Morphological relatedness and the grammaticalization of Old English -bora.
In Kastovsky's words (1992: 294) "much of the OE vocabulary is derivationally related by productive word-formation patterns" including zero-derivation, compounding and affixation. Of these widespread and deep-rooted derivational processes, this journal article focuses on affixation and, to be more specific, on the formation of nouns by suffixation. The research question that is addressed in this work concerns the status of the derivational element -bora. That is to say, whether derivation with -bora constitutes an instance of suffixation or, contrariwise, involves compounding. The reason for posing this question is that the suffix bora is a verbal element, morphologically related to the verb beran 'bear' and, as such, Kastovsky (1992) does not consider it a suffix, whilst Quirk and Wrenn (1994) do. In order to answer this question, this article analyzes the Old English nominal suffix -bora and compares it to the the suffixes -a, -e, -en, -end., -ere/-re, -icge, -estre/-istre/-ystre, -o and -u. The analysis shows that the segment at stake is no longer a free lexeme but rather a bound form. Unlike previous approaches, such as Quirk and Wrenn's (1994), evidence is gathered to hold that a process of grammaticalization has taken place that motivates the insertion of -bora into the inventory of derivational morphemes. Therefore, this paper contributes to the research project in Old English word-formation and the structure of the lexicon represented by Kastovsky (1986, 1989, 1990, 1992, 2005, 2006), who deals with the typological shift from invariable base morphology to variable base morphology that takes place in Old English; and Martin Arista (2008, 2009, 2010a, 2010b, 2010c, 2011a, 2011b, 2011c, fc.-a, fc.-b, fc.-c), who focuses on the lexical layers and derivational processes of Old English. This work also contributes to the study in the rise of analytic tendencies in English word-formation at the expense of synthetic tendencies (Haselow 2011) or, put in other words, it focuses on the coexistence of zero derivation and affixation, by means of, respectively, -a, -e, -o, -u, and -en, -end, -ere/-re, -icge, -estre/-istre/-ystre.
The data of analysis have been retrieved from the lexical database of Old English Nerthus (www.nerthusproject.com). The outline of the article is as follows. Section 2 presents the main units and patterns found in Old English noun suffixation and draws a distinction between explicit and implicit morphological relatedness. Section 3 lays the theoretical foundations of the work, discusses the relevant aspects of Lexeme-Morpheme Base Morphology (Beard 1995; Beard and Volpe 2005) and applies them to Old English word-formation. Then, section 4 discusses the grammaticalization bora > -bora. Finally, the conclusions of the article are summarised in section 5.
2. The suffixation of nouns in Old English: explicit and implicit relatedness
Kastovsky (1992) provides a list of the main nominal suffixes that comprises -d/-t/-o, -dom, -ele(e)/-l(a)/-ol, -els, -en, -end, -ere, -estre, -et(t), -had, -incel, -ing, -lac, -ling, -ness, -r&den, -scipe, -d(o)/-t, -ung/-ing, -wist. For Mitchell (1992) the most remarkable Old English nominal suffixes are -ad/-od, -end, -had, -ing, -m&l, -r&den, -d(o)/-d(u) and -ung/-ing. Quirk and Wrenn (1994) offer the largest inventory of nominal suffixes and draw a distinction between the most frequent (-dom, -end, -nes(s)/-nis/-nys, -scipe and -ing/-ung) and the least frequent (-bora,-el/-ol/-ul, -els, -en, -ere, -estre, -et(t), -had, -ing, -lac, -ling, -od/-ad, -r&den, -d(o)/-d(u)). Among other differences, Kastovsky (1992) describes -bora as a free form in compounding whereas Quirk and Wrenn (1994) consider it a bound form. This question is addressed in the next section. Meanwhile, the functions and properties of distribution of the other suffixes are presented briefly.
