Egda žena otročęmь xoditъ. On the 5^th prescription in the Old Church Slavic Folia medicinalia

3.2.1 Parallels forotročęmь xoditъ

The oldest examples of the expression ‘to walk with a child’ meaning ‘to be pregnant’ that we were able to detect come from Old Polish.^[16] The earliest one occurs in the Gniezno Sermons, from the early 15^th c. (Vrtel-Wierczyński 1953: 25). Cf. the glossing in 2.

(2) Maria geſt ona brzemēna byla

Mary-nom.sgaux-3sg3-nom.sg.f pregnant-nom.sg.f be-prf.ptcp-nom.sg.f

a ſvim ßinkem milim

and refl.poss-ins.sg.m son-dim-ins.sg beloved-ins.sg.m

geſtcy ona bila chodzila

aux-3sg-aff 3-nom.sg.f be-prf.ptcp-nom.sg.f walk-prf.ptcp-nom.sg.f

‘Mary was pregnant and (had?) walked with her beloved son’

If taken at face value, the second clause expresses an event preceding Mary’s pregnancy, since the verb is put in the pluperfect, which does not make any sense whatever the meaning of the expression ‘walk with one’s beloved son’ would be. This is but one of the famous stylistic peculiarities of the Gniezno Sermons, adding to their extreme redundancy (cf. Stieber 1952). Therefore, the form bila should not be taken into account when analyzing the passage in question, like it is the case with numerous other instances in that text.^[17] The two events can then be treated as simultaneous and, moreover, the two coordinated clauses as synonymous, with both denoting pregnancy, cf., again, numerous similar tropes in that text.^[18] What is more, the birth of Jesus is described in the subsequent periodic sentence, in an even more redundant way, so that the meaning ‘give birth’ can be ruled out for the passage cited in 2. The only problem in comparing this example to the Old Church Slavic one is that here the preposition s could simply be omitted before the word-initial s of swym <ſvim> (Nepokupnyj 1964: 73; Twardzik 1997). However, numerous other examples, only some of which are discussed below, show that a construction without a preposition was possible in this context.

The same problem concerns the example from the Song of Annunciation, written in the first half of the 15^th c. and preserved in two 15^th-c. copies that do not differ in the grammatical interpretation of the relevant passage (Maciejowski 1852: 124; Lubicz 1893: 594). The glossing of the older variant, published by Lubicz, is given in 3.

(3) Elzbÿetha synem chodzi ßwÿąthego Jana

Elizabeth-nom.sg son-ins.sg walk-prs.3sg saint-acc.sg.m John-acc.sg

porodzÿ

prf-give.birth-prs.3sg

‘Elizabeth walks with a son, she will give birth to John the Baptist’

Here, the meaning of the expression is clear, since what was and might have been said by Gabriel to Mary follows from Luke 1:36. Accordingly, in the song’s text, he speaks about Elizabeth currently “walking with a son”, i. e., being pregnant, and John the Baptist still going to be born, in the future.

The third Old Polish example occurs in a translation of the Rule of the Third Order of Saint Francis. Whereas the translation itself goes back to the late 15^th c., the copy in question may be somewhat later (Twardzik 2005: 153). The variant attested in that copy (Łopaciński 1893: 712) is glossed in 4.

(4) syosthry kthore dzyathkamy chodzą albo brzemienne

sister-nom.plrel-nom.pl.f child-ins.pl walk-prs.3pl or pregnant-nom.pl.f

‘sisters who walk with children or [in other words] pregnant’

While the coordination seems at first glance to complicate rather than facilitate the interpretation, it must be noted that it is not paralleled in the Latin original included in the manuscript, cf. <Sorores grauide> ‘pregnant sisters’ (Łopaciński 1893: 712).^[19] We are therefore dealing with a doublet translation, which is a common strategy in medieval texts, reflected among others on the same page of the manuscript.^[20] All these facts point to the interpretation of <dzyathkamy chodzą> ‘walk with children’ as ‘are pregnant’. Importantly, the omission of the preposition would be unmotivated in this case, so it can be considered an entirely reliable parallel to the Old Church Slavic hapax legomenon.

