1), 157–181.
- Worth, D. (1983). The origins of Russian grammar. Notes on the state of Russian philology before the advent of printed grammars. Columbus, OH: Slavica.
- Zupitza, J. (Ed.). (1880). Aelfrics Grammatik und Glossar. Sammlung englischer Denkmaler, 1. Berlin: Weidmann.