The suffixes in the series -d/-t/-d create deverbal nouns, as is the case with aebylgd 'anger' and haeld 'health'. The suffix -dom forms denominal and deadjectival abstract nouns with the meaning 'state, condition, fact of being, action of. Denominal nouns include caserdom 'empire', martyrdom 'martyrdom', campdom 'contest', and laecedom 'medicine'. Deadjectival nouns, among others, are freodom 'freedom', haligdom 'holiness, sanctuary', wisdom 'wisdom'. The group of suffixes -ele(e)/-l(a)/-ol/-ul are attached to action nouns, as in scendle 'reproach', dreal 'reproof and hwyrfel 'circuit, whirlpool'; agent nouns, as is the case with aeftergengel 'successor', bydel 'herald' and bccslitol 'backbiter'; object/result nouns (scytel 'dart, missile', fyndel 'invention' and bitol 'bridle'); instrumental nouns like sceacel 'shackle', tredel 'sole of the foot' and spinel 'spindle'; and locative nouns such as smygel 'burrow, retreat', stigel 'stile' and setl 'seat'. The suffix -els forms concrete masculine deverbal nouns from strong and weak verbs, as in raedels 'counsel', braedels 'carpet' and gyrdels 'girdle'. The suffix -en forms feminine nouns of action (sien 'sight', fillen 'falling', swefen 'sleep, dream'), object/result (raedenn 'reckoning, estimation', sellen 'gift', fcesten 'fortress'), instrument (hlaeden 'bucket', lifen 'sustenance', faesten 'fastener') and locative nouns (hengen 'rack, cross', byrgen 'grave'). The suffix -end forms deverbal agent nouns from both weak and strong verbs. The agent nouns are masculine, whereas the action nouns display the feminine gender. Masculine agent nouns include biddend 'petitioner', laerend 'teacher' and dclnimend 'participle', while object nouns include belifend 'survivor' and gehaeftend 'prisoner'. The suffix -ere forms nouns from other nouns and from verbs. Examples of deverbal nouns include leornere 'disciple' (agent), sceawere 'mirrow' (object), punere 'pestle' (instrumental), wordsamnere 'catalogue' (locative), dirnegeligere 'sailor' (action), etc. Denominal nouns form agent nouns like scipere 'sailor', scohere 'shoemaker' and scdere 'sower'. The suffix -estre forms deverbal and denominal feminine agent nouns. Deverbal nouns include hleapestre 'female dancer', wcscestre 'washer' and taeppestre 'female tavern-keeper'. Denominal nouns are byrdestre 'female carrier', fidestre 'female fiddler' and lybbestre 'sorceress'. The suffix -et(t) forms deverbal and denominal neuter nouns. Deverbal nouns include rewett 'rowing', hiwett 'hewing' and baernett 'burning', while diccett 'thicket', and rymet 'space, extent' qualify as denominal nouns. The suffix -had conveys the meaning of 'state, rank, order, condition, character' in instances like abbudhad 'rank of an abbot', camphad 'warfare' and cildhad 'childhood'. The suffix -incel forms neuter denominal diminutives such as bogincel 'small bough', busincel 'little house' and scipincel 'little ship'. The suffix -ing forms masculine nouns denoting 'proceeding or derived from' from nouns (wicing 'pirate'), adjectives (ierming 'poor wretch') and verbs (fostring 'fosterchild'). The suffix -lac forms masculine abstract nouns from nouns and verbs and denotes 'state, act, quality, nature of from nouns and verbs. Denominal nouns include bodlac 'decree', brydlc 'marriage, marriage gift' and lyblac 'witchcraft', while breowlac 'brewing' qualifies as a deverbal noun. The suffix -ling derives nouns from adjectives, nouns and verbs. Deadjectival nouns are deorling 'favourite' and geongling 'youth'; denominal nouns include cnaepling 'youth', fostorling 'fosterchild' and deowling 'slave'; hyrling 'hireling', rcepling 'prisoner' and hwirfling 'that which turns' are deverbal nouns. The suffix -ness and its variant forms -nis, -nes and -nys derive feminine abstract nouns from adjectives and verbs. Deadjectival nouns include cedelness 'nobility', beorhtness 'brightness' and biterness 'bitterness', clcnness 'purity'. Among deverbal nouns we find blinness 'cessation', brecness 'breach' and costness 'temptation'. The suffix -od/-ad forms masculine nouns, mainly abstract, as is the case with drohtod 'way of life', hergad 'plundering' and langad 'longing'. The suffix -roden derives feminine denominal nouns with the meaning 'state, act, condition', as in bebodrceden 'command, authority', brodorrceden 'fellowship, brotherhood' and camprcden 'war, warfare'. The suffix -scipe forms masculine abstract nouns from adjectives and nouns with the meaning 'state, act, fact, condition'. Denominal nouns include bodscipe 'message', freondscipe 'friendship' and leodscipe 'nation, people', while gecorenscipe 'election, excellence', unwcrscipe 'carelessness' and hwctscipe 'activity, vigour' are deadjectival nouns. The suffix -ung/-ing forms deverbal nouns from both strong and weak verbs. Action nouns include binding 'binding' and huntung 'hunting'. instances of agent nouns include gaderung 'gathering, assembly' and gemeting 'meeting, assembly'. Among object/result nouns we find beorning 'incense' and agnung 'possessions'. instrumental nouns include instances such as lacnung 'medicine' and wering 'dam'. cyping 'market' and wunung 'dwelling' qualify as locative nouns. Finally, the suffix -wist derives feminine abstract nouns from nouns (huswist 'household'), adjectives (loswist 'loss') and adverbs (midwist 'presence').
