Introduction

This text serves as an extension of the Reference Grammar and the MultextEast specifications, which describe the grammatical annotation of Pop Punčov Sbornik and other texts in our corpus. Although the tagset for this corpus has been designed in order to reflect both archaic and innovative features of the texts, the language of Psalterium Sinaiticum, representing one of the oldest documents of a Slavic variety, bears relevance beyond the scope of Balkan Slavic studies. For this reason, certain features have been annotated with more specific tags: variation of nominal and verbal stem, long-form adjectives, Greek equivalents, and similar phenomena. The modifications were also applied to our edition of Kiev Folia.

1. Transcript

The source file includes four script variants: coded Latin, diplomatized Latin, Cyrillic and Glagolitic. The transcript is based on the critical (Cyrillic) editions by Severjanov (et al. 1922, containing Ps 1-137, i.e. manuscript Slav. 38/O of the St. Catherine Monastery) and Mareš (et al. 1997, containing Ps 138-151, Slav. 2/N), in unclear cases the facsimile (Altbauer 1971, Tarnanidis 1988, the pictures available in the Library of Congress) or the Demetrius Psalter (Miklas et al. 2021, Slav. 3/N) were used for reference and reconstruction of damaged passages (especially Ps 67:20-36). Numbers are given with letter, not numeric equivalents, i.e. as in the edition by Mareš. In coded Latin, Cyrillic and Glagolitic, reconstructions and corrections were reflected with parentheses ( ), misplaced letters with square brackets [ ]. Otherwise, these transcripts follow the principles used in the rest of corpus, ensuring compatibility with the UTF-8 standard.

As the Psalter is written in a continuous script, the word boundaries generally reflect the critical editions. The separator character +, used in other texts of the corpus, is used in case of phonetic interactions between words. In case of haplographies, words are separated, and letters are added as reconstructions to the second word in the sequence.

original	Severjanov	coded Latin
Ⰺⰸⰴⱃⱘⰽⱏⰹ	Iз-д-рѫкъı	Izd+ rǫkъi	21:21
ⱄⱏⰱⰵⱃⱑⱅⰵⰿⱆ	Съберѣт-е-моу	Sъberěte (e)mu	49:5

As in the rest of corpus, a word divided by a line break shows the conjoining character _. If the second part of the word represents the only text on the line, a placeholder token --- (e.g. in 21:24, following prosla_vite-) is added. While the critical editions try to give exact graphical representations for interpunction and ornaments in the text, verse breaks (mostly written by smiley-like sequences in the original) are given here with the first character only, attached to the preceding token. Middle dots are used instead in the transcript of the critical edition by Mareš, following his practice.

original	Severjanov	coded Latin
Ⱉⰳ҃ⰹⰺⰸⰱⰰⰲⰻⰴⱎ҃ⱘⰿⱁⱙ :)	Ѡ г҃ı ïзбави дш҃ѫ моѭ:)	W g҃i ïzbavī dš҃ǫ mojǫ:	114:4

In the diplomatic transcript, some elements were removed: suprasegmental markers (whose function is not always clear), variation of letters denoting /i/, capitals, interpunction, and so on. In comparison to other sources in the corpus, the variation of jers is preserved. Digraphs denoting /y/ are given with y as well. The ambiguous sequence ⱐⰹ was manually transcribed as y or ii according to the context (e.g. n.sg.loc obilьi → obilii in 29:7; but m.pl.acc gradьi → grady in 9:7). Original sequences are preserved in the coded Latin transcript. Because of rather irregular use of jers (cf. Mareš et al. 1997:xvii), the diplomatic transcript uses only one jer character (ъ) for easier searching.

The character set shows minor deviations from the convention set up by Jagić and the critical editions of the Psalter, aiming at maximum Unicode compatibility and exactness of conversion between the script variants. The only Glagolitic letter without a suitable Unicode equivalent is the "open ižica" transcribed with Cyrillic у by Severjanov and Mareš (e.g. разумѣ in 39:10), given by a combination of characters Ⱁ- in our edition (Lat. ú). In the Cyrillic transcript, the rare ⱒ (the "sun-shaped" х, Miklas 2004:396) is given with х̂ (Lat. h), in an analogy to Severjanov's practice of transcribing ⰼ (ǵerv) with г̂ (Lat. ǵ).

2. Morphological Analysis

As in the rest of the corpus, the morphological analysis focuses on the form of the ending, rather than its function, as e.g. in the TOROT database. Thus, a token like vrěmję receives morphological tag Nnsnn, literally also n.sg.nom, although the same form could be used for an accusative. The actual function — an oblique argument denoting temporal specification — is apparent first within the context (vь vrěmję svoe, 1:3) and indicated in the syntactic annotation. This practice allows study of formal variation in texts under the influence of various language systems - diachronically, dialectally, and also textually. This is especially relevant for the Psalterium, a transcript going back to probably an early, if not already Moravian (with forms like svęce- in 29:1, mymъ in 39:10, etc.) translation of a Greek text, which itself continues a long Hebrew tradition.

2.1. Nouns

Nominal inflection in the Psalterium is archaic, but already showing some case syncretisms, characterizing the later Balkan Slavic developments. The annotation generally follows inflection paradigms based on gender, animacy, and stem palatality (Lunt 2001:54). The same model has been employed for the Church Slavonic texts in the rest of the corpus. In comparison to Cyrillic orthography, Glagolitic does not distinguish /ě/ and /ja/, so hard-stems like iěkově (lemma Jakov, 77:21) are treated as sg.loc (tag Nmsly), while soft-stems like i[sra]ī҃lě (lemma Izrail, ibid.) as sg.gen/acc (tag Nmsgy).

