Go to originalLocalized version was produced by AI. It may contain inaccuracies.
Preparing document pages…
9 short passages of this localized version are shown in the original language because they could not be localized reliably.
Page 1
VIRGINIJA VASILIAUSKIENĖ
Institute of the Lithuanian Language
Research areas: historical syntax, syntactic typology, the language of old writings, orthography.
TYPEFACES AND ORTHOGRAPHY IN KONSTANTINAS SIRVYDAS’ BOOK OF SERMONS
Typefaces and Orthography in Konstantinas Sirvydas’ Book of Sermons Gospel Points
ABSTRACT
The article discusses the typefaces and the main orthographic features of Konstantinas Sirvydas’ Book of Sermons (SP). The orthography of SP was influenced most strongly by Latin and Polish. The orthography and typefaces of the first (1629) and second (1644) parts of Book of Sermons often differ. Some typefaces and spelling practices occur only sporadically in both parts. The variety of diacritic-bearing typefaces used in SP is greater than in Mikalojus Daukša’s postil and Jonas Jaknavičius’s Ewangelie Polskie y Litewskie. Most diacritics in SP are very rare, but some are used quite consistently. In some cases, diacritics marked a stressed syllable or served to distinguish homoforms and homographs. Variation in the marking of fricatives and affricates may have arisen for technical reasons or may reflect changes taking place in the Vilnius dialect. Following the Latin model, words were abbreviated and omissions were marked. Unlike the main Gothic text of SP, italic Antiqua capitals used three different typefaces to represent <I> and <J>. <ł> is used consistently. Long vowels were not graphically marked (<y> and <ū> served other functions). In many instances, the marking of [e] and [ai] after a palatalized consonant, the representation of assimilated consonants, and the spelling of the prefix iš- varied (the preposition is always spelled iš). Word division was influenced by prosody, the example of other languages, technical aspects of printing, and the publishers’ linguistic competence. Clusters are often written as a single word. Whether ne was written together or separately also depended on the verb form. Capitalization followed Latin and Polish tradition and the source being quoted.
KEYWORDS: Sirvydas, typefaces, diacritics, orthography, influence of Latin.
KEYWORDS: Sirvydas, typefaces, diacritics, orthography, influence of Latin.
30
Acta Linguistica Lithuanica LXIX
Page 2
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
ANNOTATION
This article describes the usage of typefaces as well as the basic principles of orthography in Konstantinas Sirvydas’ book of sermons Gospel Points (SP). Orthography of SP was mostly influenced by Latin and Polish. Most of the different diacritics are seldom used, but in most cases they are used consistently. In certain cases these diacritics are used to mark a stressed syllable or were a means of differentiation in homographs and homoforms. Variations in marking fricative consonants and affricates can be determined both by technical reasons in printing and by reflecting changes in the speech of Vilnius’ residents. Typefaces used for marking abbreviations and omissions followed Latin traditions. Three different typefaces were used during printing for marking both the capital letters <I> and <J> in those parts of the text that were in antique cursive. No special letters were used to differentiate long vowels from short vowels. Variations are evident in marking: (1) [e] and [ai] that follow a soft consonant; (2) voiced consonants before voiceless, and voiceless consonants before voiced; (3) the prefix iš-. Principles of combining adjacent words or writing them separately were influenced by prosodical features as well as by basic standards of other languages and also technical features of the printing process and language competence of editors and typesetters. Usage of capital letters is based on Latin and Polish or is taken from quotable sources.
In the absence of a surviving manuscript of Konstantino Sirvydo’s Punktų sakymų (hereafter SP), it is difficult to determine more precisely how closely the printing house’s type collections and the types selected by its printers corresponded to the author’s own orthographic conception and the writing signs he himself used. In other words, it is unclear what the relationship was between the writing signs used in Sirvydo’s manuscript and the corresponding typefaces in the printed text1. It is useful here to recall the regret expressed in Sirvydo’s preface, written in Polish, that his book was being published without accents because of a lack of funds: “The Lithuanian language would require certain marks, called accentus in Latin, so that someone for whom Lithuanian is not a native language might be able to pronounce and read it well; but this would also require new, special printing types, and entail expenses that I am unable to cover at present”2. Thus, the author understood that a perfect correspondence was impossible. There is no doubt that printers and editors intervened and adapted the text to the technical capabilities of the printing house. This is shown not only by studies of other old texts (Subačius 2001: 129–152), but is also evident from SP itself: the excessive variation of graphemes in the same positions within the same word, the presence or absence of diacritics in the same syllables-
1 For this reason, the term typeface is used frequently in this article in order to distinguish written and printed graphemes.
2 SP I [VIII]: “The Lithuanian language would need certain marks, which are called Accentus in Latin, so that one who does not know Lithuanian by nature might be able to speak and read it well; but they would also require special printing and expense, which I cannot afford now” (quoted in Lithuanian according to: Koženiauskienė 1990: 296).
Straipsniai / Articles
31
Page 3
V I R G I N I J A V A S I L I A U S K I E N Ė
in words containing them, and the use of special word abbreviations in order to fit within the narrow confines of the “Punktai.” In addition to the human factor, which was at least partly related to the technical capabilities of printing houses at the time, variation was also caused by the relatively weak, essentially still emerging tradition of written language. Unlike Mikalojus Daukša or Danielius Kleinas (Aleknavičienė 2006: 280, 284), Sirvydas had no special typefaces made for him. He had to work within the limits of type collections (founts) designed for printing Latin and Polish texts. The complex orthography of Daukša’s postil (hereafter DP), and some of the typefaces he used, were gradually modified by later editors and printers of books in the Grand Duchy of Lithuania, with greater emphasis on functionality and ease of use. Some of the typefaces introduced by Daukša had already been abandoned in the anonymous catechism of 1605 (LK): <ē> is not used; instead of <ū>, <uo> is often written; and, owing to dialectal features, nasal graphemes were unnecessary in the catechism (Palionis 1967: 38–39; 1995: 27). Sirvydas’s orthography, being simpler than Daukša’s, became established for a long time in the eastern and later also the central variant of the written language.
1. SERMON POINTS TYPEFACES AND THEIR USE
1.1. In preparing a transcription for the critical edition of SP3, all the typefaces used there were selected and systematized. Compared with the typefaces used in DP as described by Jonas Palionis, the Lithuanian text of SP clearly displays a greater variety of typefaces with diacritics (only some consonants with diacritics used in DP are not found here: <ǵ>, <m̄>, <ṕ>, <ṕ>, <ṙ>; nor the two diphthongs <ē>, <ū>), but their frequency of use is considerably lower (stress is not marked). The smaller number of typefaces was probably due to the fact that Palionis described, in his own words, not all but only those typefaces used more regularly (Palionis 1967: 33), whereas in preparing the aforementioned transcription of SP, all typefaces were selected and recorded, regardless of how regularly or frequently they were used. As our study shows, Palionis likewise presents in his books not all the typefaces used in SP, but only those used more regularly (Palionis 1967: 39–41; 1995: 28). In presenting an overall picture of the typefaces in individual old Lithuanian books, it would be worthwhile to distinguish the frequency of use of typefaces with diacritics from the variety of those typefaces.
1.2. The second part of Punktų sakymų (SP II, published in 1644) was prepared for publication and translated into Polish by Sirvydas’s assistant and successor in his work, the Jesuit Jonas Jaknavičius. It therefore seems entirely reasonable to assume that SP, especially in its second part
3 This edition was prepared under an agreement with the Institute of the Lithuanian Language’s Department of the History of Language and Dialectology by Virginija Vasiliauskienė and Kristina Rutkovska. Its preparation in 2008–2011 was supported by the State Commission of the Lithuanian Language. The transcription of the Lithuanian part of SP was prepared by Virginija Vasiliauskienė, and that of the Polish part by Kristina Rutkovska.
32
A c t a L i n g u i s t i c a L i t h u a n i c a L X I X
Page 4
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
(SP II), the typefaces used should not differ substantially from those used in Jaknavičius’s edition of Ewangelie Polskie y Litewskie (hereafter JE, published in 1647), especially since the two were separated by only three years. According to the study conducted by Milda Lučinskienė (Lučinskienė 2005: 30), almost all the typefaces found in Jaknavičius’s Ewangelie Polskie y Litewskie were also used in the Lithuanian text of SP (with the exception of <ą>, <ò>, <ù>). However, compared with SP, the set of typefaces in JE is less extensive: JE does not use <q> or <x>, nor does it contain a considerable number of the typefaces with diacritical marks found in SP: <ă>, <ă>, <á>, <ã>, <č>, <ê>, <ē>, <í>, <ì>, <î>, <ī>, <ĩ>, <ȳ>, <ŷ>, <m̄>, <ń>, <ň>, <ô>, <ò>, <ō>, <š>, <û>, <ũ>, <ų>, <ŵ>, <ž>, <Ʒ> or <Ǯ>, <Ǯ>, <Ǯ>, <Ǯ>. In addition, SP also uses special signs borrowed from Latin to indicate word-final abbreviations: <q>, <⁹>, and <°>.
1.3. The following typefaces with diacritical marks are used in the Lithuanian text of SP (see Figs. 1–60):
a: <á>, <à>, <ă>, <â>, <ã>, <ā>, <ă>, <à>, <ā>.
c: <ć>, <č>, <ċ>.
e: <é>, <ê>, <ę>, <ē>, <è>.
i: <í>, <ì>, <î>, <ĩ>.
y: <ȳ>, <ŷ>.
l: <ł>.
m: <m̄>.
n: <ń>, <ñ>, <ň>, <ṅ>.
o: <ó>, <ô>, <ò>, <ō>, <õ>.
s: <ś>, <š>, <ŝ>.
u: <û>, <ũ>, <ū>, <ų>.
w: <ŵ>.
z: <ź>, <ż>, <z̄>.
Ʒ: <Ǯ>, <Ǯ> or <Ǯ>, <Ǯ>, <Ǯ>.
Ȝ: <Ȝ>, <ȝ>.
Some of the typefaces listed here with diacritical marks are used particularly rarely; in isolated cases, they occur only once or a few times in the entire book or in separate parts of it. In the Polish text of SP, diacritics are more varied and frequent than in the Lithuanian text (they are studied by Kristina Rutkovska).
1.4. According to their frequency and systematicity (consistency) of use, all typefaces with diacritical marks used in SP I and SP II can be classified as follows:
(1) typefaces with diacritical marks used frequently and systematically in SP I and SP II: <ć>, <ł>, <ś>, <ǯ>, <ǯ>;
(2) typefaces with diacritical marks used inconsistently but relatively frequently in SP I and SP II: <á>, <à>, <č>, <ǯ>;
(3) typefaces with diacritical marks used very rarely and inconsistently⁴:
4 A question mark denotes a typeface used particularly rarely, with an indistinctly printed diacritic.
Straipsniai / Articles
33
Page 5
V I R G I N I J A V A S I L I A U S K I E N Ė
(4) typefaces with diacritical marks are used only in pericopes and section headings (Latin roman and italic):
SP I: <š>, <ž>, <ž>; SP II: <ė>, <ž>, <ž>, <ž>.
Having recorded all typefaces with diacritical marks used in SP, we can see that their variety is greater in SP I, while their number is greater in SP II—the more frequent use of these typefaces in SP II is due to the very frequent <á>.
1.5.
Typefaces representing the consonants [š], [ž] and the affricate [č]
The following typefaces with diacritics are used in SP to represent the fricative consonants [š] and [ž]: <ś>, <š>?, <š>, <ӡ>, <ӡ>, <ӡ>, <ӡ> or <ӡ>, <ӡ>, <ӡ>, <ź>, <ź>, <ž> (see Figs. 38, 40–42, 48–58); the affricate [č] is represented by <ć>, <ċ>, <ċ>? (see Figs. 10–12). In addition, [š] can also be represented by the digraphs <ß>, <ſz>, and [č] by the digraph <cȝ> (very rarely).