Along with these suffixes, which bear an explicit derivational relationship to the base of derivation because the derivational segment is clearly distinguishable, there are other suffixes that bear an implicit derivational relationship since the same segment expresses the derivational as well as the inflectional function. Gonzalez Torres (2010) holds that there is continuity between inflection and derivation in Old English morphology because some endings function simultaneously as markers of derivation and inflection, thus -a in andettan 'confess' > andetta 'one who confesses', -e in hierdan 'protect' > hierde 'keeper', -o in fullian 'fill up' > fyllo 'fillness' and -u in giefan 'give' > giefu 'gift.' While concurring with Gonzalez Torres (2010) on the difficulty of drawing a clear-cut distinction between inflection and derivation in a language with variable base morphology such as Old English (Kastovsky 2006), i discuss these suffixes in the wider context of zero derivation, mainly from strong verbs. Such a discussion not only contextualizes the phenomenon but also allows us to introduce the sentential counterparts of the derivations at stake.
Zero derivation is derivation without derivational morphemes. It can be broken down into consonantal and vocalic zero derivation. By applying this distinction, the list of zero derivatives compiled by Pilch (1970) can be divided as follows. Within consonantal zero derivation we find nouns like rad 'ride, riding, expedition' (< ridan 'ride' str. I), gielp 'boasting, pride, arrogance' (< gielpan 'boast, exult' str. III), baer 'bier' (< beran 'bear' str. IV), aet 'eatables, food' < etan 'eat' str. V), fcr 'calamity, sudden danger' (<faran 'set forth; happen, exist, act' str. VI); denominal weak verbs of class 1 with i-umlaut or gemination such as andswerian 'to answer' (< andswaru 'answer'), daeman 'to judge' (< dom 'doom, judgement'), fyllan 'to fill' (< full 'full'), fodan 'to feed' (< foda 'food'), feran 'to go' (< far 'journey'), ferian 'to carry' (< fcr 'movement'), gremman 'to enrage' (< gram 'angry') and holan 'to heal' (< hal 'health'); denominal weak verbs of class 2 without i-umlaut or consonant gemination of the type andswarian 'to answer' (< andswaru 'answer'), lufian 'to love' (< lufu 'love'), geomrian 'to be sad' (< geomor 'sad'), hean (<*heahian 'to raise' < heah 'high'), smeagan 'to think' (< smeah 'sagacious'), tweogan 'to doubt' (< tweo 'doubt'); and causative deverbal weak verbs of class 1 derived from strong verbs like afliegan 'expel' (< flean 'flee'), drencan 'give to drink' (< drincan 'drink'), roran 'raise' (< risan 'rise'), sengan 'sing' (< singan 'sing'), settan 'make sit' (< sittan 'sit') and swebban 'put to sleep' (< swefan 'sleep'). Within vocalic zero derivation we find nouns as well as adjectives. Beginning with nouns, there are instances of derivation from strong verbs such as lyre 'loss' (< leosan 'lose' str. II) and wealda 'ruler' (< wealdan 'rule' str. VII). As for adjectives, among those based on strong verbs we find bryce 'fragile' (< brecan 'to break' str. IV), eadfynde 'easy to find' (<fndan 'to find' str. III), gefere 'accesible' (< faran 'to set forth' str. VI), lyge 'lying, false' (< leogan 'to lie' str. II), oncnowe 'known, recognised' (< cnawan 'to know' str. VII), swice 'fallacious, deceitful' (< swican 'to wander' str. I) and ungemete 'huge' (< metan 'to measure' str. V). These derivatives evidence that the category change associated with zero derivation causes a shift of inflectional paradigm that, in the case of nouns and adjectives, usually displays a vowel in the reference form, in such a way that that vowel bears the mark of inflection and derivation. The case of verbs is similar, although all of them belong in the consonantal type of zero derivation given the form of the infinitive.