		m.sg.anim				m.sg
hard	soft	tag	function	hard	soft	tag	function
-ъ	-ь	Nmsny	sg.nom	-ъ	-ь	Nmsnn	sg.nom/acc
-a	-ě	Nmsgy	sg.gen/acc ...	-a	-ě	Nmsgn	sg.gen
-u	-ju	Nmsdy	sg.dat	-u	-ju	Nmsdn	sg.dat
-e	-e	Nmsvy	m.sg.voc
-ě	-i	Nmsly	sg.loc	-ě	-i	Nmsln	sg.loc
-omь	-emь	Nmsiy	sg.inst	-omь	-emь	Nmsin	sg.inst
...
		m.pl				m.dl
hard	soft	tag	function	hard	soft	tag	function
-i	-e	Nmpny	pl.nom	-a	-ě	Nmdny	dl.nom/acc
-ъ	-ь	Nmpgy	pl.gen	-u	-ju	Nmdgy	dl.gen/loc
-omъ	-emъ	Nmpdy	pl.dat	-oma	-ema	Nmddy	dl.dat/inst
-y	(-ę)	Nmpay	pl.acc/inst
-ěxъ	-ixъ	Nmply	pl.loc

Soft m.pl stems show a slightly different distribution of the forms (see below). Inanimate m.pl/dl nouns use the same endings, tagged Nmpnn, Nmpgn etc. Animacy is usually indicated on feminine and neuter nouns, but does not affect morphology: thus, Nfsnn, Nfsgn etc. are generally equal to animate forms.

		f.sg				n.sg
hard	soft	tag	function	hard	soft	tag	function
-a	-ě	Nfsny	sg.nom	-o	-e	Nnsnn	sg.nom/acc
-y	-ę	Nfsgy	sg.gen	-a	-ě	Nmsgn	sg.gen
-ě	-i	Nfsdy	sg.dat/loc	-u	-ju	Nmsdn	sg.dat
-o	-e	Nfsvy	f.sg.voc
-ǫ	-jǫ	Nfsay	sg.acc	-ě	-i	Nmsln	sg.loc
-ojǫ	-ejǫ	Nfsiy	sg.inst	-omь	-emь	Nmsin	sg.inst
...
		f.pl				n.pl
hard	soft	tag	function	hard	soft	tag	function
-y	-ę	Nfpny	pl.nom/acc ...	-a	-ě	Nnpnn	pl.nom/acc
-ъ	-ь	Nfpgy	pl.gen	-ъ	-ь	Nnpgn	pl.gen
-amъ	-amъ	Nfpdy	pl.dat	-omъ	-emъ	Nnpdn	pl.dat
-axъ	-axъ	Nfply	pl.loc	-ěxъ	-ixъ	Nnpln	pl.loc
-ami	-ami	Nfpiy	pl.inst	-y	-i	Nnpin	pl.inst
...
		f.dl				n.dl
hard	soft	tag	function	hard	soft	tag	function
-ě	-i	Nfdny	dl.nom/acc	-ě	-i	Nndnn	dl.nom/acc
-u	-ju	Nfdgy	dl.gen/loc	-u	-ju	Nndgn	dl.gen/loc
-ama	-ama	Nfddy	dl.dat/inst	-oma	-ema	Nnddn	dl.dat/inst

Consonant stems are tagged according to the paradigms of grammar books. Thus, nouns in phrases like sъ nebese (32:13) are tagged Nnsgn and sъ nebesi (79:15) with Nnsdn, even if the use of preposition shows, that the functional distinction was fading. Differences in distribution of forms typical of short-vowel and jo-stems is indicated by new options at the animacy (5th) position of the tag.

	u-stems			i-stems
ending	tag	function	ending	tag	function
-u	Nmsgu	sg.gen/loc ...	-ь	Nmsni	sg.nom/acc
-ovi	Nmsdu	sg.dat	-i	Nmsgi	sg.gen/dat/loc
-a	Nmsau	sg.acc	-ie	Nmpni	pl.nom
-ove	Nmpnu	pl.nom	-ьi	Nmpgi	pl.gen
-ovъ	Nmpgu	pl.gen	-i	Nmpai	pl.acc
...
	soft m.pl
ending	tag	function
-i	Nmpnj	pl.nom/inst
-ę	Nmpaj	pl.acc

Feminine i-stems show the same set of endings (tagged Nfsni, Nfsgi etc.) save for plural, where the ending -i (tagged Nfpni) denote pl.nom/acc. Ending -ovi commonly appears in pl.nom nouns (e.g. Sn҃ovi tuždi, 17:45, tagged Nmpnu, with the alternative tag Nmsdu). Soft m.pl.nom nouns ending in -e (e.g. cr҃e in 2:10) are tagged Nmpny, as their endings are distinct from pl.inst forms (unlike cr҃i, e.g. in 2:2, tagged Nmpnj). Neuter jo-stems regularily distinguish pl.gen (e.g. sъvědě_neī, 24:10) from pl.inst (vъzdъi_xanьiī:, 30:11), so they don't require similar adjustments.

By the distinct tags for u- and i-stem endings, it is possible to analyze the variation of forms of the common lemma Gospod. Among them, the form g҃i is etymologically tagged as an i-stem vocative (Nmsvi), if the context does not imply another function.

g҃ъ	Nmsny	1:6
g҃ě	Nmsgy	2:2
Tъi že g҃i	Nmsvi	3:4
vъ imję g҃i	Nmsgi	19:6
gju҃	Nmsdy	7:18
gv҃i	Nmsdu	2:11

2.2. Adjectives

The annotation of adjectives does not differ much from the system employed in the rest of corpus, with paradigms similarly distributed as nouns. The only form not attested in later sources is the uninflected adjective of the type īsplъnь 'full' (e.g. 32:5), tagged A---i. An additional position in the tag has been added to distinguish adjectives of animate and inanimate m.sg nouns: while in Blaženъ mǫžь, the adjective (tagged Amsnn, 1:1) expresses clearly sg.nom case, in dx҃ъ burenъ (Amsnn-n, 10:6) the form could express an accusative too. As animate m.sg nouns show adjectives more often in the text, only the adjectives of inanimate noun phrases receive the distinctive tag. The relevant options are the following:

		m.sg				m.sg anim
hard	soft	tag	function	hard	soft	tag	function
-ъ	-ь	Amsnn-n	sg.nom/acc	-ъ	-ь	Amsnn	sg.nom
-yi	-ii	Amsny-n	sg.nom/acc.pron ...	-yi	-ii	Amsny	sg.nom.pron
-a	-a	Amsgn-n	sg.gen	-a	-a	Amsgn	sg.gen/acc
-aego	-aego	Amsgy-n	sg.gen.pron	-aego	-aego	Amsgy	sg.gen/acc.pron