Attestations for the expression ‘to walk with a child, son, etc.’ in the meaning ‘to be pregnant’ continue into the Middle Polish period. For example, it occurs several times in Mączyński’s dictionary (Maczinsky 1564: 121 d, 133 d, 149 c, [390]c, 509d) and is still noted in Cnapius’s (1621: 67; 1626: 327) dictionary and Mesgnien’s (1649: 132) grammar. The material from the 16^th c. (SPXVI 3: 281) shows that the preposition s > z was used only optionally at that time, having possibly been introduced hypercorrectly, as though deleted (simplified) only before the s- of syn ‘son’ (cf. the Old Polish examples above). Apart from the words for ‘child’ or ‘son’, the construction occurs with other words referring to unborn children as well as animals. Interestingly, while the lexemes syn ‘son’, dziewka ‘girl, daughter’, płód ‘foetus’, and brzemię ‘burden’ as well as pronouns and personal names are normally put in the singular (cf. the examples in SPXVI 3: 281; Sławski 1961: 308), dzieci and dziatki ‘children’ usually occur as plural in spite of a single pregnant woman being referred to (not necessarily with multiples), and the collective bydło ‘cattle, livestock’ is attested in reference to a single animal.^[21] The expression clearly begins to decline no later than the second half of the 18^th c., cf. Chodzi Synem ‘walks with a son’ as an entry in Włodek’s (1780: 6) dictionary of obsolete expressions. Nevertheless, it was attested in the 20^th c. in Polish and Kashubian dialects, without the preposition in Mazovian (e. g., Example 5) and with the preposition z as well as o (with the locative) elsewhere (SGP 3: 623).

(5) χoʒ́i dvunastėm ʒ́ećakėm

walk-prs.3sg twelfth-ins.sg.m child-ins.sg

‘walks with the twelfth child’

Mention should also be made of chodzić as a bare verb meaning ‘be pregnant’ and as adjoining various manner or time expressions. The former possibility is reflected in a single citation for the Greater Polish dialect: I̯ag͜ χoʒ́iu̯am ‘when I walked’, i. e., ‘when I was pregnant’ (SGP 3: 623). Manner expressions, involving adjectives in the nominative or the instrumental, are characteristic of the Middle Polish period, cf. nom.sgbrzemienna, ins.sg-ą ‘pregnant’, nom.sgciężka, ins.sg-ą ‘heavy’ (SPXVI 3: 281), and can co-occur with nominals denoting or pronouns referring to offspring, cf. (i) (płód) którym brzemienną chodźiłá ‘(the fetus) with which she was walking pregnant’, (ii) choćby nim náten cżás mátká brzemienną chodźiłá ‘even if the mother were then walking pregnant with him’ (SPXVI 2: 468).^[22] Also time expressions may co-occur with such nominals in the 16^th c. The independent use, reflected in gdy iuż nadedniem chodzą ‘when they already walk over the day’ meaning ‘when they are shortly before the child’s birth’ (SPXVI 3: 281; cf. s.v. dzień SPXVI 6: 396), is paralleled in modern dialects in the north, e. g., kobıu̯a χoʒ́i jedenåśće ḿeśėnʒı ‘a mare walks eleven months’ meaning ‘a mare’s pregnancy lasts eleven months’ (SGP 3: 623).

The Czech material is virtually limited to expressions including the preposition s ‘with’, as in the single Old Czech record known to us, from 1484: s dětmi našimi, s Michalem, s Štěpánem, s Margaretú i s tiem, s kterýmž chodí ‘with our children, with Michael, with Stephen, with Margaret, and with that with which she is walking’ (Kniezsa and Király 1952: 50).^[23] Worth mentioning are some related expressions without the verb choditi but including the noun in the bare instrumental, attested since the 14^th c., cf. dietětem těžka bieše ‘she was heavy with a child’, počala jest dietětem ‘she conceived with a child’.^[24]

For Middle Czech, cf. s djtětem choditi ‘walk with a child’ s.vv. Počjti ‘conceive’ and Těhotnau býti ‘to be pregnant’ in Adam Veleslavín’s dictionary (Adamus 1598: 1061, 1568). Importantly, the same work includes a German equivalent, cf. die mit eynem kind gehet, mit eynem kind gehen (Adamus 1598: 1568–1569, and see Section 3.2.2 on the German material). Nedochodiła s tim djtětem ‘she did not make it to walking with this child’ is one of several expressions recorded in Rosa’s 17^th-c. dictionary that denote (un)successful delivery by combining the noun djtě ‘child’ with various, mostly participial, forms of the perfective verb dochoditi ‘make it to walking’ (cf. Rosa 2008).