1.5.1. Uppercase <ӡ>, <ӡ> occur sporadically only in SP II: Žydu SP II 162₁₉–₂₀, 202₄, 332₃; Žynio SP II 104₈, Žodžiey SP II 83₂₆. In all other cases they are replaced by <ӡ>⁵. <ś> is found very rarely in pericopes in SP I printed in italics: graši 317₈, while the only instance of <č> occurs in the main text of SP I: lynčiuoguofna 373₁₀ (the form of the diacritic is unusual here). Also very rare in SP II is <ś>, which should essentially be regarded as a poorly printed <ś>: ráßtas s. SP II 188₁₃; páráśitam 254₆, karaliśčiu SP II 14₁₆–₁₇.
1.5.2. <ź>, <ż>, and <ž> are used in text set in Latin roman and italic type. In the SP II roman typeface, the diacritic above <ž> (see Fig. 50) resembles a hollow dot and differs in form from <ź> and <ż>, which are used elsewhere in the text: Nežadžia 51₁₆, Bažnīčia 153₁₇, žìames 153₁₉, Vžftokles 180₈, ażuturet 239₃. The latter two typefaces are used in both SP I and SP II. <ź> is used consistently in the affricate <dź> in SP II: żodźiey 248₂₂, żuǯwes 248₂₀, Ižkur 116₂₅, didžio 116₂₆; compare SP I: ižpažino 163₁₆, ižpilde 163₁₇, žodieys 244₂₁, žiames 269₉.
1.5.3. The forms of the diacritics vary greatly on the typefaces <ӡ>, <ӡ>, <ӡ>, <ӡ> or <ӡ> used in the main text set in Schwabacher type (see Figs. 51–58). These typefaces use diacritical marks in the form of a dot, acute accent, dash, and a curved mark and/or a caron. Because of the print quality, it is often difficult to distinguish one from another: smudg-
5 Other uppercase letters were also used without diacritical marks. Franz Specht had already written about this in 1929: “In Unzialschrift stehen C, S, Z, L auch č, š, ž, ł, d. h. es fehlen die diakritischen Zeichen” (Specht 1929: 13).
34
A c t a L i n g u i s t i c a L i t h u a n i c a L X I X
Page 6
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
An acute accent or hyphen may look like a dot, while a poorly printed dot may resemble a small arch. For example, in the word žmones SP I 326₁₂, the arch looks like a smudged dot, whereas the arch in the word gražuolė SP II 244₂₇ is most likely a dot that has not printed fully. Diacritics in the shape of a caron or small arch above <ž> and <ž> occur only a few times in the SP text. In SP I they look more like carons, and in SP II more like small arches: tulžis SP I 321₂₁; didžiu SP II 199₅–₆.
In doubtful cases, an attempt was made to enlarge the type, but when enlarged further, the shape of the diacritic began to differ from what was visible to the naked eye; thus, relying solely on this method of identifying an unclear diacritic would have been quite risky. In SP II, the diacritic above <ž> (see Figs. 57, 58), as in the case of the roman type <ž>, resembles a horizontally elongated dot (cf. Fig. 50). By convention, it was represented as a hyphen above the type, in order to distinguish it from the traditional dot-shaped diacritic.
A comparison of individual types representing the same grapheme shows that neither the types themselves nor their diacritics had identical forms in the printing fonts; for example, some <ž> differ from one another. The same can be said of a considerable number of other types, especially those with diacritical marks.
1.5.4. In SP, the fricatives [š], [ž] and the affricate [č] can often be represented by types without diacritical marks <ſ>, <s>, <з>, <z>, <c>, and conversely, <ś> may occur in place of <ſ>, <s>: šilpnas SP II 12₂ and filpnas SP I 211₁₇; žiemes SP I 65₁₆ and žiemes SP I 261₁₃; žimus SP II 41₁₆–₁₇ and žimus SP I 68₈; karſčiaus SP II 50₁₆, karščiaus SP II 38₁, karſčiuole SP I 244₁₇; čielas SP I 247₁₅ and celas SP I 67₂₅; ſiurkſčiev SP II 57₁₅ and šiurkſčtibes SP II 107₂; ſzwynčiaufias SP II 89₈ and ſzwynčiausiami SP II 239₁–₂; ſirdiy SP II 16₈ and sirdiy SP II 239₂₂, as well as βyrdies SP II 169₁₂; parajzie SP 217₃₀, parašie SP I 349₂₈ and parafie SP I 307₅–₆; and so on. Let us also compare the spelling of SP words with the root śvies-: (fwief- / fwies- / świef- / świes- / βwief- / świes- / βwies-): fwielibes SP I 20₂₉, SP II 32₂₉; fwiesibe SP II 28₅; świeſibe SP I 353₂₄, SP II 197₁₁; świesibe SP I 354₆; βwieſibes SP I 202₆, SP II 215₇–₈; świeſibes SP I 353₂₇; świeśibi SP II 33₁₃; βwieśiaus SP II 201₂₆.
Given that the listed cases were neither isolated nor accidental, and that they also occur in other printed works of the period (Palionis 1967: 37)⁷, at least some of them should be attributed to the conventions governing the use of type in printed books of that time. Some diacritics on these types may simply have failed to print; however, the alternation between a type with a diacritic and one without it in the very same, often even
⁶ Franz Specht’s statement, “Für den dentalen Sibilanten wird ſ und s promiscue gebraucht, nur pflegt ſ nicht im Anlaut zu stehen” [For the dental sibilant, ſ and s are used interchangeably; only ſ is generally not used in initial position], most likely contains a proofreading error—it was presumably meant to say: “...nur pflegt s nicht...”
⁷ In Merkelis Petkevičius’s 1598 catechism, the consonants š, č are represented by the single letters s, ſ, c or the diacritic single letters ś, ć.
Straipsniai / Articles
35
Page 7
V I R G I N I J A V A S I L I A U S K I E N Ė
words used in close proximity (cf. SP I, pp. 20 and 353)—and throughout the SP text rather than sporadically—would suggest that diacritics, at least during the printing process, were not always considered an essential distinguishing feature of a grapheme⁸. Other causes of such variation may have been the absence of types with diacritics for capital letters and the pronunciation of Vilnius, influenced by its multilingual environment. For example, words with the root silpn- are written 21 times with <l>, <s>, and 11 times with <ś>. In SP II, these types occur at roughly equal frequencies (12x and 10x, respectively): filpniby SP I 185₁₉, filpnina SP II 57₁₄, silpnu SP I 211₂₃, silpſta SP II 75₂₄, śilpnu SP I 61₃₁, śilpnibes SP II 218₁₃.
Variation between types representing the palatalized [s] and [ś] in the same words is more frequent in borrowings from Polish and Belarusian. Moreover, <ś> is chosen relatively more often than <s> and <ſ> in SP II; for example, silos occurs 6 times in SP I and 3 times in SP II; śilos occurs 2 and 3 times, respectively; świetas is used more often in SP II than ſwietas; whereas śwletas is very rare in SP I and may be considered a proofreading error; ſmertis predominates in both parts of SP; śmertis occurs five times, only in SP II. As noted above, this spelling most likely reflects variation in pronunciation in the Lithuanian spoken in multiethnic Vilnius. According to Zigmas Zinkevičius, Poles had some difficulty distinguishing Lithuanian [š] from [s] (Zinkevičius 1981: 30–32). In both cases they heard Polish [ś], and consequently wrote very inconsistently: <s> or <sz>.
Whether variation between <ź> and <z>, and between <č> and <c> in the same words can be linked to a particularly strong opposition between [ž] and [z], [č] and [c] in the speech of the Vilnius population of that period cannot be confirmed or ruled out on the basis of SP data alone.
Thus, variation between types with and without diacritics for representing fricatives and affricates must have been caused by several factors, not just one: the conventions of use in contemporary printed works, pronunciation, the absence of types for capital letters with diacritics, and print quality.
1.6. Typefaces with diacritical marks over <a>
The greatest variety of diacritical marks is found over <a>: <á>, <à>, <à>, <á>, <à>?, <á>?, <â>, <ã>, <ā> (see Figs. 1–9).⁹ Of these, only <á> is used more frequently and systematically; it may be associated with the Polish “raised” <á>. In the Lithuanian text of SP II, this typeface quite often marks short, unstressed [a]. <á>
8 In Mikalojus Daukša’s postil, the following spellings were used: dabókites, artés 16₃₃, netikéfites 16₃₆, paáugſtinos 87₁₀, Bažniczios 87₁₈, Bažniczia 87₁₉, gárſtĭczios 87₆, garſtĭczios 87₃₈, and so on.
9 <à>, <â>, <ã>, and <ā> are described in greater detail together with other typefaces bearing diacritical marks of the corresponding form (see 1.7; 1.8; 1.13).
36
Acta Linguistica Lithuanica LXIX
Page 8
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
The use of <à> in SP is likewise consistent: kárálifti SP II 87₁₅, ážulayko 89₂₅, reykálu 103₅, Sámarítonas 119₁₄–₁₅, niekádu 120₂₈, átaimo 126₁₄, ážuláyko 149₂₃–₂₄, márindámi 155₂₃, iftáte 153₂₂, áźndingie SP I 223₁₃–₁₄, gieldžimis SP II 155₁₃–₁₄, átleyde 167₅, ádunt 177₂₀, eydámás 178₂₉, ažufirakina SP I 332₃, prieżáftu 365₄–₅. Visually, these two typefaces are also not always distinguishable from one another. They are used much less frequently in SP I, where <á> is most often chosen not, but <à>. In SP II, <à> occurs a few times and its use does not differ from that of <á> and <à>, while the one instance of <à> in this part is probably a poorly printed <á>; the function of <à>, used a couple of times in SP II, is also unclear—the diacritic, because of its slight inclination to the right, resembles an apostrophe and an acute accent, while in the word gálu SP II 247₁₃ it may also be a dot that was slightly smudged in printing and consequently changed in shape.
1.7. Typefaces with a grave accent
Gravis ant <à> dukart pavartotas lotyniškojoje antikvoje (Strockis 2004: 294–295): à kayre labay ilga bus SP I 240₂₀–₂₁. Gravio ženklas kituose dviejuose švabachu išspausdintuose žodžiuose atrodo neįprastai, kiek pasislinkęs į šoną (žr. 3 pav.). Be <à>, gravio formos diakritikai dar atsitiktinai gali būti pavartoti ant <č>, <è>, <ì>, <š> (žr. 12, 18, 19, 42 pav.) ir yra būdingesni lotyniškajai antikvai: <è> (SP II) ir <š> (SP I) aptinkami tik antikvoje, nelabai aiškios formos <ì> antikva ir švabachu surinktame tekste, o <č> (SP I) vienąkart pavartotas švabachu išspausdintoje teksto dalyje.
1.8. Typefaces with a circumflex accent
Circumflex-shaped diacritics are used sporadically in SP I over <â>, <ê>, <î>, <ô>, and <û> (see Figs. 7, 14, 20, 38, 43): kuriâs (acc. pl.) 171₂₆, 373₁₉, gieriâus (adv. cmp.) 198₁₇, mariâs (acc. pl.) 269₁₈, iiêmus (dat. pl.) 367₂₅; čía 267₉, marios (nom. pl.) 268₇, takiuôs (acc. pl.) 367₁₁, iuôs (acc. pl.) 198₂₄, tuôs (acc. pl.) 367₂₄, kuriûo (instr. sg.) 368₁₁, punciûofna (il. pl.) 373₈, kuriû (instr. sg.) 352₈, Iûoda (nom. sg.) 229₂₃, gieriaûs 199₂₃, 360₁₈, perfḳrodûs 14₁₉, perfmegiûs 14₂₁. They most often mark stressed syllables in the accusative plural of pronouns, and occasionally in the instrumental singular. These syllables are more often long than short. In other isolated words and in unstressed long syllables, the appearance of this diacritic is probably accidental or should be regarded as a proofreading error (the function of <î> is unclear). In some words, the typeface bearing this diacritic may have served not only to mark stressed syllables but also to distinguish homographs and homoforms (e.g., gieriâus, gieriaûs, kuriâs, mariâs, perfḳrodûs,
Straipsniai / Articles
37
Page 9
VIRGINIJA VASILIAUSKIENĖ
perſmegiùs). Attention should also be drawn to the distribution of this diacritic: it is not evenly distributed throughout the text, but is often concentrated in particular passages and on pages where other forms of diacritical marks (acute accents, circumflexes, strokes) are also found more frequently.