In the Psalterium, adjectival long forms showing a sequence of vowels are irregularily contracted. In fact, they reflect various stages of the process. For example, m.sg.gen of ništ is attested in the "regular" long form ništaego (10:4), in the form niśtaago (13:6) with an assimilated suffix, as well as in the fully contracted form niśtago (21:25). Uncontracted forms receive an alternative pronominal tag. The forms are common for masculine and neuter adjectives (the latter receiving tags like Ansgy:Pp3nsg, Ansdy:Pp3nsd etc., e.g. Sъ gotovae^go žiliśta in 32:14), adjectival tags may also be extended with inanimate m.sg markers as described above (e.g. o_tъ kameni dragaego, 20:4, tagged Amsgy-n:Pp3msg).

hard	soft	tag	alt.tag	function	late CS equivalent
-aego	-aego	Amsgy	Pp3msg	sg.nom/acc.pron	-ago
-uemu	-juemu ...	Amsdy	Pp3msd ...	sg.nom/acc.pron ...	-omu
-ěemь	-iimь	Amsly	Pp3msl	sg.loc.pron	-omь
-yixъ	-iixъ	A-pgy	Pp3-pg	pl.gen/loc.pron	-ixъ
-yimъ	-iimъ	A-pdy	Pp3-pg	pl.dat.pron	-imъ
-yimi	-iimi	A-piy	Pp3-pg	pl.inst.pron	-imi

• The adjectival long form is interpreted on the browser level as "pronominal" (.pron), so adjectives attaching pronominal endings on the root, like in na vьsěkъ denъ: (7:12), are always tagged so (Amsny-n). Lemmas thus handled include mostly possessive adjectives like moi or naš, as well as determiners like vse (OCS m.sg.nom vьsь) or the mentioned vsěki (OCS vьsjakъ or vьsěkъ).
• Soft-stem m/n.sg.loc and inst long form -iimь is mostly homographic in OCS. It is sometimes (Trunte 2022:88) even considered regular, although the form reflects assimilation (sg.loc: < -i+jemь) or analogy (sg.inst: < -emь+jimь; Lunt 2001:66). In the Psalterium, the loc/inst ending is rare, e.g. vyšьnimъ (72:11). An alternative tag indicates the homography (Amsly:Amsiy).
• As in the rest of the corpus, participles are regularily tagged with both verbal and adjectival tags. Adjectival endings of present and old past (*-us-) active participles are inflected according to soft paradigms, showing similar oscillations as that of jo-stems like m.pl.nom cr҃e vs. cr҃i: e.g. sǫdęštei (2:10) vs. nadějǫštii (2:12). The variation is presently only commented in the notes, and not integrated to the tagset, as soft m/n.pl.inst short forms are not attested in the text, and m.pl.nom and inst long forms are distinct. The long-form marker position may be used to distinguish palatality of stem (e.g. *ništi or *nadějǫšti: Ampnj, *ništę, *nadějǫštę: Ampaj).

2.3. Pronouns

Annotation of nominal (personal/reflexive) pronouns strives to reflect the attested forms. Specific for Psalterium are the contracted 1sg.gen forms like u mne (2:8), which receive an alternative dative tag (Pp1-sg:Pp1-sd). In contrast to the rest of corpus, long sg.dat (mьně, tebě) and pl.gen (nasъ, vasъ) forms show extended tags to reflect case syncretisms specific for OCS.

	1sg			2sg			refl
form	tag	function	form	tag	function	form	tag	function
azъ	Pp1-sn	1sg.nom	ty	Pp2-sn	2sg.nom
mene	Pp1-sg	1sg.gen	tebe	Pp2-sg	2sg.gen	sebe	Px---g	refl.gen
mьně	Pp1-sdl	1sg.dat/loc ...	tebě	Pp2-sdl	2sg.dat/loc ...	sebě	Px---dl	refl.dat/loc ...
mi	Pp1-sd	1sg.dat	ti	Pp2-sd	2sg.dat	si	Px---d	refl.dat
mę	Pp1-sa	1sg.acc	tę	Pp2-sa	2sg.acc	sę	Px---a	refl.acc
mьnojǫ	Pp1-si	1sg.inst	tobojǫ	Pp2-si	2sg.inst	sobojǫ	Px---i	refl.inst
...
	1pl			2pl
form	tag	function	form	tag	function
my	Pp1-pn	1pl.nom	vy	Pp2-pn	2pl.nom
nasъ	Pp1-pgl	1pl.gen/loc	vasъ	Pp2-pgl	2pl.gen/loc
namъ	Pp1-pd	1pl.dat	vamъ	Pp2-pd	2pl.dat
ny	Pp1-pa	1pl.acc	vy	Pp2-pa	2pl.acc
nami	Pp1-pi	1pl.inst	vami	Pp2-pi	2pl.inst
...
	1dl			2dl
form	tag	function	form	tag	function
vě	Pp1-dn	1dl.nom/acc	va	Pp2-dn	2dl.nom/acc
naju	Pp1-dg	1dl.gen/loc	vaju	Pp2-dg	2dl.gen/loc
nama	Pp1-dn	1dl.dat/inst	vama	Pp2-dn	2dl.dat/inst

Adjectival pronouns can be alternatively tagged as adjectives (lemmas like vse) or personal pronouns (e.g. if a demonstrative pronoun is used as a 3rd person pronoun). Unlike adjectives, the animacy of m.sg forms like togo is not distinguished, as pronouns appear often alone (e.g. togo vъziśtǫ, 26:4). If the gender is unclear, either the masculine or feminine tag is used in forms shared with neuter (as Pd-msg in the example from 26:4), or none, as in the case of shared plural forms.