In order to further discuss these types it is necessary to introduce the relevant aspects of Lexeme-Morpheme Base Morphology. This is done in the next section.
3. Theoretical aspects: Lexeme-Morpheme Base Morphology
The theoretical framework chosen for this study is Lexeme-Morpheme Base Morphology, as proposed by Beard (1995) and Beard and Volpe (2005). The advantage of this framework is that it decomposes a complex concept such as derivational relationship into simpler categorial and functional categories and, moreover, because it provides a unified inventory of derivational and inflectional functions compatible with explicit and implicit morphological relatedness.
Lexeme-Morpheme Base Morphology is known for its strict distinction between lexemes and grammatical morphemes. Morpheme-based morphology assumes that language contains only one type of meaningful unit, the morpheme, which includes stems and affixes, all of which are signs. Lexeme-based morphology, on the contrary, assumes that only lexemes, derived or underived, are signs, and that affixes, reduplication, re-vowelling, metathesis, subtraction, stem mutation, and the like, are means of phonologically marking independent derivational operations which a lexeme might have undergone. This means that lexemes refer to something in the real world, whereas morphemes refer exclusively to universally available closed class grammatical categories (such as Tense, Aspect, and Number) and may consist of independent phonemic strings, affixes, infixes, changes in accent or tone, or even predictable omissions (zero morphemes).
The basic idea, therefore, is that the lexicon contains exclusively noun, verb and adjective stems, whereas grammatical morphemes are the output of phonological operations independent of the semantic operations they realize. In this framework, affixation is reduced to an exclusively phonological operation. This is called the Separation Hypothesis. The Separation Hypothesis splits derivation, both lexical and inflectional, into three processes: lexical (L-) derivation, inflectional derivation, and morphological spelling. Derivation comprises operations on abstract lexical and inflectional category functions such as [+Plural, -Singular], [+Past, -Present], [+1st], and the like. Spelling is the purely phonological realization of the morphological categories of any base lexeme that has undergone such derivation. Its function is to distinguish stems that have undergone derivation from those which have not. If the derivation is inflectional, the marker may be attached to the lexical stem or assigned independently to a structural position in syntax in ways which syntax alone cannot predict. Lexical derivation takes place in the lexicon and inflectional derivation in the syntax. Beard (1995) distinguishes four kinds of lexical derivation: transposition, functional derivation, feature switches and expressive derivations. Transpositions change the lexical category of a lexeme. Functional derivations add a semantically interpretable category function, such as Subject, Object, Locus and Manner. Lexical switches change the value of inherent lexical features, such as Gender and expressive derivations comprise the Augmentative and Diminutive and reflect the attitude of the speaker. The base rule component of the theory cannot be syntactic only but must accommodate both lexical operations (derivations) and high-level syntactic operations (inflections). The types of lexical derivation rules that are available to grammars, therefore, are determined by the categories of the base rule component and the lexicon. This is called the Base Rule Hypothesis.
The Universal Grammatical Function Theory stipulates that the functions of inflectional and lexical derivation are the same.
Given this overview of the theory, instances of implicit morphological relatedness such as ridan 'to ride' ~ ridda 'rider' imply three types of lexical derivation: a transposition whose input is a verb and whose output is a noun, a functional derivation that assigns the subjective role, and a featural switch. These three types of lexical derivation are illustrated, respectively by figures 1-3, based on Beard (1995, 2005) [where NP stands for Noun Phrase, C for Complementiser, CP for Complementiser Phrase, IP for Inflectional Phrase and VP for Verb Phrase; the basic parallel is with a sentence, in which IP contains a word level category such as will, must, etc. expressing verbal inflection and the Complementiser such as that introduces clausal complements; notice in this respect that the Old English relative is not analyzed as a pronoun but as a conjunction].
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
These tree diagrams ultimately indicate that Old English has two ways of expressing the same meaning by using an identical lexeme: one is lexical (ridda 'rider'), the other is syntactic (sum oe rided 'one who rides'). The same functions are found in both expressions: there is a subjective function and an unexpressed objective function. As it is shown in the next section, however, the objective function can be overt.