1.9. Other isolated type sorts with diacritics
The isolated type sorts <ó> and <ò> may have marked a stressed syllable or distinguished forms: wiró (gen. sg.) 219₂₁, ó¹⁰ 175₂₄, Maldõs (nom. pl.) 67₂₄, Iõnuy (cf. Iãnowi) 140₁₇, neflidabōdami 84₃₀. It is impossible to determine more precisely the functions of <ó>, <ò> and <ô> in SP (see 1.8), as, for example, in the texts of Baltramiejus Vilentas (Kabašinskaitė 2005: 56–62), because there are too few examples.
An acute accent over <é> marks a stressed, usually long syllable: kuriuosé (loc. sg.) 177₁₁, kurié (nom. pl.) 162₈, praiéwu (gen. pl.) 177₂₆, gaylétus (cnd. 3) 158₁₉, apé 176₁₁; however, there were also a couple of instances in an unstressed long syllable: Wiéta (nom. sg.) 203₇ and éſſibi (acc. sg.) 171₃.
In SP, two consonant type sorts with an acute accent, <ń> and <ẃ>, and one consonant type sort with a dot, <ṅ>, were used a few times. In these particularly rare instances, the type sorts <ń> and <ṅ> most often mark a palatal [n], as in Polish. The purpose of the diacritic in the illative endings of the words fkrinion SP I 369₃₀ and kloniñ SP II 248₁₈ is unclear: <ń> may have been confused with a stroke-shaped diacritic over a vowel, or that <ń> may have been added later when restoring an earlier abbreviation (cf. peklõ SP I 279₁₂, wietō SP II 247₂₁).
Twice in SP I, <ẃ> was probably used accidentally, in the words wúnduo 264₅ and wífi 286₁₁. Its function is unclear.
1.10. Type sorts with diacritics and stress
As already mentioned in the introduction, SP did not use special type sorts with diacritics solely to mark stress. The rare cases of stress marking tentatively identified in sections 1.6, 1.7, 1.8, and 1.9 are intertwined with other functions performed by these type sorts and are not consistent or unidirectional when all available examples of any particular diacritic are considered together. Since rarely or especially rarely used diacritics are concentrated in specific parts of the text, it is highly likely
10 Diacritics are also found in this instance on the corresponding words in the Polish part of the text, cf.: mężą 219₂₁ and á.
38
Acta Linguistica Lithuanica LXIX
Page 10
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
It is possible that they entered the text here from other, most likely manuscript, sources that Sirvydas may have used. Because they are rare, intermingle with other functions, and are not used consistently, it was impossible to isolate and precisely define the cases in which typefaces with diacritics were used to mark stress.
1.11. Typefaces with a cross-stroke
In SP, typefaces with a cross-stroke <a>, <ę>, and <ų> were used several times (only in SP I; see Figs. 15 and 44). As in DP, they did not mark nasal sounds, since nasal sounds had already disappeared from the Eastern Aukštaitian dialect at that time; thus they may have entered Sirvydas’s text accidentally from the Polish part of the text, or may have been adapted for other functions. The function of the typeface <ų> in the words geriaus 36826 and fų 3708 is unclear (does it mark a long stressed syllable?). The use of <ę> should most likely also be considered a proofreading error in the word metu (instr. sg.) SP I 14, as well as in the words βwęčiaufi SP I 3698 and paļipeneimo (the cross-stroke on <ę> is faint) SP II 1763, where the marking of <n> in the word is duplicated. In the other three cases, <a> and <ę> indicate omitted <n>: Kadąg I 9622, nutrėkimu SP II 3411, bedrautumite SP II 1682.
1.12. The typeface <ł>
The typeface <ł> is used relatively consistently and systematically throughout the SP text to denote a hard (1) or depalatalized (2) [ł] before vowelism of the [e], [ei], [ė], [i] type (Palionis 1967: 84–85):
(1) łūsis SP I 29521, fumala SP I 3106, mieláśirdingus SP II 2728, łaymibe SP II 247);
(2) łeyzdawos SP I 1286, łeydźiaji SP II 927, łeyftu SP I 34318, naßle SP I 3148, nupleßt SP II 2475–6, łepinimay SP I 3014–5, gayłeimu SP I 14131, pergałeimas SP II 22910, łingwa SP I 5130, SP II 361.
After [ł] is hardened, the changed vowels are often represented by <a>, <y> (more often in SP II11): pałayftułe SP I 3749, pełanuołe SP II 17928, ʃłankfti SP II 17915, ne ápłayft SP II 17721, ápłaydus 17724, apłayʒdamas 17727, nułynkiame SP II 1486, nułynkias SP I 10524, ßakieły SP I 36930. The retracted [ė] is represented by <e>: pergałeio SP I 18517, parpuołe SP I 604, gułeio SP I 2692saute SP II 427, tiłetu SP II 7324.
11 In SP II, other consonants can also be hardened, for example: graytay 619; trácias 7612, žynot 19622, prywerʃt 18627, ßytkuose 17930, ßyrdies 16912, but examples also occur in SP I: žynklina 2527.
Straipsniai / Articles
39
Page 11
VIRGINIJA VASILIAUSKIENĖ
[ł] remains hard in some consonant clusters:
[łpn] before [e]- and [i]-type vocalism: filpniby SP I 18519, filpnefnio SP II 9714, filpniem us SP II 6931, filpnina SP II 5714, šilpnesnia SP II 129, šilpni SP II 976, šilpnefnio SP I 12919, Silpnibe SP I 1837 (the soft <ł> is written in only three of the 20 occurrences of root weak-form words);
[łb] before [e]-type vocalism: kalbet SP I 21321, SP II 635, kalbefis SP I 22727, kalbedami SP II 2408, kalbečia SP 1203, kalbeio SP 20210 (the soft <ł> is written only in the infinitives of two root kalb- words in SP II), giebek SP I 27829, gialbeio SP II 16528–29.
There are not many errors in marking [ł]. The cases in which hard <ł> is used incorrectly are more reliable, since, owing to the quality of printing, the stroke through <ł>, that is, the mark of hardness, is not always discernible. The following inaccuracies in marking <ł> are listed in Franz Specht's list of SP errata (Specht 1929: 48–61) (SP I 1–100):
The word fuodele (instrumental singular) SP I 3615, in which the hardness of [ł] is marked, can be emended in two ways: fuodele and fuodely (restoring the dialect form). Specht prefers the latter.
In the SP original used by Specht, on which the facsimile edition was based (ŠPS 1929), the diagonal stroke indicating the hardness of <ł> is not always discernible; however, in some words included in his list of errors, hardness was marked and is clearly visible in other surviving copies of SP¹², for example: algoia SP I 2027, bilo SP I 721, pelenays SP I 8512, tulu SP I 9713.
Taken as a whole, this very detailed list of errors reflects the state of research into old writings and dialects at the time, the then-current understanding of proofreading errors, and the characteristics of the original used to prepare the facsimile (in not all copies are fine details of the type, especially diacritics, printed equally clearly).
12 Plg. Lietuvos diplomatui Jurgiui Šauliui (1879–1948) priklausiusį ir šiuo metu Pensilvanijos universitete saugomą Punktų sakymų konvoliutą: Annenberg Rare Book and Manuscript Library, University of Pennsylvania; Location: Rare Bk & Ms Library RBC; Call Number: 491.922 Si78.2.
40
Acta Linguistica Lithuanica LXIX
Page 12
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
1.13. Typefaces with a tilde diacritic shaped like a hyphen and/or a circumflex
Diacritics resembling a circumflex (SP I) and a hyphen (SP II) above vowel typefaces <ã>, <ā>, <ē>, <ĩ>, <ȳ>, <ȳ>, <õ>, <ō>, <ū>, <ǖ> (see Figs. 8, 9, 17, 21–24, 38, 39, 45, 46) consistently, though infrequently, mark omitted tautosyllabic [m] and [n] in the SP text (Palionis 2005: 31–33; Strockis 2004: 294–295)13 and should be regarded as tildes—special abbreviation marks: darbūp6 SP I 526, nuteriojųiā SP I 14929, bud{ūlp SP I 28713, peklō SP I 27912, wietō SP II 24721, füti SP II 94–5, pafkūdoy SP II 1537, idāt SP II 3425, ūt SP II (10x), pafjrētū SP II 9727, βwētas SP II 7327, trūpay SP II 4611, βw{ȳ}tas SP II 815. In two cases in SP II, omitted [n] is also indicated by a typeface with a crossbar <ę> (see 1.11.). The letters <ñ> (in SP I most likely <ñ>) and <m> (see Figs. 29, 33, 34) most often indicate geminates in loanwords: Paños (cf. Polish 3 Pānny) SP I 6716, Hofāna (cf. Greek hōsanna) SP II 14410, fūmu (cf. Latin summa) SP II 1642. The abbreviation marks introduced, following the example of Latin, were intended to save space in the narrow lines of the SP where necessary.
1.14. Word-final abbreviation typefaces
The typefaces used for this purpose include: <g>=us, <ᵍ>=us, <ᵒ>=go, <ĝ>=go14 (see Figs. 61, 62, 63), and a colon (most often in SP II): Tūlg SP I 25110, mg SP I 10028, Chrift9 SP I 10023, Kaln9 SP II 394, kitiem9 SP II 2817, žimčiū9 SP II 2713, širg9 SP II 25718, Wieβ: SP II 14613, Wieβp: SP II 16512, Pfal: SP II 5211, Pfa: SP II 14912, Dowid: SP II 1224.
1.15. Typefaces <I>, <J>, <J> and <j>
In the main text of the SP, printed in Gothic type, a single capital letter is used to represent the sounds [i] and [j] at the beginning of a word (Subačius 2003: 227–228)15. Antiqua
13 The use of diacritics (tildes) shaped like a circumflex and/or a hyphen in the Lithuanian text of the SP was adopted from Latin, where tildes marked omitted tautosyllabic <m> and <n>. Such marking is also common in the Latin marginalia of the SP (Petreikis 2011: 67–85).
14 Word-final abbreviations are marked with these signs even more frequently in the Polish text of the SP and in the Latin marginalia. This method of word-final abbreviation was adopted from Latin.
15 In texts printed in Gothic type, <I> and <J> were not distinguished. When Antiqua was introduced in Lithuania, printers continued for a time to follow the conventions of Gothic type and likewise did not distinguish <I> and <J>: they chose one or the other typeface to represent both sounds [i] and [j], or used them interchangeably.
Straipsniai / Articles
41
Page 13
V I R G I N I J A V A S I L I A U S K I E N Ė
In the italic-printed pericopes, three differently shaped typefaces¹⁶ are used to denote both these sounds: <I>, <Ƒ>, <J> (see Figs. 25, 26, 27; Gelumbeckaitė 2003a: 81–82; 2003b: 302; cf. Juda 2001). SP I uses only the first two: in nine pericopes, only <Ƒ> in all instances (SP I: 1, 69, 122, 214, 234, 262, 281, 296, 317); in one pericope, only <I> in all instances (see SP I: 195); and in four pericopes <I> and <Ƒ> occur intermixed (see SP I: 32, 163, 338, 358), for example:
SP I 32: fžgirdis, Jonas, Jezus, Jonuy, Jefus, fonu;
SP I 338: IezauIp, fƀeio, fr 2x, fumus, fra;
SP I 195: Iezus, Ierozolimu, Jezus, Ierozolimoy, Jeruzalem, Ir 4x, Jezus;
SP I 122: fudeios, ftureios, fonop, fr 2x, fordano, flaiiofias.