	hard stems
	m.sg			f.sg			n.sg
form	tag	function	form	tag	function	form	tag	function
tъ	Pd-msn	m.sg.nom	ta	Pd-fsn	f.sg.nom	to	Pd-nsn	n.sg.nom/acc
togo	Pd-msg	m.sg.gen/acc	toę	Pd-fsg	f.sg.gen	togo	Pd-nsg	n.sg.gen
tomu	Pd-msd	m.sg.dat ...	toi	Pd-fsd	f.sg.dat/loc ...	tomu	Pd-nsd	n.sg.dat
			tǫ	Pd-fsa	f.sg.acc
tomь	Pp1-sl	m.sg.loc				tomь	Pd-nsl	n.sg.loc
těmь	Pp1-si	m.sg.inst	tojǫ	Pd-fsi	f.sg.inst	těmь	Pd-nsi	n.sg.inst
...
	m.pl			f.pl			n.pl
form	tag	function	form	tag	function	form	tag	function
ti	Pd-mpn	m.pl.nom	ty	Pd-fpn	f.pl.nom/acc	ta	Pd-npn	n.pl.nom/acc
...			těxъ	Pd--pg	pl.gen/loc
...			těmъ	Pd--pd	pl.dat
ty	Pd-mpa	m.pl.acc
...			těmi	Pd--pi	pl.inst
...
	m.dl			f/n.dl
form	tag	function	form	tag	function
ta	Pd-mdn	m.dl.nom/acc	tě	Pd-fdn	f.dl.nom/acc
...			toju	Pd--dg	dl.gen/loc
...			těma	Pd--dd	dl.dat/inst
...
	soft stems
	m.sg			f.sg			n.sg
form	tag	function	form	tag	function	form	tag	function
sь	Pd-msn	m.sg.nom	si	Pd-fsn	f.sg.nom	se	Pd-nsn	n.sg.nom/acc
sego	Pd-msg	m.sg.gen/acc	seę	Pd-fsg	f.sg.gen	sego	Pd-nsg	n.sg.gen
semu	Pd-msd	m.sg.dat ...	sei	Pd-fsd	f.sg.dat/loc ...	semu	Pd-nsd	n.sg.dat
			siǫ	Pd-fsa	f.sg.acc
semь	Pp1-sl	m.sg.loc				semь	Pd-nsl	n.sg.loc
simь	Pp1-si	m.sg.inst	sejǫ	Pd-fsi	f.sg.inst	simь	Pd-nsi	n.sg.inst
...
	m.pl			f.pl			n.pl
form	tag	function	form	tag	function	form	tag	function
si	Pd-mpn	m.pl.nom	się	Pd-fpn	f.pl.nom/acc	si	Pd-npn	n.pl.nom/acc
...			sixъ	Pd--pg	pl.gen/loc
...			simъ	Pd--pd	pl.dat
się	Pd-mpa	m.pl.acc
...			simi	Pd--pi	pl.inst
...
	m.dl			f/n.dl
form	tag	function	form	tag	function
sia	Pd-mdn	m.dl.nom/acc	si	Pd-fdn	f.dl.nom/acc
...			seju	Pd--dg	dl.gen/loc
...			sima	Pd--dd	dl.dat/inst

As in the rest of corpus, adverbial pronouns denoting locations and directions, even if based on a demonstrative or interrogative stem, are primarily annotated as adverbs. An alternative tag may reflect the form of the suffix, e.g. allative sěmo 'hither' (72:10), tagged R:Pr---a. The same system can be analogically applied for interrogative root k- and its derivates (e.g. kъde 'where' in 41:4, Pq:Pr---n). The adverbial/relative category (Pr) serves to distinguish these "adverbial cases" from nominal ones like kogo (tagged Pq---g, e.g. in 26:1) However, these morphems are often used without clear distinction (cf. vъnǫ_trъǫdu: for locative 'inside, ἔσωθεν', 44:14).

<

	root
adv. case	s-	t-	k-	tag
locative	sьde	tu	kъde	Pr---n
ablative	sǫdu	tǫdu	kǫdu	Pr---g
allative	sěmo	tamo	kamo	Pr---a

2.4. Verbs

In most situations, the annotation of verbs follows the model used in the rest of the corpus. Still, some forms do not appear in later texts. The supine is one of them, although it is sometimes replaced by the ti-infinitive, e.g. Ěko grjędetъ sǫdītъ zemli- sǫdīti vъseleněi vъ pravъdǫ (95:12, Dem. showing sǫditъ in both clauses). Supine form is tagged on the mood/type position: Vmu for sǫdītъ, on the basis of MultextEast specification for Slovene. Another archaism is the use of root aorist forms. Later ox-aorists like 1sg mogox (Vat.slav.2), replacing most root forms but 2/3sg, are not used in the Psalterium. The distinction does not seem to reflect any distinctive function in comparison with sigmatic aorists in OCS, but it may be used for statistical comparisons. A tag extension is used, when the verb form shows the thematic suffix in this sense.

1sg	vьzdvigъ	Vmia1ser	24:1
2/3sg ...	reče	Vmia3ser	2:7
1pl	Proīdomъ	Vmia1per	65:12
3pl	dopadǫ	Vmia3per	15:6
3dl	Ízne_možete ...	Vmia3der	17:37

An exception is the rare (formally) 3pl.aor bǫdǫ used in the imperative — or injunctive (Večerka 1984:162) — sense (e.g. in 108:8, tagged Vmm-3pe). A similar archaism, attested only once in the text, is the 1sg optative Otъpaděmъ (7:5, tagged Vmm-1pe because of the -mъ ending). Aorist forms of i-verbs, appearing on place of Greek futures, like Ízbavi 'he will deliver me' (17:20, for 3sg.fut.mid ῥύσεταί), are annotated as aorists. An alternative tag may be used in unclear cases: e.g. Ízbavi could formally be an imperative as well, so it is tagged Vmia3se:Vmm-2se in the instance. These are also used, when endings were corrected by the scribe or the modern editor, like the second Privedǫ(tъ) in 44:16 (tagged Vmip3pe:Vmia3pe, for 3pl.fut.pass ἀχθήσονται). In general, the annotation does not distinguish the person of 2/3sg aorists and imperatives — the former are added as quasi-3rd, the latter as 2nd person, regardless of the form of the sentence subject (e.g. Bǫdi pǫtь i^xъ tъma, 34:6, verb tagged Vmm-2se, for 3sg.aor.imp.pass γενηθήτω).