4. The grammaticalization bora > -bora
This section argues for the existence of the derivational suffix -bora as a result of the grammaticalization of the free form bora 'bearer'. The grammaticalization bora > -bora, in terms of the Lexeme-Morpheme base Morphology framework, is based on a feature switch as depicted by Figure 3 and leading to a functional derivation like the one in Figure 2 (both requiring a transposition as described by Figure 1). The feature switch displayed by Figure 4 represents the objective function required by the transitive beran 'bear', in contradistinction to the intransitive ridan 'ride'.
[FIGURE 4 OMITTED]
This said, let us consider this evolution in the wider setting of the noun-deriving suffixes of Old English. The lexical database of Old English Nerthus turns out 480 nouns derived from verbal bases by means of -bora and the following affixes: -a, -e, -en, -end, -ere, -estre, -icge, -o and -u, which can be broken down as follows:
a. -bora (23) b. -a(90), -e (11), -en (14), -end (192), erel-re (96), -estre/-istre/- ystre (12), -icge(2), -o(2), -u(33)
With the exception of complex nouns with -bora, the derivatives tabulated in (1) have a verbal base of derivation, which allows the researcher to determine whether a subjective or an objective relationship holds between base and derivative. This point is illustrated by (2), which displays instances with all the affixes in (1b) along with their verbal bases of derivation:
(2) (ge)spreca 'spokesman' ~ (ge)sprecan 'to speak' syde 'a decoction' ~ (ge)seodan 'to boil' byrgen 'burying place' ~ (ge)byrgan 'to bury' unrihthomend 'adulterer' ~ unrihthoman 'to commit adultery' halsere 'soothsayer' ~ halsian 'to adjure' hoppestre 'female dancer' ~ hoppian 'to dance' acennicge 'mother' ~ acennan 'to bring forth' gehlytto 'lot' ~ gehleotan 'to cast lots' sacu 'reproof; affliction; persecution' ~ sacan 'to struggle'
The derivatives in (2) can be classified, on the grounds of the distinction drawn between fully derivational and non derivational affixation, into a group comprising instances like byrgen 'burying place', unrihthomend 'adulterer', halsere 'soothsayer', hoppestre 'female dancer' and acennicge 'mother'; and another one consisting of (ge)spreca 'spokesman', syde 'a decoction', gehlytto 'lot' and sacu 'persecution'. In these groups, lexical switches produce pairs like those in (3). Notice that m stands for masculine, f for feminine and n for neuter:
a. acennend m 'parent' acennicge f 'mother' b. aoswara m 'oath-swearing, oath' aoswaru f 'oath-swearing, oath' c. byroestre f 'female carrier' byrore m 'bearer, supporter' d. cennend m 'parent' cennestre f 'mother' e. forspennend m 'procurer' forspennestre f 'procuress' f. fylgend m 'follower, observer' fylgestre f 'female follower' g. galdre m 'wizard, magician' galdricge f 'enchantress' h. holend m 'Saviour, Christ' holestre f 'saviour' i. hleapere m 'runner, courier; wanderer, leaper, dancer' hleapestre f 'female dancer' j. leornere m 'learner, disciple; scholar; reader' leornestre f 'a student' k. oferswioend m 'vanquisher' oferswioestre f 'victrix' l. plegere m 'player' plegestre f 'female athlete'
To focus on functionally motivated lexical derivations, example (4) analyses the available evidence as to pairs or triplets of derivatives that share a base of derivation in such a way that the attachment of an affix performs a different function:
a. andetla m 'declaration, confession' (objective) andetta m 'one who confesses' (subjective) andettere m 'one who confesses' (subjective) b. byrgen f 'burying place' (objective) byrgend m 'grave-digger' (subjective) c. foreoingere m 'intercessor, mediator' (subjective) foreoingiend m 'intercessor' (subjective) foreoingroden f 'intercession' (objective) d. gehlyta m 'companion' (subjective) gehlytta m 'partner, fellow' (subjective) gehlytto ? 'fellowship, lot' (objective) e. (ge)reora m 'rower, sailor' (subjective) (ge)reoru np 'oars' (objective) f. (ge)saca m 'opponent, foe' (subjective) (ge)sacu f 'conflict, strife, war' (objective) g. giefa m 'donor' (subjective) giefend m 'giver' (subjective) giefu f 'giving, gift' (objective) h. gripa m 'handful, sheaf' (objective) gripu f 'kettle, caldron' (subjective) i. manswara m 'perjurer' (subjective) manswaru f 'perjury' (objective) j. nydnima m 'one who takes by force' (subjective) nydnimend f 'rapine' (objective) nydnimu f 'rapine, forcible seizure' (objective) k. sceaoa m 'injurious person' (subjective) sceaou f 'injury' (objective) l. selen f 'grant, gift; tribute' (objective) sellend m 'giver; betrayer' (subjective) m. slaga m 'slayer, homicide' (subjective) slea f 'slay, weaver's reed' (objective) n. unna m 'favour, approval; grant' (objective) unnend m 'one who grants' (subjective) o. wioercwida m 'contradicter' (subjective) wioercwide m 'contradiction' (objective)