In the first three pericopes of SP II, <I> and < Ƒ> are likewise used intermixed (see SP II: 1, 21, 43–44); however, in the last seven, another typeface of a different shape, not used previously, appears: <J> (SP II 77–78, 108–109, 143–144, 172–173, 225, 235–236, 248–249):
SP II 1: IEZVS, Jezus, IEzus, fr, fey 2x;
SP II 77–78: Iezus, Jezus, Iž, Jeme, Jog, Jra;
SP II 143–144: Jezus, Ieroſolimosъ, Jezus, Jezus;
SP II 172–173: Jezus, fr, Iudaƶius, Iƀkarioto, Ir, Jezus 2x, fey, Ius, Jey.
The examples show that all three typefaces under discussion could denote both [j] and [i], and were not associated with only one or the other; that is, their functions were not differentiated. In SP I, one can discern efforts to manage the use of two distinct typefaces by limiting their variation within the text and more often choosing <Ƒ> for both functions. In SP II, variation seems to be regarded as the norm and is not restricted in any way. One reason why the notation of [i] and [j] is not differentiated in the antiquae text of SP must certainly have been its setting amid Gothic text—the Gothic, rather than the antiqua, typeface was the principal typeface of the SP text (Aleknavičienė 2006: 286). Another factor likely to have caused confusion was the prothetic [j] at the beginning of words before front vowels, characteristic of eastern and central Lithuanian dialects (Palionis 1995: 50; Zinkevičius 1996: 102) (in the pericopes: Jeme SP I 358₃, Jeme SP II 21₃, Jeme SP II 78₁₃; in the sermon text: iiebƀoio SP II 97₃₀₋₃₁).
The typeface <j> is not used in SP I. In SP II, this typeface occurs 20 times (see Fig. 28). Almost all instances are concentrated on pp. 1–47: in the pericopes, they are most often forms of the pronoun jis: ijƀai 1₈, (2x), ij 1₁₆ (acc. sg.) (6x), jiemus 21₇, juos 44₂, priima 44₃¹⁷; while in the sermons,
¹⁶ In SP, these typefaces differ somewhat in form from the counterparts chosen to represent them in this article. The first typeface has the precise form of a Roman numeral one; the second has an elongated, slightly curved vertical stroke with a horizontal bar across the middle; the third has an elongated, slightly curved vertical stroke.
¹⁷ JĘ is written in the corresponding places: ijƀai 40₁₆, ii 40₂₃, iemus 42₂₀, juos 46₁₃, priima 46₁₆.
42
A c t a L i n g u i s t i c a L i t h u a n i c a L X I X
Page 14
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
without ij 34₁₈ (6x), with <j> still written as igija 6₂₃ and kalaviju SP II 18₇. On a couple of occasions, <j> is recorded in later passages of the sermon text (not the pericopes): bijos SP II 52₅, wienatij SP II 215₂₉. In all the words, <j> follows <i>: <ii> is written¹⁸ (Jolanta Gelumbeckaite 2009:19); therefore, during the research process, the question arose whether ij¹⁹ could be the definite form jįjį of the pronoun jis, but it is also written in non-final positions of this word and in other words. The occasional definite forms of pronouns in SP II are usually written with <y>: toy (nom. sg.) 60₄, 100₂₇, tiey 1₇, 132₁₈, nekuriey 4₁₁, 42₂₀. Thus, on pages 1–47 of SP II, which cover almost the first three sermons (together with the pericopes), there was an intermittent attempt to distinguish the notation of [i] and [j] by introducing lowercase <j>. In addition, as already noted, this part of the text is also distinguished by the fact that <I> is not used in its pericopes.
In SP I, <j> is used neither in the main Gothic text nor in the pericopes printed in Antiqua. Here [i] and [j] always correspond to <i> and <I> in the text set in Gothic type, and to <i>, <I>, and <j> in Latin Antiqua.
1.16. Examples of typefaces used in SP
FIG. 1. SP II 65–5
FIG. 2. SP I 364–3
FIG. 3. SP I 333–24
FIG. 4. SP II 170–32–4
FIG. 5. SP II 179–2
FIG. 6. SP II 191–23
FIG. 7. SP I 198–17
FIG. 8. SP I 149–29
FIG. 9. SP II 73–23
FIG. 10. SP I 35–6
FIG. 11. SP I 303–29
FIG. 12. SP I 373–10
¹⁸ In Jono Rėzos Psalmtere Dowido (Der Psalter Davids Deutsch und Litawisch [RPs 1625]) <ij> and <gh> are allographs of <j>.
¹⁹ In SP II, the pronoun form jį is written in three ways: most often ii, occasionally ij and gi. In SP, [j] is represented by <i>, <y>, and occasionally, in SP II, also by <j> (pp. 1–47) and <g> (SP II, pp. 53–58, 112–113, 121–126, and once in SP I 211₃₁, if this is indeed not the particle gi).
Straipsniai / Articles
43
Page 15
V I R G I N I J A V A S I L I A U S K I E N Ė
1 3 P A V. SP I 162-8
1 4 P A V. SP I 367-25
1 5 P A V. SP II 34-12
1 6 P A V. SP I 372-30
1 7 P A V. SP II 73-27
1 8 P A V. SP II 236-12
1 9 P A V. SP I 332-2
2 0 P A V. SP I 268-7
2 1 P A V. SP II 104-18
2 2 P A V. SP II 172-10
2 3 P A V. SP II 103-4
2 4 P A V. SP II 14-3
2 5 P A V. SP II 75-22
2 6 P A V. SP II 76-10
2 7 P A V. SP II 76-19
2 8 P A V. SP II 1-16
2 9 P A V. SP II 164-2
3 0 P A V. SP I 369-30
3 1 P A V. SP II 242-18
3 2 P A V. SP I 373-16
3 3 P A V. SP I 93-7
3 4 P A V. SP II 70-19
3 5 P A V. SP I 219-21
3 6 P A V. SP I 140-17
3 7 P A V. SP I 367-24
3 8 P A V. SP II 235-4
3 9 P A V. SP II 247-21
4 0 P A V. SP II 40-12
4 1 P A V. SP II 188-13
4 2 P A V. SP I 317-18
44
A c t a L i n g u i s t i c a L i t h u a n i c a L X I X
Page 16
K onstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
FIG. 43. SP I 373–8
FIG. 44. SP I 368–26
FIG. 45. SP II 108–25
FIG. 46. SP II 138–23
FIG. 47. SP I 264–5
FIG. 48. SP I 318–9
FIG. 49. SP II 78–19
FIG. 50. SP II 153–17
FIG. 51. SP I 380–17
FIG. 52. SP I 321–21
FIG. 53. SP I 326–12
FIG. 54. SP II 199–6
FIG. 55. SP I 298–17
FIG. 56. SP II 15–3
FIG. 57. SP I 355–5
FIG. 58. SP II 56–5
FIG. 59. SP II 104–8
FIG. 60. SP II 202–4
FIG. 61. SP I 194–13
FIG. 62. SP II 39–4
FIG. 63. SP II 27–13
2. ORTHOGRAPHY OF THE SERMONS
2.1. The orthography of SP was most strongly influenced by the most important language of scholarship of that period—the Latin language—and only afterwards by Polish. This influence must have taken place in two stages: (1) through the use of the aforementioned languages for scholarly and communicative purposes, their spelling principles were naturally adopted and adapted to Lithuanian; (2) through typefaces—by systematizing, generalizing, combining, or dividing the functions of written signs according to the technical capabilities of printing presses. The first stage is more closely associated with the act of writing itself, the second with adapting a manuscript for print. Thereafter, one may also speak of the influence of printed text back upon the orthographic system (Subačius 2001: 129–152).
Straipsniai / Articles
45
Page 17
V I R G I N I J A V A S I L I A U S K I E N Ė
2.2. Marking long and short vowels
In SP, as in other Catholic writings of the period, long and short vowels were generally not distinguished graphically. Doubling vowels to mark long vowels is not found throughout the entire work. It occurs only sporadically, mostly in the middle of the text of Part II of SP, and is considered an intervention by an anonymous editor: tiiko 142₁₄, prekii 26₂₂, iir 170₉, Karaluus (nom. pl.) 202₁₄, kuu 78₁₀, geruu darbuu (gen. pl.) 175₁₃ (Zinkevičius 1971: 153–167; 1972: 79–100; 1977: 237–244; 1988: 257–259). Infrequently and inconsistently, long vowels may also be marked by <y>: piktyby (acc. sg.) SP I 147₁₅, świełbyby (acc. sg.) SP I 138₁₁, żalty (acc. sg.) SP I 182₉, didy (acc. sg.) SP II 13₂; instances of confusion also occur: dydi (acc. sg.) SP II 206₁₂. In the word ybklaujfiu SP I 198₉–₁₀, <y> most likely marks not [i], but [ii] (together with an added [j]). Most commonly and consistently, <y> is used:
1. In the diphthongs <ay>, <ey>, <uy>: łaymingay SP I 280₁₀, weydu SP I 352₁, żmoguy I 58₂₉; pageydimay SP II 188₈; instances with <i> are isolated: kaip SP I 205₆; funkiei SP I 234₁₃ (in pericopes!). In terms of Sirvydas’s orthography, the latter could be considered proofreading errors. A comparison of SP pericopes and the corresponding passages in JĖ would suggest that Jaknavičius’s hand was most likely involved here. Where the marking of these diphthongs differs between SP pericopes and the corresponding passages in JĖ, JĖ usually has <ai>, <ei>, <ui>.
2. (a) In reduced locative endings in the [i] position (more systematically in SP I): śirdiy SP I 138₂₃, naktiy SP II 66₃₂; ugniy SP I 160₆; żodiy SP I 354₁₄; żmogifłey SP I 138₁₁; pakaũy SP I 370₁₀; in SP II, <i> sometimes already appears in the [i] position, while <y> marks a long vowel: śirdyi 29₁₈, 63₃₁, żodyi 61₅; (b) in the [ii] position in pronominal forms in SP II: Seniey 134₁₃, tikroy tieła 141₁₉, wirefniey 54₂₀, świetifβkiey Vkiniḱay 54₂₁.
<y> is used less consistently:
1. After depalatalized consonants (especially in SP II): prywało SP I 5₁₂, ſtypray SP I 7₂₃; nułynkias SP I 105₂₄, ſtyprafnis SP II 146₁₂, Sżytays SP II 18₈, βyłkuoſe SP II 179₃₀, Żynoio SP II 104₇, niepryimtinas SP II 137₅, βyrdies SP II 169₁₂, żyba SP II 184₁₇.
2. In the singular locative ending: paguldimy SP I 83₃; puody SP II 107₅; raβty SP I 13₅; daykty SP II 98₁₃; <i> also occurs in this position, but infrequently: paguldimi SP I 332₂₀, daykti SP II 112₄, raβti SP I 47₁₆. In SP I we generally find <y>, whereas in SP II <i> is more frequent (cf. kuny 11x in SP I and 2x in SP II; kuni 9x in SP II).
3. In loanwords: żywatas SP II 59₁₈, tyranay SP I 13₂₆, żyday SP I 138₂₆, Synagoga SP I 333₃, herezya SP I 254₁₄.
In SP II, the functions of <y> have narrowed compared with SP I. In many cases, <i> may be used in the same positions as <y>.
46
Acta Linguistica Lithuanica LXIX
Page 18
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
<u> was used to denote both short [u] and long [ū]²⁰. The doubled <uu> denoting a long sound occurs only sporadically in SP II and is of the same kind as <ii>: Karalus (nom. pl.) SP II 202₁₄; geruu darbuu (gen. pl.) SP II 175₁₃. The typeface <ū> with a macron in SP marks omitted tautosyllabic [m] and [n] (see 1.13.).