The annotation of verbs in this edition of the Psalterium focuses on a more detailed analysis of verbal stems. The aspect of individual verbs may differ from other texts or the data given in the Lemma Dictionary, built with focus on modern varieties. The annotation of aspect in the Psalter and Kiev Folia is based on the data from the dictionaries available at the Gorazd website. If both aspects are given in the dictionary, the stem is considered anaspectual, and the tag position is left blank, e.g. in 2sg.imp vīždъ (9:14), tagged Vmm-2s.

So far, the following lemmas are considered anaspectual: běžati, blagoslovja, blagoslovestviti, ida, iskydati, izpitam, jam, konьčati, liša, naslědьstvovati, obetъšati, obidja, obitati, obonjati, otvěštati, peja, pija, prěditi, prezra, prosja, prostraniti, poida, pojasati, poletěti, pomnja, porabotati, porugaja, poslušam, progněvati, protivja, prověštati, razumeja, razьnьstvovati, sadja, slyšati, svętja, sъdja, sъgrěti, sъmotriti, sъtęžati, sъvěštati, uča, ugotovati, venčaja, věrovati, vesti, vidja, vъskolěbati, zasъvědělьstvovati, znamenati, živiti.
Non-present forms classed under the lemma bъda (e.g. 2/3sg.aor bystъ, imp bǫdi, cond bi, ptcp.aor.act byvъ) are also handled as anaspectual.

Lemma Dictionary distinguishes four conjugation classes based on present stems (a-, e-, i- and athematic verbs, cf. examples here), each with its own aorist stem. In OCS, one aorist stem may be shared between two present stems, e.g. 2/3sg.aor sъkaza could be related to 3sg.prs forms sъkažetъ (perfective je-stem) or sъkazaetъ (imperfective aje-stem, cf. Kamphuis 2016:295), both probably reflecting forms productive as markers of a specific aktionsart at different periods, but still lexicalized. For this reason, the stem of each verbal form is indicated in a separate column of the source file. The database LOVe serves as a dictionary for the attested options. The stems are classified according to suffix (if identifiable) or the root phonem.

Aorist stems are relevant, alongside the aorist tense itself, also for infinitives and l-participles.

stem	example	verse ...
a-	3pl.aor sobrašę	2:2	a-suffix on root
aH-	3pl.aor sta	1:1	root auslaut in -a
C-	3sg.aor dastъ	17:14	consonant root with no sigmatic/thematic suffixes
e-	3sg.aor sěde	1:1	thematic suffix (-e/o-) on a consonant root
eH-	3sg.aor pětъ	7:1	root auslaut in a -ě from a PSl diphthong or long *ē
eN-	3sg.aor priętъ	6:10	root auslaut in a nasal vowel (< *-em-t)
eR-	3pl.aor otvrěsę	21:14	root auslaut in a metathetic -ě (< *-ver-s-n̥t)
ě-	3sg.aor vъznenavidě	5:6	PIE fientive suffix *-eh₁-
i-	3pl.aor poučišę	2:1	PIE causative/iterative suffix *-ei-
iH-	inf ubiti	9:29	root auslaut in -i
nǫ-	inf vъskrъsnǫti	40:9	PIE present suffix *-neu-
ova-	3sg.aor vъzdradova	15:9	a-suffix following a denominal ov-suffix (?)
s-	3pl.aor sъmęšę	6:3	s-suffix merging with a root consonant (< *-meNt-s-n̥t)
u-	1sg.aor čjuxъ	34:15	root auslaut in -u
y-	3sg.aor zaby	9:13	root auslaut in -y

Present stems are also relevant for both finite present forms and present participles (both active and passive).

stem	example	verse ...
a-	3pl.prs imǫtъ	17:39	a-suffix on root
aje-	3sg.prs upadaetъ	1:3	thematic suffix following an a-suffix (?)
i-aje-	ptcp.prs vъzvěštaę	2:7	aje-suffix following an old *ei-suffix
v-aje-	3sg.prs prizyvaetъ	41:8	aje-suffix following a long root vowel or *eh₁-suffix
e-	3sg.prs idetъ	1:1	thematic suffix on a consonant root
ěje-	3sg.prs udolěetъ	9:26	thematic suffix following an *eh₁-suffix
C-	1sg.prs esmъ	2:6	consonant root with no sigmatic/thematic suffixes
i-	3sg.prs poučitъ	1:2	PIE -ei- or essive suffix -h₁i-
ije-	1sg.prs počijǫ	4:9	thematic suffix following a root -i
je-	2pl.prs ištete	4:3	PIE present suffix *-je-
ne-	3sg.prs pogybnetъ	1:6	PIE *-neu-
oje-	1sg.prs pojǫ	7:18	thematic suffix following a root -oi
uje-	2sg.prs trěbueši	15:2	thematic suffix following an ov-suffix
ve-	ptcp.prs živǫštei	23:1	thematic suffix following a long root vowel
yje-	ptcp.prs pokryeši	30:21	thematic suffix following a root -y

In LOVe, verbs are given with aorist and present stems for classification, which more or less reflects the traditional system by Leskien (1922:121f.). Still, one lemma may show various constructions of stems within tense paradigms, e.g. 1sg.aor daxъ is analyzed as an aH-stem, 2/3sg dastъ as a C-stem; among e-/ne- words, nǫ-suffix appears in infinitive, but not in aorist forms. Generally, 3sg.prs and 2/3sg.aor forms were taken as decisive, with 1pl and 3pl forms distinguishing between sigmatic and thematic aorists (e.g. idъ, idǫ vs. rěxъ, rěšę). Following combinations of aor/prs stems are common in OCS:

Leskien I: prs. e-stems	e-/e- (e.g. iti, 1sg.aor idъ), s-/e- (rešti, 1sg.aor rěxъ < *rēk-s-om), eN-/e- (načęti), eR-/e- (zavrěti), a-/e- (bьrati), iH-/ije- (biti), u-/ve- (pluti)
Leskien II: ne-stems	e-/ne- (izběgnǫti, 2/3sg.aor izběže), nǫ-/ne- (minǫti, 2/3sg.aor minǫ)
Leskien III: je-stems	a-/je- (drěmati, 3sg.prs drěmletъ), a-/aje- (izbirati, 3sg.prs izbiraetъ), a-/je- or aje- (with various present stems attested: sъkazati, 3sg.prs sъkažetъ or sъkazaetъ), ě-/ěje- (želěti, 3sg.prs želěetъ), ova-/uje- (negodovati, 3sg.prs negoduetъ), eR-/je- (mlěti), y-/yje- (kryti, 3sg.prs kryetъ), ě-/je- (xotěti, 3sg.prs xoštetъ)
Leskien IV: i-stems	i-/i- (causative/iterative: gasiti, 3sg.prs gasitъ; denominal: ljubiti, 3sg.prs ljubitъ), ě-/i- (stative/fientive: bъděti, 3sg.prs bъditъ; old perfect: gorěti, 3sg.prs goritъ)
Leskien V: athematic	aH-/athematic (dati, 3sg.prs and 2/3sg.aor both dastъ, 1sg.aor daxъ), ě-/a- (iměti, 3sg.prs imatъ)

Lemmas categorized according to attested variation can be viewed via the stem class view.

Stems used in imperfects, imperatives, and past participles are classed separately, as their markers interact with other suffixes differently. Imperfect forms, showing a similar sharing for multiple present stems as aorists (e.g. 3pl ponošaaxǫ in 41:11 could be related to both i-stem ponošętъ and aje-stem ponošajǫtъ), are generally annotated with lemmas relevant for the aje-stem.

stem	type	example	verse ...
a-	imperfect	3pl.impf oblygaaxǫ	37:21	a-suffix on root
i-a-	imperfect	3pl.impf nǫždaaxǫ	37:13	a-suffix following an *ei-suffix
ě-	imperfect	1sg.impf goněaxъ	37:21	ě-suffix on root
...
ai-	imperative	2sg.imp prěvitai	10:1	imperative/PIE optative *-ih₁- on an a-suffix
C-	imperative	2sg.imp daždъ	19:5	shortened *ih₁-suffix on a root -d
ě-	imperative	2pl.imp bǫděte	31:9	PIE full-grade optative *ieh₁-suffix on root (used in plural)
ěi-	imperative	2pl.imp razuměite	2:10	ih₁-suffix following fientive eh₁-suffix
i-	imperative	2sg.imp zabǫdi	9:33	ih₁-suffix on root or merging with an ei-suffix
ni-	imperative	2sg.imp vъskrъsni	3:8	*ih₁-suffix on nasal present suffix
oi-	imperative	2pl.imp poite	9:12	*ih₁-suffix merging with a root -i
ui-	imperative	2pl.imp raduite	2:11	*ih₁-suffix on a root -u
vi-	imperative	2sg.imp živi	36:27	*ih₁-suffix following a root -i
...
a-	ptcp.aor.act	m.sg.dat blagoděavъšjumu	56:3	participal *-us- following an a-suffix
aH-	ptcp.aor.act	m.pl.nom vъstavъše	34:11	participal suffix following a root -a
C-	ptcp.aor.act	m.pl.acc rekъšję	11:5	participal suffix directly on root
i-	ptcp.aor.act	pl.gen ostraštъšixъ	16:9	participal suffix following a shortened *ei-suffix
...
a-	ptcp.aor.pass ...	m.pl.nom vygъnani	36:28	participal -n- following an a-suffix
aH-	ptcp.aor.pass	m.sg.nom prědanъ	9:35	participal -n- following a root -a
e-	ptcp.aor.pass	m.sg.nom zabъvenъ	30:13	participal -en- directly on root
i-e-	ptcp.aor.pass	m.sg.nom postavlenъ	2:6	participal -en- following a shortened *ei-suffix
eN-	ptcp.aor.pass	m.pl.nom sъpęti	17:40	participal -t- following a root long vowel

Reflecting the given suffix variation, OCS verbs can be analyzed into chains of following components. The maximal model is roughly based on aje-stems, derived from both types of i-verbs.

	position	-2	-1	0	+1	+2	+3	+4	+5
		prefix	prefix	root	*-ei-/-ov-	-a-	tense/mood	person/number	case
aor a-	sobrašę		so-	br-		-a-	-š-	-ę
aor aH-	sta			sta			(-*s-)	(*-t)
aor e-	sěde			sěd-			-e	(*-t)
aor ě-	vъznenavidě		vъz-	nenavid-	-ě		(*-s-)	(*-t)
aor i-	sъkrušilъ (3:8)		sъ-	kruš-	-i-		-l-		-ъ
aor nǫ-	vъskrъsnǫti		vъs-	krъs-	-nǫ-			-t-	-i
aor ova-	vъzdradova		vъzd-	rad-	-ov-	-a	(*-s-)	(*-t)
prs aje-	upadaetъ		u-	pad-		-a-	-e-	-t-	-ъ
prs i-aje-	pomyšlějǫtъ (9:23)		po-	myšl-	(-j-)	(-a-)	(-jo-)	-ǫt-	-ъ
prs v-aje-	razuměvajęi (13:2)			razum-	-ě-	-va-	(-je-)	-ę-	-(jь)
prs e-	idetъ			id-			-e-	-t-	-ъ
prs ěje-	udolěetъ		u-	dol-	-ě-		-e-	-t-	-ъ
prs i-	poučitъ		po-	uč-	-i-			-t-	-ъ
prs i-	prěvьznosjęšta (36:35) ...	prě-	-vъz-	nos-	(-i-)		-ęšt-		-a
prs je-	ištete			išt-			-(j)e-	-te
prs je-	pospěetъ (1:3)		po-	spě-			-e-	-t-	-ъ
prs uje-	trěbueši			trěb-	-u-		-e-	-š-	-i
impf a-	oblygaaxǫ		ob-	lyg-		-aa-	-x-	-ǫ
impf i-a-	nǫždaaxǫ			nǫžd-	(-j-)	-aa-	-x-	-ǫ
imp ai-	prěvitai		prě-	vit-		-a-	-i	(*-s)
ptcp.aor.act i-	ostraštъšixъ		o-	strašt-	(-j-)		-ьš-		-(i)ixъ
ptcp.aor.pass a-	vygъnani		vy-	gъn-		-a-	-n-		-i
ptcp.aor.pass i-e-	postavlenъ		po-	stavl-	(-j-)		-en-		-ъ