In order to pursue the question of functional derivations that add semantically interpretable functions such as the subjective or the objective, 480 suffixed nouns have been analysed, out of which 391 are subjective and 89 objective. Therefore, the subjective function is clearly favoured. The suffixes analysed can be divided into three groups on functional grounds: those suffixes that always perform the same function, those suffixes that practically always realize the same function and those suffixes for which no predominant function can be identified. These groups are given in (5):
a. -bora (21 subjective), -estrel-istrel-ystre (18 subjective), -icge (2 subjective), -o (2 objective) b. -e (8 subjective, 3 objective), -en (13 objective, 1 subjective), -end(190 subjective, 2 objective), -ere/-re (95 subjective, 1 objective), -u (29 objective, 4 subjective) c. -a (58 subjective, 32 objective)
As can be seen in (5), the suffix -a is selected for the subjective and the objective functions. The suffix -e is selected for the objective function mainly. The suffix -en is selected for the objective function mainly. The suffix -end is clearly subjective. The situation with the suffix ere/-re is comparable. It is overwhelmingly subjective, although there is an instance of the objective function. The suffix -estre/-istre/-ystre is subjective only. The suffix -icge is exclusively subjective. The suffix -o is objective only. Finally, the suffix -u is clearly objective, although there are four instances that can be considered subjective. These results are in accordance with the Universal Grammatical Function Theory, which predicts that the functions of inflectional and lexical derivation are the same. Indeed, suffixes involved in explicit derivational relations such as -estre perform the same function, namely subjective, as other suffixes partaking in implicit derivational relations, such as -a. The same applies to the objective function. Suffixes taking part in explicit derivational relations such as -en perform the subjective function, as other suffixes involved in implicit derivational relations, like -o, do. Finally, the fact that most of the suffixes under scrutiny perform the subjective and the objective function is in keeping with the Separation Hypothesis, in terms of which grammatical morphemes are the output of phonological operations independent of the semantic operations that they realize. That is, affixation is a phonological operation of affix selection, whereas lexical derivation entails lexical categories and functional relations.
Regarding (5), it is remarkable that the suffix -bora is selected for the subjective function exclusively. This is the case because bora is a verbal element, morphologically related to the verb beran 'bear'. Bora as a free element means 'bearer', so that the agentive function is expressed by the inflectional suffix -a, which corresponds to the nominative singular masculine of the weak declension of nouns. In other words, the morphological relation holding between the strong verb and the deverbal noun is implicit. Bora turns out in the following complex words:
caegbora 'key-bearer, jailor' candelbora 'acolyte' ceacbora 'yoke for buckets' feorhbora 'life-bearer' hornbora 'horn-bearer, trumpeter' leohtbora 'light-bearer' locbore 'one who has long hair, free woman' mundbora 'protector, preserver, guardian, advocate; prefect' raedbora 'adviser, counsellor; (Roman) consul' raesbora 'counsellor, leader, guide' rodbora 'cross-bearer' segnbora 'standard-bearer' soobora 'soothsayer, astrologer' strolbora 'archer' sweordbora 'sword-bearer, swordsman' tacnbora 'standard-bearer; guide' waegbora 'wave-bearer' waepenbora 'weapon-bearer' wigbora 'fighter' witumbora 'bridesman' woobora 'orator, speaker, seer, prophet, poet, singer' wrohtbora 'accuser; the devil' wudubora 'wood-carrier'
These derivatives strongly indicate that the grammaticalization of a free lexeme has taken place, but also that the process of grammaticalization is blurred by lexicalization. Even though -bora derivatives are relatively transparent, we also come across some instances of lexicalization such as candelbora 'acolyte' and wrohtbora 'the devil'. Beginning with grammaticalization, this term is used in the sense of change from grammatical status into lexical status (Hopper and Traugott 2003) with the desemanticization of lexical forms, which develop more abstract meanings (Givon 2009). The process of grammaticalization must have taken three steps. The first involves the transitive type represented by wopenbora 'weaponbearer', which has