2.3.1. Notation of [e] after a palatalized consonant
The notation of [e] after a palatalized consonant varies greatly: in SP I, <e> or <ie> was more often written (additionally marking the consonant’s palatalization), while SP II also used <ia> (throughout in some passages of SP II) (Specht 1929: 17)²¹. This would suggest that the dialect of the editors of the SP II section had a more open [e] than Sirvydas’s dialect and more intensive consonant palatalization (Zinkevičius 1996: 100–101). Taking both parts of SP together, the following variants are found:
de / dia-:degina SP I 302₂₀, diaǧina SP II 76₂₀, diagintu SP II 160₂, dešine SP I 241₈, SP II 244₂, diaβne SP II 186₃₀, deβne SP II 229₂₅–₂₆, deftis SP I 370₂₃, SP II 70₁₆, diaftis SP II 170₄;
ge- / gie- / gia-:geria SP I 46₁₆, gieria SP II 50₁₅, giaria SP II 92₁₄, gema SP I 104₂₂–₂₃, SP II 51₁₇, gema SP I 22₃, SP II 123₂₅, giema SP I 106₂₉, SP II 97₂₀, geray SP II 184₂₉, giera (adj. nt.) SP I 284₁₇, gieras SP I 201₁₉–₂₀, giara SP I 172₂₆, SP II 186₃₁, giaras SP II 55;
kie- / kia-:kielo SP I 161₁, SP II 240₃₂, kialo SP II 87₉, kieltis SP I 178₂, SP II 228₁₆, kialtis SP II 158₁₁, kieltuwu SP I 269₃, SP II 157₆, kialtuwu SP II 95₁₃–₁₄, βakiele SP I 311₁₂, βákialu SP II 171₂₀ (there are no words beginning with ke-);
me- / mie- / mia-:metas SP I 331₅, SP II 140₂₈, miatas SP II 169₂₃, medžiu (instr. sg.) SP I 310₂₀, SP II 65 (acc. sg.), miadžiu (gen. sg.) SP II 170₃₀–₃₁, ftameni SP I 103₄, ftamenio SP I 76₉, akmenimis SP I 186₂₃, SP II 45₁₆, akmienimis SP I 377₉;
ne- / nie- / nia-:nuniaβa SP II 188₁₁–₁₂, nendres SP II 99₈, niendri SP II 166₇, niandri SP II 166₁₁, neprietelaus SP I 25₂, SP II 12₃₀, niepretelaus SP II 74₂₀–₂₁;
pe- / pie- / pia- / pea-:péciu SP II 204₂₀, pécio SP I 194₇, piéciu SP II 218₁₆, piácio SP II 205₂, pekłos SP I 275₁₁, SP II 52₈, pieklįβkay SP II 42₂₀–₂₁, piakliβkas
20 Following the Latin-language tradition, [u] at the beginning of a word is usually represented by the typeface <v>. <u> was used at the beginning of words only 8 times and should be regarded as a proofreading error: užgimimo SP II 124₁₆; uolos SP I 270₁; uolas SP I 50₂₁; uoga SP I 321₂₀; uoday SP I 260₁₈; 316₁₂; unt SP II 200₈; 204₁₄.
21 The highest concentration of <ia> was recorded by the author of the article in SP II 166–183. Franz Specht also drew attention to this (1929: 17). However, the first to notice and describe this variation was another German linguist, Richard Garbe, in the 1884 edition of SP I. According to him, <ie> is most often written after the consonants <g>, <k>, <š>, <ž>, whereas it is rarely found after other consonants (Garbe 1884, 13).
Straipsniai / Articles
47
Page 19
V I R G I N I J A V A S I L I A U S K I E N Ė
SP II 206₂₄–₂₅, piakłos SP II 75₃₂, Petras SP I 49₈, SP II 207₁₅, Piatras SP II 17₂₈, Peatras SP II 186₁₁, 207₁;
re- / rie- / ria-: retos SP I 75₇–₈, Riatas SP II 182₂₈, regiedami SP I 19₆, SP II 34₁₀, riegiedamas SP II 48₁₉, riag SP II 177₁₅;
se- / sia-: fenu (acc. sg.) SP I 56₁₃, SP II 123₃–₄, fianu SP 171₂₆;
ste- / stia-: ftebįli SP I 329₂₈, II 71₅, ftiabili SP II 176₁₉;
šie- / šia-: šiesielis SP I 125₂₇, plišieli SP II 27₆, kaŭšielis SP I 314₁₄, šieši SP I 214₁₀, šiaßtu SP I 317₁₀ (pericopes) (there are no words beginning with <še>);
we- / -wie- / wia-: welinas SP I 293₃, SP II 46₁₂, wialinas SP II 77₁₇–₁₈, wede SP II 95₆, wiade SP II 177₃₁, giwena SP I 308₁₃, giwiena SP I 308₂₂–₂₃, giwianimo SP II 169₂₁;
že- / žie- / žia-: žemes SP II 138₄, žiames SP I 12₇, SP II 151₃₂, žiemes SP II 246₁₄, žemibe SP II 216₂₅–₂₆, žiemay SP II 48₂, žiamious SP II 183₂₅, nužemintus SP II 47₁₄, nužiamintays SP II 47₁₇–₁₈, Nužiemindawau SP II 67₃.
In SP I, [e] is represented more consistently, most often by <e> or <ie>; <ia> is concentrated mainly in one defined passage of the SP II text; for the rest, the representation of [e] essentially does not differ from SP I. [k] must have been softened most intensively in the position before [e]—after <k> a mark of palatalization is always written (Palionis 1995: 30). A mark of palatalization is also written after [š], represented by the typeface <š>; the <ß> found in two words was probably intended to denote a velarized [š], or else a proofreading error occurred here: βeymina SP II 181₆ and βeymina SP I 317₅ (pericopes); cf. šiey-min-, written 5 times, mostly in SP I.
2.3.2. Notation of [ai] after a soft consonant
The diphthong [ai] after a soft consonant is represented in four ways: <ey>, <iey>, <iei>, <iay>: pageydimu SP II 140₃₁, pagieydimus SP I 266₁₀, SP II 259₁₀, pagiaudimuołe SP II 178₂₁–₂₂, gieyde SP I 127₃, SP II 27₈, geyde SP II 180₂₃, weydo SP I 90₁₀, SP II 148₄, wiaydo SP II 219₁₉, wieykiey SP I 268₁₀, weykiey SP I 52₂₉–₃₀, SP II 106₂–₃, meyłes SP I 89₂₀, SP II 107₂, miayłes SP II 184₁, funkiey SP I 198₂₇, SP II 88₂₅, funkiei SP I 234₁₃ (pericopes), βeymina SP II 181₆, šieymiu SP I 254₁₁–₁₂. The most frequent variant is <iey>; <iay> is characteristic of certain passages in the text of SP II.
2.4.1. Notation of voiced and voiceless consonants at their junctions
In cases of assimilation, consonants are more often written according to the morphological principle, but no single principle is followed consistently: pabaygtu (4x) SP II 62₂₆–₂₇, pabaiktu SP II 182₁₃–₁₄, àžudingie SP II 18₂₂–₂₃, àžudingt SP I 255₂₆, àžudingta SP I
48
A c t a L i n g u i s t i c a L i t h u a n i c a L X I X
Page 20
Konstantino Sirvydo Punktų sakymų spausdmenys ir rašyba
310₁₇–₁₈, užjudenktu SP II 18₂₇, ažudinktu SP II 40₁₀, ažudinkta SP II 209₃₂–₃₃, apjidenktumite SP II 175₂₁, ažudegtas SP I 144₉–₁₀, prabyngtay SP I 53₂, prabinktay SP I 273₂₁, užjudegtas SP II 76₅–₆, ažudektu SP II 240₁₃–₁₄, ažudirbtu SP I 323₂₅, dirbt SP I 330₁ (4x), dirps SP II 35₂₆, Dirpkite SP II 177₃₂, iżwerβtu SP I 85₁₃, iżweržto SP I 13₁₇. The second-person singular imperative is written twice in SP II: nunuoging 169₂₉; saugog 141₈ (-k + g(i) = g: nunuogink gi?).
The spelling of half-participles is very often based on the phonetic principle: nuleyzdama SP I 333₁₈, pamezdami 258₅, ibrizdami 187₂, drauždami 258₄, žiždamas 213₁₉; fuleyzda-mas SP II 12₈, žayzd amas 58₃, melždami 103₁₃, rauždami 176₁, fkuzdamafis 142₅, melždamies 202₁₀–₁₁.
But examples with <s> also occur: fukdamafis SP II 55₁₅, ibmesdamas SP II 72₁₀,
At the junction of a root and a suffix, two consonants belonging to different morphemes may occasionally be placed side by side without being merged: apjidingkite SP II 132₅, Džiaugkites SP I 326₂, biegkt SP II 172₁₇ (cf. per bekt SP II 217₁₉), nulaužsiu SP I 310₂₈. Against the background of Sirvydas’s linguistic competence, this was most likely intended to reveal the process of verb-form formation obscured by assimilation (cf. iżližios SP I 7₆, užJet SP I 285₅). This may have been one of the proposed spelling practices: keeping morphemes intact and unchanged at their junctions.
It is unclear whether the following isolated instances should be attributed to proofreading errors or to attempts at spelling standardization (hypercorrections): ungβtu SP II 104₁₂, lisdus SP I 305₁, mafgot SP II 173₅ (cf. mazgofi 173₁₀ on the same page).
Overall, consonants are more often written phonetically in SP II.
2.4.2. Notation of [š] and [s] before [č]
Both [š] and [s] in the position before [č] were not always represented in accordance with their origin (cf. tuščias and skai tus), and their notation varied greatly. In addition, the affricate [č] was also represented differently in the resulting consonant cluster. Evidently, rendering these consonants with written signs must have posed problems for the authors and book publishers of the time, leading to considerable variation. The following variants are found in SP:
[š] in the position before [č]: <fc>, <lć>, <lċ>, <sć>, <ść>, <ść>, <śċ>, <lcз> <lcзć>, <βć>: karfciomis SP I 232₁₃, tufcias SP II 40₁₂–₁₃, karfciaus SP II 50₁₆, kumfciomis SP II 199₂₃, Krikfcionis SP II 202₁₀, tewikfcion SP II 187₂₈, karfčiuofe SP I 244₁₇, krikfcionie SP I 162₁₀, tewikfčiu SP I 217₂₈, tufčiu SP II 156₄, krikfčioniu SP I 320₁₁–₁₂, waykčioia SP I 351₄, paukčiey SP I 308₁₂, paukśiey SP I 311₄, karśčiaus SP I 65₂₂, tusčias SP I 44₂₂, krikścionimus SP I 360₂₁, tewikścion SP I 343₇–₈, paukś-cio SP I 107₁₈, krikścionis SP I 112₁₆, paukśiey SP I 189₂, tusčiey SP II 6₂₇, krikś-
Straipsniai / Articles
49
Page 21
VIRGINIJA VASILIAUSKIENĖ
čionis SP I 239₁₈, SP II 179₃₀, tuščio SP I 244₁₉, tułczus SP II 202₁₃, namikfčėiey SP II 186₉, brikfčionie SP II 232₂–₃, karbčio SP I 329₁₇; [s] in the position before [č]: <lc>, <lć>, <sć>, <ść>, <sć>: rupefčiu SP I 134₂, rufcioy SP I 105₂₇, fkayfčiaufi SP II 271, karalifciu SP I 134₁₇ (cf. DP: rupefčiu 16₃₄, ižgafcio 13₂₀), rupefčiu SP I 248₄–₅, fkayfčiaufia SP I 211₂, karalisčiu SP I 132₁₄, rusčiaufia SP I 272₂₉, rupesčiu SP I 144₂₇, SP II 62₁₀, ižwerščiey SP II 14₃, wagiščiu SP I 147₁₁, karališčiu SP II 154₁₅–₁₆, fkaiščiaus SP II 38₂₂, karališčiu SP II 14₁₆–₁₇.
The most likely reasons for these spellings have already been identified: inconsistent use of the same typefaces with and without diacritics (see 1.5.4.) and the absence of more definite orthographic norms. In representing the consonants mentioned, neither the morphological nor the phonetic principle was followed consistently. Other causes should not be ruled out in this case: this notation may have reflected features of Vilnius speech—features that were regional or arose from language interference, and that involved fricative consonants and affricates.