• Verbs with two prefixes are rare. Some stems show even more, but likely because they were lexicalized with prefixes as new roots: e.g. in vъznenavidě, the negative marker cannot be detached, so the root is actually nenavid--. Denominal verbs like razuměite based on prefixed roots work similarily. In 1sg.prs upodoblǫ (27:1), the prefix at the position +1 is u-, the root being based on podoba 'likeness, manner'.
• The position +1 shows multiple options: *-ei- (causative/iterative), denominal -i-, denominal -ov-/-u-, *-eh₁- (fientive), and maybe *-h₁i- (essive) and -n- (new fientive? cf. Stang 1942:54).
• The position +2 shows the a-suffix, which can be combined with causative or denominal markers, as a marker of iterative or durative aktionsart (cf. Yamazaki 2023:297). The imperfect tense marker likely developed alongside this suffix (cf. Stang 1942:81, Kamphuis 2016:299). Verbs like sъkonьčavaetъ (83:3, Dem.: iskonъčavaetъ) show a reduplication of the suffix, what may be a diachronic development.
• The next position +3 shows markers of tense (aorist -s-, possibly present -je-), mood (imperative/old optative *-ih₁-/ieh₁-) and participal suffixes (-l-, -ǫšt-, *-us-, -n/en/t-). The thematic suffix (-e/o-) also appears here: it may have served as a present tense marker in OCS, given aor/prs alternations as a-/aje- or iH-/ije-. It does not appear in i-verbs, which mostly denote states and indefinite movements (like učiti), so it may have worked as a marker of a definite (non-perfect, narrated?) event, or a definite object (but that would have to occur way earlier, cf. Kortlandt 1983:312). Forms like 3sg.prs obladatъ (21:29) may show an early omission, as in modern varieties.
• The position +4 falls out in 2/3sg.aor and imp because of phonetic developments (together with the sigmatic suffix).
• The case (position +5) is relevant for participles and infinitive/supine distinction. Historically, the same slot seems to have been relevant for distinction of present ("primary") endings (Fortson 2004:84). The long form in ostraštъš-ixъ, actually a clitic pronoun, could be interpreted as the position +6.
• The synchronic variation was often conditioned by the opposition of stem and the endings, even if the stem was composed of diachronically different components — as in the case of originally a je-stem pospěetъ (root spě-) and an ěje-stem udolěetъ (root dol-).
• In any case, the variation at positions +1 and +2 is of a derivational (and partly lexical) character (cf. Kamphuis 2016:103).

3. Syntactic Analysis

Syntactic relations are annotated according to the Universal Dependencies model, and its application described in the Reference Grammar. Generally, most verbal elements are chosen as sentence roots. In grammatical periphrastic constructions (like perfects, passives), the semantic element is selected as the root: ěko xvalimъ estъ grěšъnoi (9:24). Elsewhere, the verbal root principle is applied more consistently as in the rest of the corpus. Modal verbs, generally rare in OCS, are preferred over infinitives as roots (e.g. i ne mogǫtъ stati, 35:13). In predication, the copula is selected as the root - the relation "cop" is not used. If the copula is left out, the subject is promoted to the root (Blaženъ mǫžъ iže ne idetъ ..., 1:1), according to the promotional hierarchy shared with the rest of the corpus:

advmod > nsubj > obj > obl > nmod

Additional options may be added only in the column for extensions for comparison with Greek, or in combinations of already existing options.

UD_type	UD_ext
advcl	csubj	subject expressed by a subordinate clause ...	ěko něstъ vъ sъmrъti pominaęi* tebe*	6:6
advcl	dabs	root of a dative absolute clause	davъšu tebě imъ sъberǫtъ	103:28
obl	agent	agent in passive constructions	Azъ že postavlenъ esmъ cr҃ъ otъ nego:	2:6
root	fut	unmarked future constructions	i pojǫ* imeni g҃ju vyšъnumu*	7:18
root	opt	unmarked optative constructions	Vъzvratętъ sę i ustydetъ sę źělo	6:11

• In contrast to later varieties, the text contains many clausal subjects, expressed by participal clauses. These are occasionally promoted to roots in predicative sentences: e.g. blaženъ razuměvaę na ništa (40:2, UD_type "root", UD_ext "advcl:csubj").
• Placeholder tokens --- used after words, part of whose stay alone in a line (see above), are annotated as dependents of those words. The term diapsalma is also annotated as a dependent of the preceding word. Psalm numbers, usually written on the side of the main text, depend on the root of the title.
• In titles, the word psalm (e.g. in 4:1) or its phrase attribute are handled as roots. In attributive sentences translated with 'as for...' by Brenton like Gь҃ na nb҃si prěstolь ego (10:4, gr. κύριος ἐν οὐρανῷ ὁ θρόνος αὐτοῦ), the predicative clause is handled extension of the receiver of the attribute, which is promoted to the root:

Gь҃	1	0	root
na	2	3	case
nb҃si	3	4	obl	loc
prěstolь	4	1	acl	nsubj
ego	5	4	nmod	poss

In comparison to the previous version of the corpus, modifications were made also in the Syntax Browser, used for viewing sentence structures. First, it generates verse number for easier switching between Psalter and Syntax views. Details on individual tokens (lemma, morphology, links to dictionaries) are given on the same page. A generator for the downloadable sentence diagram in the .svg format has been added. These features (save for the verse numbers) are available also for the rest of the corpus available at the website.