a clausal correlate of the transitive type ('someone bears a weapon') and, moreover, involves an agent. The second step in the grammaticalization of bora must have consisted of the stative (possessive) type illustrated by locbore 'someone with long hair', with a clausal correlate 'someone has long hair', thus requiring a possessor (experiencer) rather than an agent. Finally, the third step of this process of grammaticalization must have comprised the stative (locative) type exemplified by ceacbora 'yoke for buckets', which diverges from the transitive clausal correlate ('the buckets are on the yoke') in displaying a non-possessor experiencer and a location. That is, there has been a process of meaning weakening and generalization whereby a transitive clausal correlate has yielded way to an intransitive clausal correlate of the copulative type (transitive > possessive > locative lcopulative). Overall, fully lexical meanings like 'to bear', 'to hold', 'to bring', 'to carry', etc. are replaced by the less specific grammatical meaning of possession. This loss of semantic weakening must have favoured the presence of abstract goals like mundbora 'protector', which appear hand in hand with concrete goals like and wudubora 'wood-carrier'.
Additional evidence in favour of the bound character of bora can be found in corpus analysis and in other sets of derivatives. From the perspective of corpus frequency, it turns out that bora as a free form is extremely infrequent. According to The Dictionary of Old English, there is a single occurrence of bora 'bearer' in the corpus. Considering derivational morphology as a whole, there are other suffixes morphologically related to the strong verb beran 'bear', along with bora itself. That is to say, the vocalic grade of the infinitive beran 'bear' corresponds to the one of the adjectival suffix -berende, the vocalic grade of the 3rd. person singular preterite indicative beer is identical with the one displayed by the adjectival suffix -beere and the vocalic grade of the past participle boren is also staged by the suffix with which this article deals. Thus, Old English has pairs like feorhbora 'life-bearer' and feorhberende 'living', sweordberende 'sword-bearing' and sweordbora 'swordsman', federbeere 'having feathers, winged' and federberende 'feathered'; as well as the triplet leohtbore 'brilliant, luminous', leohtberende 'light-bearing, luminous' and leohtbora 'lightbearer'. According to Kastovsky (1992: 50) -berende and -beere have dialectal distribution, in such a way that the former represents the Anglian form and the latter the West Saxon one. No such distribution has been proposed for bora. Nevertheless, the existence of these two suffixes reinforces the bound status of bora. It must not be forgotten, in this respect, that other genetically and areally related languages like modern German make use of a derivational morpheme that represents a reflex of bora.
This article has drawn a distinction between explicit and implicit morphological relatedness and applied it to Old English suffixed nouns. It has also classified deverbal nouns on the grounds of the functions identifiable with respect to a clausal correlate. It has been found that bora bears both explicit and implicit morphological relatedness and that it is selected for the subjective function exclusively. Evidence has been gathered in favour of the bound character of bora, including textual frequency and comparison with other suffixes. The grammaticalization of bora has been explained by means of the theoretical proposals of Lexeme-Morpheme Base Morphology, which distinguishes inflection from derivation but unifies derivations that assign the same function. Such a process of grammaticalization can be divided into three steps. The first involves the transitive type with a clausal correlate of the transitive type and an agent. The second step consists of the stative (possessive) type that requires a possessor. The third step the stative (locative) type that diverges from the transitive clausal correlate because it stages a non-possessor experiencer and a location.
Aceptacion comunicada: 21/12/2012
6. Bibliographical references
Beard, R. 1995. Lexeme-Morpheme Base Morphology. Albany: State University of New York Press.
Beard, R. and M. Volpe. 2005. Lexeme-Morpheme Base Morphology. In P. Stekauer and R. Lieber (eds.), Handbook of Word-Formation. 189-205. Dordrecht: Springer.
Givon, T. 2009. The Genesis of Syntactic Complexity. Amsterdam: John Benjamins.
Gonzalez Torres, E. 2010. The Continuum Inflection-Derivation and the Old English suffixes -a, -e, -o, -u. ATLANTIS 3211: 103-122.
Haselow, A. 2011. Typological Changes in the Lexicon. Berlin: Mouton de Gruyter.