2.4.3. Notation of [š], [ž] at the junctions of the prefixes iš-, iš-, už- with other morphemes
The spelling of the prefixes iš-, iš-, už- at junctions with other morphemes (especially with the reflexive particle si) varies: i{ž}liveržtu SP II 48₂₅, i{ž}taria SP II 5₁₃, išpa{ž}ini-mo SP II 172₄, ižleyzdawo SP II 28₂₂–₂₃, ižtrinta SP II 163₂₃, iβtrint SP II 231₂₈, Ižjim{á}no SP II 83₇, iβsimano SP I 332₂₃, iɪmano SP I 228₂₃, i{ž}naykinta SP II 154₄, iβnaykie SP I 185₂₁–₂₂, iβfimano SP I 305₁₈, iβfikalbet SP II 53₁₅, iβluoia SP II 72₉; ižlipilde SP II 220₁₂–₁₃, ižagtas SP I 174₂, iſſipildis SP I 1₁₈, ižizios SP I 7₆, if{ž}iotu SP II 69₂₅–₂₆, i{ž}metimo SP II 61₁₉–₂₀, ižmeft SP II 56₃, iβmes SP II 57₃, vžfet SP I 285₅, vži̇kielunčio SP I 310₃₀, vžtekiet SP I 347₇, vž{ž}ingг SP II 234₃₂, vſimatis SP II 177₁₂, vſimeftu SP II 176₈, VJžwefdamas SP II 37₆, vβžerſi SP I 349₂. Because of the spelling variation, it is not always clear which verb is being used: įstatyti or išstatyti:
From one’s own folly, the devil leads a person into sin SP I 232₃–₅. For he leads [them] into the Church, as if into a kingdom of some kind SP II 151₁₇–₁₈.
SP clearly makes an effort to represent the prefixes under discussion consistently in different positions, but variation nevertheless persisted, so their spelling cannot be called fully consistent (Palionis 1995: 30). The examples show that distinguishing and recording these prefixes was most difficult where they merged with the reflexive particle si. Even then, however, the variant most often chosen was užsi-. SP uses two forms of the prefix iš-: iż-, iż-, iż-, iż, iz-, iż-, iż- (with <ž>) or isz-, iβ- (with <š>). The choice of one of these depends on the beginning of the root
50
Acta Linguistica Lithuanica LXIX
Page 22
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
in terms of phonetic expression: isz-, iβ- is not written before voiced consonants, whereas iž- can be found in almost all positions (not found before ey, e: IS3eio SP 350₂₉, iβeyt SP I 39₁₅, SP II 121₅, iβeymay SP I 351₁₁, iβeykite SP II 168₈, Iβeiau SP II 242₂₁), so the spelling of some words beginning with a voiceless consonant or semivowel varies: ižkaktu SP I 352₁₅, ižkako SP II 243₂₂, ižkafa SP I 208₂₄, SP II 209₂₆, iβkafa SP I 208₂₀, ižmanimu SP I 251₂₄, iβmanimu SP I 94₁₉, ižmintis SP II 215₃–₄, iβmintis SP I 63₂₂, SP II 156₁₉, ižreyβkia SP I 128₁₂, iβreyβkia SP I 82₁₈, SP II 12₁₂, ižfakit SP I 159₈, iβfakit SP I 161₂₈, ižkayto SP I 126₁₄, SP II 149₁₃ iββkayto SP I 333₂₇, SP II 122₁₆, ižwidis SP I 129₂₇, SP II 18₂₄, iβwidis SP I 209₂₈. iβ- is found much more frequently than iž-... before semivowels. In the case of the preposition iš, position is irrelevant—it is systematically written with <ž> in SP: iž, iz, iž, iž, iz, iž, iž. Three occurrences with iβ (SP I 95₂, 310₂₅, SP II 172₂₄ (pericopes)) should be regarded as accidental.
2.4.4. Marking [n] before [b], [p]
The form of the prefix im- is used fairly consistently before [b] and [p]: imbers SP II 107₉, impuola SP I 108₂₀ (23 occurrences in total in the forms of the verbs ipulti and ipuldinėti); only once is inpuolys written.
2.5. Consonants, digraphs, and geminates of foreign origin
<f> and <h> are used in international words, while <x> is used in Lithuanian words and loanwords to represent the consonant cluster [ks]: affektays SP I 294₁₅, Iozafu SP I 181₂₃, heretikay SP I 204₁₁, hijtoríu SP I 205₁₃, herefiju SP I 251₁₂, holofe SP I 255₂, džiauxmu SP I 201₁₂, linxmibes SP II 97₂, paduxi SP II 99₉, uxnoy SP II 243₃, uxufo SP II 163₁₅, auxi SP II 36₂₄, axtinays SP II 63₂. <Ch> is used in non-Lithuanian proper nouns—personal names and ethnonyms: Chrijtus SP I 367₂₆; SP II 242₆, Chaldaycзykų SP II 168₇–₈, Chaneneus SP II 7₁₄, Cherubinay SP II 197₁₆–₁₇, Chrijtoftomas SP II 47₁₀–₁₁. <C> and <c> are written instead of <k> in loanwords from Latin or words that entered via Latin: cedrus SP I 141₅, Casus SP I 306₂₁, cometa SP II 140₂₁, Cyrillus SP II 93₇. The letter <q> is also used in Latinisms: queftiiu SP II 306₈. In international words based on Latin and Greek, <Th>, <th>, and <rh> are written: theologay SP I 238₂₆, katholiku SP I 226₆, Thomas SP I 99₇, Theophilaktus SP II 178₃₁, myrrhu SP I 225₂₅. For the same reason, <y> may be used or consonants may be doubled: tyrannu SP II 146₂₇ (cf. Lat. tyrannus), affektays SP I 294₁₅ (cf. Lat. affectus). The geminates <fs> and <ff> in the present-tense forms of the verb būti probably also arose under the influence of Latin.
Straipsniai / Articles
51
Page 23
VIRGINIJA VASILIAUSKIENĖ
the verb esse: efsi (27 occurrences), effi (15 occurrences), efsiby SP I 17211, effuntiemus SP I 27817, efsigu SP I 8214. Foreign-origin words also contain other geminates (see the previous examples in this subsection)22.
2.6. Writing words together and separately
In the absence of grammars and the norms codified in them, whether words were written together or separately was determined by several internal and external factors: (1) prosody; (2) the example of other languages; (3) technical features of printing and page layout, and the linguistic competence of printers and publishers. Much of the spelling in SP follows from an intuitive sense of the intonation of one’s own language and its variations, as well as nuances of meaning. Thus, in the absence of explicit rules, word combinations are quite often written as one word, while monosyllabic words lacking independent stress (enclitics and proclitics) are naturally attached to longer words possessing intonational and semantic autonomy. Intonational pauses, word order, and contrastive stress are chosen more consciously and are more closely connected with the rhetoric of sermons.
Adverbs or adverbialized combinations may be written as one word when formed from a preposition and an oblique case (1); a pronoun and an oblique case (2); the preposition be and a pronoun (3); or combinations of pronouns, and of pronouns and particles (4):
(1) In SP I, be is written together (with one exception) in combinations expressing quantity: begalo (14 times), be galo (once) SP I 20414, bemieros SP I 1711–2, beikayčiaus SP I 31326; in SP II, by contrast, be galo (8 times) is written as two words (most such cases are concentrated on pp. 211–218 of SP II). In combinations expressing quality and location: ižtolo SP II 18718; ižkur SP II 18420.
(2) In SP I: tuometu (15 times), tuo metu (9 times); in SP II: tuometu (2 times), tuo metu (17 times).
(3) Ne turi įbmonios / ney luofo noro / a beto ne deftis ne vienas nufideimas SP I 28321.
(4) kunorint SP II 15132, kokinoris SP I 8829, kokiaf=norint SP II 19830, iogipaties (i.e. jo gi paties) SP II 2068, kurys giwienas (i.e. kuris gi vienas) SP II 8631.
The preposition be is more clearly distinguished in SP I from the second component be of the compound particle nebe:
Not without mockery or temptation did she leave of her own accord SP I 9714. For, no longer finding a pretext, he did not speak of the Gospel; he cast out the devil SP II 616.
22 We should regard darīsime SP I 1409 as a rare, inconsistent spelling. This word occurs in a quotation from the Vulgate (Luke 3:14). Therefore, some unclear and complicated questions concerning SP spelling can be answered more precisely only once quotations (which may make up 20–25 percent of the entire SP text) have been distinguished from the authorial text and the two textual layers have been examined separately.
52
Acta Linguistica Lithuanica LXIX
Page 24
Konstantino Sirvydo Punktų sąkymų spaudmenys ir rašyba
In the dual forms of pronouns, the numeral du is not fused into a single word with the pronouns:
The two disciples, speaking about the sufferings of the Savior, entreated Him, that the Lord might come to them SP II 240₇. Thus, when these two disciples listened to the Lord expounding the Scriptures, they were enlightened and believed SP II 240₂₆.
It is assumed that writing ne separately from verbs arose under the influence of Polish (Palionis 1995: 31). However, ne is not written separately with all verbs, much less with all verb forms (especially passive participles), with equal frequency. Moreover, there are quite a few instances of variation here, for example, SP I: negal (52x), ne gal (19x), negali (4x); SP II: negal (7x), ne gal (32x), negali (1x), ne gali (3x); SP I: nežino (15x), ne žino (6x); SP II: nežino (3x), ne žino (1x); ne kinčia SP II 216₂, ne kinčia SP I 117₇; nebiokites SP I 87₂₈, ne biodamies SP I 64₇. Ne is written together with reflexive verbs: nefirado SP I 49₆, nefileyde SP II 242₁₅. On one occasion ne was written separately, most likely by mistake: ne įf̄augo SP I 372₁₄. With passive participles functioning adjectivally, ne is more often written together than with finite verb forms: neižagtas, neižāgtas (SP I 2x; SP II 1x); ne i(n)žagtas (SP II 2x); neįßlakitas, neižßlakitas (SP I 12x; SP II 2x); ne ibßlakitas, ne ižßlakitas (SP I 7x; SP II 7x); SP I: neful̄twertas (4x), ne futwertas (2x).
Ne is written separately from verbs more often in SP II. With some verbs ne is more often written together, while with others it is written separately, without any clearer rationale.
In SP II, prepositions and prefixes can be confused with one another: prepositions are written together with the words that follow them (1), while prefixes are written separately from the rest of the word (2):
Instances of prepositions written together also occur in SP I (ižkuriu 153₂₉, pawundeni-mis 269₂₈), but they are much more concentrated in SP II; moreover, the reverse phenomenon is also recorded here—that is, prefixes written separately.
One of the reasons for confusing prefixes and prepositions should apparently be considered the printers’ and editors’ inadequate command of Lithuanian and their linguistic competence in general. Nor should we forget the fact that SP II was prepared for printing by someone other than the author himself.
Other reasons are technical:
(1) a change in the word-break point when setting the text for printing from the manuscript (the location of the hyphen changes because the lines do not correspond):
Straipsniai / Articles
53
Page 25
VIRGINIJA VASILIAUSKIENĖ
fu=manim SP II 22923, iš=giditoiu SP II 22319, I=eyk SP II 2588, I=eyt SP II 8520, fu wieninta SP II 1137 and so on (see also those listed above).
(2) uneven spacing between separate words on the same line; it is not always possible to determine whether there is a space between words or not; sometimes there is insufficient room for a hyphenation mark on a narrow line.
In most cases, words retain their stable structures throughout the SP text and are not fused with the other words surrounding them. Rendering compounds as a single word is often motivated by lexicalization (begalo, ištolo, tuometu, etc.). As already noted, some rules for writing words together or separately were adopted from Polish, but in this case it would be very important to check for a possible influence of Latin as well, since that influence is unquestionable in certain other areas of spelling and typeface use. This question can be answered more precisely only by excluding from the SP text biblical and other quotations translated from Latin and examining their spelling features separately.