Literature

Altbauer, Moshé (1971) Psalterium Sinaiticum. An 11th Century Glagolitic Manuscript from St. Catherine's Monastery, Mt. Sinai. Skopje: MANU.
Amse-De Jong, Tine H. (1974) The Meaning of the Finite Verb Forms in the Old Church Slavonic Codex Suprasliensis. A Synchronic Study. (= Slavistic Printings and Reprintings 319) The Hague: Mouton.
Beekes, Robert S.P. (author) & de Vaan, Michiel (rev., 2011) Comparative Indo-European Grammar: An Introduction. Second Edition. Amsterdam & Philadelphia: John Benjamin's.
Bunina, I.K. (1959) Sistema vremen staroslavjanskogo glagola. Moskva: Akademija Nauk SSSR.
Derksen, Rick (2008) Etymological Dictionary of the Slavic Inherited Lexicon (= Leiden Indo-European Etymological Dictionary Series Vol. 4). Leiden & Boston: Brill.
Dimitrova-Vulchanova, M. (2009) Linearization Patterns in the Old Bulgarian Nominal Expression. Dimitrova-Vulchanova, M. & Tomić, O.M. (ed.) Invesigations in the Bulgarian and Macedonian Nominal Expression. Trondheim: Tapir. 74-94.
Duridanov, I., Dogramadžieva, E. & Minčeva, A. (et al., 1991) Gramatika na starobălgarskija jazik. Sofia: BAN.
Feldstein, Ronald (2020) Law of Open Syllables (Rising Sonority). Greenberg, Marc L. (ed.) Encyclopedia of Slavic Languages and Linguistics Online. doi:10.1163/2589-6229_ESLO_COM_036020
Flier, M.S. (1974) Aspects of Nominal Determination in Old Church Slavic. The Hague: Mouton.
Fortson, Benjamin W. IV (2004) Indo-European Language and Culture: An Introduction. Malden: Blackwell.
Geitler, Leopold (1883) Psalterium. Glagolski spomenik manastira Sinai brda. Zagreb: Knjižarnica Jug. Akademie L. Hartmana.
GORAZD: Digitální portál staroslověnštiny. link
Hansen, B. (2004) The life cycle of a definite marker: the development of the short and long form of the adjective in Russian, Old Church Slavonic and Serbian/Croatian. Zbornik za lingvistiku i filologiju XLVII, 1-2. 51-73.
Holzer, Georg (2011) Glasovni razvoj hrvatskoga jezika. Zagreb: Institut za hrvatski jezik i jezikoslovlje.
Jagić, Vatroslav (ed., 1907) Psalterium Bononiense. Vindobonae: Gerold & Soc.
Kamphuis, J. (2016) Verbal Aspect in Old Church Slavonic. Leiden: University of Leiden (dissertation). link
Keipert, Helmut (2017) Cerkovnoslavjanskij jazyk: krug ponjatij. Slověne 2017:1. 8-75.
Kortlandt, Frederik (1983) Proto-Indo-European verbal syntax. Journal of Indo-European Studies 11. 307-324. link
Leskien, August (1922) Handbuch der Altbulgarischen (Altkirchenslavischen) Sprache: Grammatik - Texte - Glossar. Sechste Auflage. Heidelberg: Carl Winter.
Rix, Helmut (ed., 2001) Lexikon der Indogermanischen Verben. Wiesbaden: Dr. Ludwig Reichert.
Lunt, Horace G. (2001) Old Church Slavonic Grammar, 7th edition. Berlin: Mouton de Gruyter.
Mareš, Franciscus V. (red., 1997) Psalterii Sinaitici pars nova. Wien: ÖAW.
Miklas, Heinz (2004) Zur Relevanz des neuen sinaitischen Materials für die Entwicklungsgeschichte der Glagolica. Dürrigl, Marija-Ana; Mihaljević, Milan; Velčić, Franjo (ured.) Glagoljica i hrvatski glagolizam. 389-399.
Miklosich, Franz (1865) Lexicon palaeoslovenico-graeco-latinum. Emendatum auctum. Vindobonae: Guilelmus Braumueller. link
Mirčev, Kirill (2000) Starobălgarski ezik. Sofia: Faber.
Olander, Thomas (2015) Proto-Slavic Inflectional Morphology: A Comparative Handbook (= Brill's Studies in Indo-European Languages & Linguistics 14). Leiden & Boston: Brill.
Severjanov Sergei (red., 1922) Sinajskaja psaltyrʹ. Glagolitičeskij pamjatnik XI veka. Petrograd: Akademija Nauk SSSR.
Stang, Christian (1942) Das slavische und baltische Verbum. Oslo: Jacob Dybwad
Tarnanidis, Ioannis C. (1988) Slavonic Manuscripts Discovered in 1975 at St Catherine's Monastery on Mount Sinai. Thessaloniki.
Trubetzkoy, Nikolaus S. (1954) Altkirchenslavische Grammatik: Schrift-, Laut- und Formensystem. Wien: Rudolf M. Rohrer.
Trunte, Nicolina (2022) Slověnьsky językъ: Ein praktisches Lehrbuch des Altkirchenslavischen. 6. Auflage. Bonn.
Večerka, Radoslav (1984) Staroslověnština. Praha: Státní pedagogické nakladatelství.
Werner, Heinrich (1997) Die ketische Sprache. Wiesbaden: Harrasowitz.
Yamazaki, Yoko (2023) Diathetic problem of the Baltic ā-preterits to the simple thematic presents. Historical Linguistics Vol.134 No.1. 290-311. link