Healey, A. diPaolo ed. 2003. The Dictionary of Old English in Electronic Form A-F. Toronto: Dictionary of Old English Project, Centre for Medieval Studies, University of Toronto.
Hopper, P. and E. Closs-Traugott. 2003 (1993). Grammaticalization Cambridge: Cambrige University Press.
Kastovsky, D. 1986. Deverbal nouns in Old and Modern English: from stem-formation to word-formation. In J. Fisiak (ed.), Historical Semantics-Historical Word Formation. 221-261. Berlin: Mouton de Gruyter.
Kastovsky, D. 1989. Typological Changes in the History of English Morphology. In U. Fries and M. Heusser (eds.), Meaning and Beyond. Ernst Leisi zum 70. Geburstag. 281-293. Tubingen: Narr.
Kastovsky, D. 1990. The typological status of Old English Word Formation. In S. Adamson, V. Law, V. Nigel and S. Wright (eds.), Papers from the 5th International Conference on English Historical Linguistics. 205-224. Amsterdam: John Benjamins.
Kastovsky, D. 1992. Semantics and vocabulary. In R. Hogg (ed.), The Cambridge History of the English Language I: The Beginnings to 1066. 290-408. Cambridge: Cambridge University Press.
Kastovsky, D. 2005. Conversion andlor zero: word-formation theory, historical linguistics, and typology. In L. Bauer and S. Varela (eds.), Approaches to Conversion/Zero Derivation. 31-50. Munster: Waxmann.
Kastovsky, D. 2006. Typological Changes in Derivational Morphology. In A. van Kemenade and B. Los (eds.), The Handbook of The History of English. 151-177. Oxford: Blackwell.
Martin Arista, J. 2008. Unification and separation in a functional theory of morphology. In R. Van Valin (ed.), Investigations of the Syntax-Semantics-Pragmatics Interface. 119-145. Amsterdam: John Benjamins.
Martin Arista, J. 2009. A Typology of Morphological Constructions. In C. Butler and J. Martin Arista (eds.), Deconstructing Constructions. 85-115. Amsterdam: John Benjamins.
Martin Arista, J. 2010a. OE strong verbs derived from strong verbs. SKASE Journal of Theoretical Linguistics 7: 36-56.
Martin Arista, J. 2010b. Lexical negation in Old English. NOWELE-North-Western European Language Evolution 60l61: 89-108.
Martin Arista, J. 2010c. Building a lexical database of Old English: issues and landmarks. In J. Considine (ed.), Current projects in historical lexicography. 1-39. Newcastle: Cambridge Scholars Publishing.
Martin Arista, J. 2011a. Projections and Constructions in Functional Morphology. The Case of Old English HREOW. Language and Linguistics 1212: 393-425.
Martin Arista, J. 2011b. Adjective formation and lexical layers in Old English. English Studies 92l3: 323-344.
Martin Arista, J. 2011c. Morphological relatedness and zero alternation in Old English. In P. Guerrero Medina (ed.), Morphosyntactic Alternations in English. 339-362. London: Equinox.
Martin Arista, J. The Old English Prefix Ge-: A Panchronic Reappraisal. Australian Journal of Linguistics. Forthcoming-a.
Martin Arista, J. Lexical database, derivational map and 3D representation. Revista Espanola de Linguistica Aplicada. Forthcoming-b.
Martin Arista, J. Recursivity, derivational depth and the search for Old English lexical primes. Studia Neophilologica. Forthcoming-c.
Mitchell, B. 1992. A Guide to Old English. Oxford: Blackwell.
Quirk, R. and C. L. Wrenn. 1994 (1955). An Old English Grammar. DeKalb, Illinois: Northern Illinois University Press.
(1) This research has been funded through the project FFI2011-29532. Revista Electronica de Linguistica Aplicada (ISSN 1885-9089) 2012, Numero 11, paginas 137-149
Miguel Lacalle Palacios
Universidad de La Rioja
|Printer friendly Cite/link Email Feedback|
|Author:||Palacios, Miguel Lacalle|
|Publication:||RaeL Revista Electronica de Linguistica Aplicada|
|Date:||Jan 1, 2012|
|Previous Article:||Modificaciones sintacticas basadas en la reordenacion de complementos del verbo con utilidad en esteganografia linguistica.|
|Next Article:||One Mind, Two Languages. Bilingual language processing.|