2.7. Use of capital letters
The Latin and Polish traditions of capitalization were most often followed. Proper names and titles begin with a capital letter, and biblical names are used abundantly. Moreover, following the example of Latin and Polish, ethnonyms, various religious denominations and movements are written with capital letters:
Murinas SP I 29420, Tatariev, Turkay SP I 3468, Ziday SP I 4812, Zyday SP II 9322, Amonitu SP I 3529, Afsironis SP I 17015–16, Aegipcianus SP I 26020, Aegypčionu SP II 2084–5, Chaldavczyku SP II 1687–8, Philitinų SP I 22619, Korinthianump SP I 475–6, Medanus SP II 15022–23, Madionitu SP II 91, Niniwityav SP II 5718–19, Pharizeußav, Lewitas SP I 8123, Arionay, Eutichionay, Monothelitay SP I 20514–15, Neftoriany, Lut-heranu SP 2901–2, Luteru ir Kalwinu SP I 25121.
The stylistic use of capital letters is not always consistent. Capitalization is most consistent in the word Dievas (‘God’) (the four instances of lowercase spelling should be considered proofreading errors). The word is intentionally written with a lowercase letter when referring to pagan deities:
Teyp ir Ziday telu iš aukfo Jaw nuleii diewu ii dare SP I 7229.
In both parts, there is greater variation in the spelling of the word bažnyčia (‘church’): SP I: Bažnićia 63x, bažníčia 57x; SP II: Bažníčia 28x, bažníčia 5x.
Words are more often written with a capital letter for stylistic reasons or by chance in SP II. For example, in SP I the nouns kunigas (‘priest’) and kunigaikštis (‘duke’) are written with lowercase letters,
54
Acta Linguistica Lithuanica LXIX
Page 26
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
In SP II, kunigas ‘priest’ is already written with a capital letter in a quarter of cases, while kunigaikštis ‘duke’ is more often written with a capital than a lowercase letter.
Similarly, the spelling of Velykų, velino also varies in SP II (in SP I, velinas is written only with a lowercase letter). Velinas is written with a capital letter in quotations from John Chrysostom (Homiliae LV in Acta apostolorum. XLI; PG 60: 293) (1) and the First Letter of Peter (1 Pet 5:8) (2):
(1) Nufideimas piktefnis / ira negi Welinas / nes welinas daro žmones nužemintus. Er ne regite apfleftuiu / kad pafweykfla ába luofays nuog pikto Welino tumpta kaip ira nužamintays SP II 4712–17. Daemone gravius est peccatum; ille namque humiles reddit. Annon videtis daemoníacos, quando a morbo recreantur, quam sint demissi.
(2) Neprietelus / iufų Welinas / kayp lewas rekdamas fukafi / iefbkodamas ku praritu SP II 25629–32.
The word demonas is used by John Chrysostom in the Greek original; it is also retained in the Latin translation of this work.
In the adjacent quotations in Polish, this word is also written with a capital letter. However, in the first quotation the word welinas, additionally introduced by the translator (it is absent from the Greek and Latin texts), is treated differently: it is translated into Polish not as Dyabel but as czárt, while in the Lithuanian quotation it is written with a lowercase letter; cf.:
Grzech / gorby ieft / niž Dyabel / bo czárt czyni ludzie pokorne. A zá nie widzićie opełtánych / gdy do zdrowia przychodzą / y wolnemi od Dyabłá ftáją się / iáko są pokorni SP II 4711–15.
SP I uses capital letters more systematically, consistently, and less frequently than SP II. The use of capital letters in the text of SP II seems in some cases to be random, and their use was not fully regularized during the publishing process:
iop priſigrįbt / Nes ne wieno daykto futwerto ne regieio SP II 1143–4; kuríu Dáwe CHRyſtufas taridamas SP II 926–7; Ir todrin ſtebiſi S. Auguſtinas / Krikſčionies to durnibey SP II 714–6; iž kalno aba iž Pilies prieβ neprietelus muβáſi SP II 1924–26.
CONCLUSIONS
1. The typeface sets adapted for Latin and Polish texts could not fully reflect Sirvydas’s orthographic conception. They had to be adapted to the capabilities of the printing house. For this reason, in discussing the writing signs used in SP, priority was given to the concept of a typeface. The aim was to avoid equating handwritten and printed graphemes, since the same typeface could denote several handwritten graphemes, and vice versa.
Straipsniai / Articles
55
Page 27
VIRGINIJA VASILIAUSKIENĖ
Diacritic-bearing typefaces are used very extensively in DP, but their variety is greater in SP. SP also uses more and more varied diacritic-bearing typefaces than JE. Approximately 51 typefaces bearing diacritics of various forms are found here. Most of them are very rare and sporadic, although some are used consistently, if infrequently. Only 9 diacritic-bearing typefaces are used frequently: <ć>, <ł>, <ś>, <ź>, <ž>, <á>, <à>, <č>, <ȥ>. The variety of diacritic-bearing typefaces is greater in SP I, while their frequency is greater in SP II because of <á>. The typefaces <è>, <š>, <ź>, <ž>, <ž̄> occur only in text set in Latin antiqua and italic. The form of the diacritics varies considerably on <ź>, <ž>, <ȥ>, <ž> or <ž>. The greatest variation and uncertainty occur in the representation of fricative consonants and affricates. Here, typefaces both with and without diacritics are confused. On the basis of research on SP alone, it is currently difficult to say whether this confusion arose for technical reasons (carelessness in typesetting, print quality, treatment of diacritics), or whether it reflected certain linguistic phenomena and processes taking place in the language (language interference, traces of a particular dialect).
In SP II, <á> is very often used in an unstressed syllable. The circumflex-shaped diacritics <â>, <ê>, <î>, <ô>, and <û> may have marked a stressed syllable or served to distinguish words and their forms. They are used too infrequently for broader conclusions to be drawn. The typefaces <ę> and <ų> entered the Lithuanian text of SP accidentally and did not represent nasal sounds. At times, <ę> serves the same function as <ē>. The typeface <ł> quite consistently denotes a hard or depalatalized [l]. Following the model of Latin, vowel typefaces bearing dash- and circumflex (tilde)-shaped marks consistently indicated omitted tautosyllabic <m> and <n>. Some typefaces used to abbreviate word endings were also adopted from Latin.
Because such uses are rare, overlap with other functions, and lack consistency, it is impossible to isolate and define more precisely the cases in which diacritic-bearing typefaces were used to mark stress.
At that time, a diacritic was probably not always considered an essential distinguishing feature of a typeface.
In the main SP text printed in Gothic type (Schwabacher), a single typeface is used to represent uppercase <I> and <J>. In antiqua italic, three different typefaces represent <I> and <J>. <j> is used sporadically on SP II, pp. 1–47.
2. The most significant influence on SP orthography came from the most important scholarly language of the period—Latin—and only afterwards from Polish.
Vowel length in SP was not represented graphically. <y> is often used in the position of <j>: almost systematically in the diphthongs <ay>, <ey>, <uy>; it is also found after depalatalized consonants, in reduced locative endings, and in loanwords. <u> represents both a short and a long vowel. Doubled typefaces to indicate long vowels occur only a few times in SP II.
56
Acta Linguistica Lithuanica LXIX
Page 28
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
The notation of [e] after a palatalized consonant varies considerably: in SP I, <e> and <ie> are written, while SP II also has <ia> (the greatest concentration of <ia> is on pp. 166–183 of SP II). [ai] is most often represented by <iey> and <iay>, and less often by <ey> and <iei>.
The notation of consonants at the junctions of voiced and voiceless consonants varies: overall, the morphological principle is used more often; in SP II, the phonetic principle is applied more frequently than in SP I; participles are quite often written according to the phonetic principle.
The notation of [š] and [s] before [č] varies particularly greatly. This variation may have arisen from the previously mentioned unsystematic use of diacritics, or it may also have been conditioned by certain changes in the system of fricatives and affricates in the pronunciation of Vilnius residents of that period.
The spelling of the prefixes iž-, iš-, and už- also varies at their junctions with other morphemes. It was most difficult to preserve the prefix form unchanged next to the reflexive particle si. Two forms of the prefix iš- occur in SP: iž-, iž-, iš-, iš, iz,- iš-, iż- (with <ž>) or isz-, iƀ- (with <š>). Variants with <š> are never used before voiced consonants, whereas variants with <ž> are possible in all positions. The position of the prefix iš is neutralized, and it is written with <ž> in every case.
Consonants of foreign origin <f> and <h> are used in international words, while <x> is also used in Lithuanian to denote the cluster [ks]. Following Latin usage, <c>, <th>, <rh>, <fs>, and <ff> (in the present tense of the verb būti) may also be used.
The joining or separation of words in SP was determined by several internal and external factors: (1) prosody; (2) the example of other languages; (3) technical aspects and form of printing, and the linguistic competence of printers and publishers. Most words retain their stable forms throughout the SP text and are not run together with the words surrounding them. Rendering clusters as a single word is often motivated by lexicalization. The joining of prepositions to other words, and the separation of prefixes in SP II, could be associated with the linguistic competence of printers and publishers. In SP I, spacing between words varies particularly greatly.
Ne is more often written separately from verbs in SP II. With some verbs or their forms, ne is more often written together (especially with passive participles), while with others it is written separately, without any more specific motivation.
Capitalization followed Latin and Polish orthographic traditions. In addition to the cases familiar to us, SP capitalizes ethnonyms, the names of various religious movements and denominations, and—though not entirely consistently—the most important religious concepts. Stylistic capitalization varies. Capital letters are used more consistently in SP I. SP II contains a considerable number of proofing errors and slips. In some cases, Sirvydas’s use of capital letters was taken over from the Greek or Latin source he was citing.
Straipsniai / Articles
57
Page 29
V I R G I N I J A V A S I L I A U S K I E N Ė
SOURCES AND LITERATURE
A l e k n a v i č i e n ė O n a 2006: [Review:] On the Old Printing Types of Lithuania. – Knygotyra 46, 279–294.
THE BIBLE, or THE HOLY SCRIPTURE. Ecumenical edition. Old Testament translated from Hebrew, Aramaic, and Greek by Antanas Rubšys; New Testament translated from Greek by Česlovas Kavaliauskas. Vilnius: Lithuanian Bible Society, 1999.
DP – Poltilla CATHOLICKA. Tái éłt: Iżguldimas Ewangelìu kiekwienos Nedelos ir ßwetęs per wiłłús metús. Per Kúnigą MIKALOIV DAVKSZĄ Kanonîką Mėdnikų, iż łėkibło pergūldita. Su walá ir daláīdimu wîreuliuių. W Wilniui / Drukárnioi Akadėmios SOCIETATIS IESV, A.D. 1599.
G a r b e R[i c h a r d] 1884: Szyrwid’s punkty kazań: (punktay sakimu) vom Jahre 1629: mit einer grammatischen Einleitung. – Göttingen: Vandenhoeck & Ruprecht.
G e l u m b e c k a i t ė J o l a n t a 2003: Pirmas lietuviškas pamokslų rinkinys – Wolfenbutelio postilė (1573). Rankraščio kritinio komentuoto leidimo principai ir tyrimo strategija. – Archivum Lithuanicum 5, 51–96.
G e l u m b e c k a i t ė J o l a n t a 2009: Lietuvių kalbos ortografijos reformos (Mažoji ir Didžioji Lietuva). – Lietuvos Didžiosios Kunigaikštystės kalbos, kultūros ir raštijos tradicijos. Vilnius: Lietuvių kalbos institutas, 17–27.
JE – POLISH AND LITHUANIAN GOSPELS, both for Sundays and for all the Feast Days observed throughout the year in the Catholic Church according to the Roman rite. Published with the permission of the [elders. [...] in VILNIUS. At the Printing House of the Academy of the SOC. IESV, in the year / 1647.
J u d a M a r i a 2001: Pismo drukowane w Polsce XV–XVIII wieku. – Lublin: Wydawnictwo Uniwersytetu Marii Curie-Skłodowskiej.
K a b a š i n s k a i t ė B i r u t ė 2005: Diakritiniai ženklai Baltramiejaus Vilento Enchiridione ir Evangelijose bei Epistolose (1579): <ũ>, <ô>, <ò>, <ó>. – Archivum Lithuanicum 7, 45–66.
K o ž e n i a u s k i e n ė R e g i n a 1990: Prefaces and Dedications of the Sixteenth–Eighteenth Centuries. Vilnius: Mokslas.
LK – KATHECHISMAS ABA PAMOKIMAS WIENAM KVRIAMGI KRIKS CZONIVY RAYKIAMAS PARASZITAS Nuog D. IOKVBO LEDESMOS Theologo Soc: IESV. E pergulditas Lietuwißkay / ir Vntrnkárt ißßáultas WILNIVY Drukórnioy Acadėmios SOCIETATIS IESV, Metūlu vźgimimo Wießpatės / 1605.
L u č i n s k i e n ė M i l d a 2005: Jonas Jaknavičius ir 1647 metų katalikiškos Evangelijos lietuvių kalba. – Jono Jaknavičiaus 1647 metų „Ewangelie polskie y litewskie“, par. Milda Lučinskienė. Vilnius: Lietuvių kalbos instituto leidykla.
58
A c t a L i n g u i s t i c a L i t h u a n i c a L X I X
Page 30
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
Palionis Jonas 1967: Lietuvių literatūrinė kalba XVI–XVII a. Vilnius: Mintis.
Palionis Jonas 1995: Lietuvių rašomosios kalbos istorija. Vilnius: Mokslo ir enciklopedijų leidykla.
Palionis Jonas, Strockis Mindaugas 2005: Jono Rėzos 1625 m. psalmy no kirčiavimas. – Baltistica 40(1), 31–44.
Petreikis Tomas 2011: Marginalijos kaip knygos istorijos šaltinis: istoriografinis aspektas. – Knygotyra 57, 67–85.
PG – PATROLOGIA GRAECA 60: S. P. N. JOANNIS CHRYSOSTOMI, ARCHIEPISCOPI CONSTANTINOPOLITANI, OPERA OMNIA QUAE EXSTANT, VEL QUAE NOMINE CIRCUMFERUNTUR, AD MSS CODICES GALLICOS, VATICANOS, ANGLICOS, GERMANICOSQUE; NECNON AD SAVILIANAM ET FRONTONIANAM EDITIONES CASTIGATA, INNUMERIS AUCTA; NOVA INTERPRETATIONE UBI OPUS ERAT, PRAERFATIONIBUS, MONITIS, NOTIS, VARIIS LECTIONIBUS ILLUSTRATA; NOVA SANCTI DOCTORIS VITA, APPENDICIBUS, ONOMASTICO ET COPIOSISSIMIS INDICIBUS LOCUPLETATA OPERA ET STUDIO D. BERN. DE MONFAUCON, MONACHI BENEDICTINI E CONGR. S. MAURI. EDITIO NOVISSIMA, IIS OMNIBUS ILLUSTRATA QUAE RECENTIUS TUM ROMAE, TUM OXONII, TUM ALIBI, A DIVERSIS IN LUCEM PRIMUM EDITA SUNT, VEL JAM EDITA, AD MANUSCRIPTORUM DILIGENTIOREM CRISIM REVOCATA SUNT, ACCURANTE ET DENUO RECOGNOSCENTE J.-P. MIGNE, BIBLIOTHECAE CLERI UNIVERSAE, SIVE CURSUUM COMPLETORUM IN SINGULOS SCIENTIAE ECCLESIASTICAE RAMOS EDITORE. TOMUS NONUS. VENEUNT 18 VOLUMINA 160 FRANCIS GALLICIS. EXCUDEBATUR ET VENIT APUD J.-P. MIGNE EDITOREM, IN VIA DICTA D’AMBOISE, OLIM PROPE PORTAM LUTETIAE PARISIORUM VULGO D’ENFER NOMINATAM, SEU PETIT-MONTROUGE, NUNC VERO INTRA MOENIA PARISINA. 1862.
SP I – POINTS OF SERMONS from Advent until Pentecost / in the Lithuanian language, with an explanation in Polish, BY Reverend KONSTANTY SZYRWID / Theologian of the Society of Jesus / PUBLISHED WITH THE PERMISSION OF THE ELDERS. In VILNIUS, at the Printing House of the Academy of the Society of Jesus, in the Year M. DC. XXIX.
SP II – POINTS OF SERMONS FOR LENT, in the Lithuanian language By Reverend Father CONSTANTY SZYRWID, Theologian of the Society of Jesus, composed. And now translated into Polish / and published in both languages. With the permission of the Elders. In Vilnius / At the Printing House of the Academy of the Society of Jesus. In the Year of Our Lord, 1644.
Specht Franz 1929: Šyrwids Leben und Schriften. – Šyrwids Punktay sakimu (Punkty kazan), Teil I: 1629, Teil II: 1644, litauisch und polnisch mit kurzer grammatischer herausgegeben von Dr Franz Specht, Professor an der Universität Halle-Wittenberg. Göttingen: Vandenhoeck & Ruprecht, 7–61.
Strockis Mindaugas 2004: Graikų ir lotynų kirčiavimo įtaka lietuvių kirčio žymėjimui. – Baltistica 39(2), 289–307.
Straipsniai / Articles
59
Page 31
V I R G I N I J A V A S I L I A U S K I E N Ė
S u b a č i u s G i e d r i u s 2001: The influence of printers on spelling and its standards. – Archivum Lithuanicum 3, 129–152.
S u b a č i u s G i e d r i u s 2003: The graphemes <I> and <J> in Lithuanian texts of the 18th–19th centuries. – Archivum Lithuanicum 5, 227–262.
ŠPS – Šyrwids Punktay sakimu (Punkty kazan), Teil I: 1629, Teil II: 1644, litauisch und polnisch mit kurzer grammatischer herausgegeben von Dr Franz Specht, Professor an der Universität Halle-Wittenberg. Göttingen: Vandenhoeck & Ruprecht, 1929.
Z i n k e v i č i u s Z i g m a s 1971: On the genesis and language of K. Sirvydas’s “Punktai sakymų”. – Baltistica 7(2), 153–167.
Z i n k e v i č i u s Z i g m a s 1972: The eastern Lithuanian literary language of the 17th century, its origin and disappearance. – Baltistica 8(1), 79–100.
Z i n k e v i č i u s Z i g m a s 1977: On the origin of the Lithuanian literary language. Baltistica 13(1), 237–244.
Z i n k e v i č i u s Z i g m a s 1981: Sirvydas or Širvydas? – Baltistica 17(1), 28–41.
Z i n k e v i č i u s Z i g m a s 1988: History of the Lithuanian Language 3. The Language of Old Writings. Vilnius: Mokslo.
Z i n k e v i č i u s Z i g m a s 1996: The History of the Lithuanian Language. – Vilnius: Mokslo ir enciklopedijų leidykla.
Typefaces and Orthography in Konstantinas Sirvydas’ Book of Sermons “Gospel Points”
S U M M A R Y
Typefaces available at the Vilnius university printing house could not comprehensively reflect Sirvydas’ orthographic conception. They were mostly adapted for printing books in Latin and Polish. Sirvydas had to bear in mind the technical limitations of the printing house. Considering this we use the noun „typeface“ to differentiate printed from handwritten letters when speaking about graphemes in SP. The same printed letter type could represent a few different handwritten letters, and the same handwritten letter could be printed using different typefaces.
In the oldest books printed in Lithuanian the variety of diacritics and their frequency of use differ. Diacritics are used very often in Daukša‘s Book of Sermons, however their variety is larger in SP.
SP orthography was mostly influenced by Latin which was considered the standard basic scientific language at that time. The influence of Polish was less significant.
There are more different typefaces used in SP than in Jonas Jaknavičius’ “Evangelie Polskie y Litewskie”: 51 typefaces with different diacritic marks are used in this book. Most of them
60
A c t a L i n g u i s t i c a L i t h u a n i c a L X I X
Page 32
Konstantino Sirvydo Punktų sakymų spaudmenys ir rašyba
were rarely used, but in spite of that their usage was consistent. Only 9 diacritics appear very often: <č>, <ł>, <ś>, <ź>, <ż>, <á>, <à>, <ċ>, <ž>. The variety of diacritics is more significant in SP I, but we find them more frequently in SP II (mostly because of the especially frequent <á>). <è>, <š>, <ź>, <ż>, <ž> are only used in texts that were typeset in Latin antique. Shapes of diacritics on letters <ź>, <ż>, <ž>, <ż> or <ž> vary significantly within the whole text. Most inconsistencies and confusion were found in marking fricative consonants and affricates. Letters with and without diacritics alternated with each other with no apparent reason. It is difficult to say only using data from SP if inconsistencies were caused by technical peculiarities of printing or reflected deeper processes in language such as interference or some traces of different dialects.
<á> was found most frequently in an unstressed syllable. <â>, <ê>, <î>, <ô> and <û> could mark stressed syllables or serve as a means of differentiating word forms. However, these diacritics appear too rarely to allow one to make definite conclusions about their real functions. <ę> and <ų> did not mark nasal sounds and were accidental in that part of the SP text printed in Lithuanian. In some cases <ę> is used as <ê>. <ł> consistently marks the hard or depalatalized consonant [l]. Vowels with diacritics in the form of a dash and tilde were used to mark an omitted <m>, <n> as in Latin. A way of marking shortened endings was also taken from Latin.
At that time some printers probably did not pay much attention to whether a type had a diacritic or not.
In the basic SP text that was printed in the schwabacher typeface, only one letter was used for marking both <I> and <J>. Three different letters were used during printing for marking both the capital letters <I> and <J> in those parts of the text that were in antique cursive. A small letter <j> was occasionally used in SP II, p. 1–47.
2. The long vowels were not differentiated in any way from short ones in SP. <y> was used as follows: (1) for marking <j>; (2) in diphtongs <ay>, <ey>, <uy>; (3) after depalatalized consonants; (4) in reduced endings of the locative case. <u> was used to mark both a short and a long vowel. Geminates for marking long vowels, occur only a few times in SP II.
The way of marking an [e] that follows a palatalized consonant varies: in SP I it is marked as <e> or <ie>, and in SP II also as <ia> (the largest concentration of <ia> is found in SP II, p. 166–183). [ai] is mostly marked as <iey> and <iay>, sometimes also as <ey> and <iei>.
Marking of voiced consonants before voiceless, and voiceless before voiced often vary, but the morphonological principle was used most often in the spelling. In SP II consonants are marked according to the phonological principle more often than in SP I, especially in writing half-participles.
The most inconsistent way of marking [š] and [s] is when they precede [č]. Such inconsistencies might be caused by the unsystematic usage of diacritics as well as by certain changes in pronunciation in Vilnius residents’ speech.
The marking of the prefixes iž-, iš-, už- is also inconsistent at their points of contact with other morphemes, especially when these prefixes precede the reflexive marker si. Prefixes were written in two different ways – with <ž> (iž-, iž-, iš-, iž, iz-, iż-, iż) and with <š> (isz-, iż-). Prefixes with <š> never occur before voiced consonants, whereas prefixes with <ž> are possible in any position. The preposition iš is always written as iž in SP. Its position here is neutralized.
Straipsniai / Articles
61
Page 33
VIRGINIJA VASILIAUSKIENĖ
The two consonants of foreign origin <f> and <h> are used in international words. <x> is also used in Lithuanian words to mark the two-consonant group ks. The letters <c>, <th>, <rh>, <ſs> and <ſſ> (in the present tense of the verb būti) can be used according to Latin traditions.
Combining adjacent words together or writing them separately depended on a few things: (1) prosodics; (2) influence of other languages; (3) technical restrictions of printing and the shape of a page; (4) language skills (competence) of editors and typesetters. In SP the size of spaces between words differs a lot.
Ne is combined together with verbs more often in SP II than in SP I. However no clear explanation exists why ne is more often combined with some verbs (such as passive participles), but in other cases it is separated from other verbs.
Capital letters can be used in similar ways as in Latin and Polish. Besides the traditional usage of capital letters in modern Lithuanian, they were also used for ethnonyms, religious trends and basic religious concepts. These letters were used more consistently in SP I. More errors and more misprints in reading proofs were found in SP II. Stylistic usage of capital letters tends to vary. In some cases Sirvydas used capital letters from quotable Greek or Latin sources.