Later lexicographers usually follow Adam Veleslavín but sometimes add related examples, e. g., kráwa chodj 40 neděl ‘a cow walks for 40 weeks’ (Tham 1805: 40). Jungmann (1835: 809) includes a translation of one Polish example, and it is probably this quotation that gave him an opportunity to lemmatize the expression with the preposition as only optional, cf. DJTĚTEM neb S DJTĚTEM CHODITI ‘walk (with) a child or with a child’. Whether with or without the preposition, the phraseme shows up extremely rarely in literary works, into which it may have been artificially introduced from dictionaries, e. g., počawši synem chodila k sedmemu měsjci ‘having conceived, she walked with a son until the seventh month’ in Vlček’s translation of the Iliad (Wlčkowskj 1842: 327).^[25] Ultimately, one may suspect that it has never been used in colloquial language (possibly except for the Old Czech period), having been calqued for lexicographic purposes, from German, by Adam Veleslavín.^[26]

The preposition is obligatory in Upper Sorbian, cf. z dźěsćom zakhodzić ‘to start walking with a child’ meaning ‘to be pregnant for the first time’ (Pful 1866: 966), probably also: Tehdy sym ja z prěnim dźěsćom khodźiła ‘then I walked with the first child’ (Radyserb-Wjela 1905: 115). However, since bare instrumentals were ousted by prepositional phrases in Sorbian (Stone 1993: 614), z dźěsćom chodźić may have replaced an earlier *dźěsćom chodźić.

Apart from Old Church Slavic and West Slavic, we find corresponding examples in the East Slavic group. For Belarusian, Mjacel’skaja & Kamaroŭski (1972: 77) note the phraseme dzicem xadzic’ ‘walk with a child’, paraphrased as byc’ cjažarnaj ‘to be pregnant’, from the dialect of Savoni in the Stowbtsy District. For Ukrainian, analogous expressions are attested in Hrinčenko’s dictionary as xodyty dytynoju (divčynoju) ‘walk with a child (girl)’, paraphrased as byt’ beremennoj ‘to be pregnant’ (SUMHrin 3: 407). The examples come from dialectal records from the Borzna and Chernihiv areas in North Ukraine, a work by Hanna Barvinok (born in or near Borzna in 1828), and a story recorded near Dnipro (then Ekaterinoslav) in South East Ukraine by Hrinčenko’s wife (before 1897). Finally, we find parallel constructions in Russian dialects, lemmatized as xodit’ (s) kem-l[ibo] ‘walk with somebody’ (SRNG 51: 78–79). The prepositionless variant appears beyond the Ural Mountains, namely in the Priuralsky District (Tyumen Oblast; 1946–1965) and the Tomsky District (Tomsk Oblast; 1986, 2002). One example which includes the preposition was recorded in the European part of Russia, in the Solikamsky District (Perm Krai; 1973). Apart from regular personal nouns denoting children (mal’čikom ‘boy’, (s) devkoj ‘girl’), a nominal employed in the expression can refer to the number of children carried, cf. odnem ‘one’, dvojnikami ‘twins’, or the number of previous pregnancies, cf. četverimi ‘four’. Some nouns in the instrumental – without a preposition – refer to the woman’s body, either directly or metaphorically, cf. brjuxom ‘belly’, košёlkoj ‘braided bag’ (cf. SRNG 15: 144), or as manner expressions, cf. goroj ‘as a mountain’.

In Russian dialects, the verb xodit’ can also stand alone as expressing pregnancy, cf. kotora ženščina xodit ‘which woman is walking’ meaning ‘a woman that is pregnant’ (SRNG 51: 78), and can be adjoined with time and manner expressions (SRNG 51: 78–79). The only example combing a nominal in the instrumental case denoting offspring and a further adjunct is attested within the material from the Priuralsky District, cf. the prepositional phrase v položenii ‘pregnant’ (originally: ‘in condition’) in Ja xodila ej [dočkoj] v položenii ‘I was walking pregnant with her (a daughter)’.^[27] Among various adjuncts occurring independently, adjectives are used not only in Russian dialects, cf. nom.sgčežela ‘heavy’, but also in standard Russian, cf. ins.sgberemennoj ‘pregnant’ (BAS 1: 552–553), and standard Ukrainian, cf. nom.sgvažka, ins.sg-oju ‘heavy’ (SUM 1: 279). For time expressions, cf. dolgo ‘long’, devjatʹ mesjacev ‘nine months’, in Russian dialects (SRNG 51: 78).

We can conclude that expressions of the type ‘walk with’ in the meaning ‘to be pregnant’ are well attested in East and West Slavic. This clearly speaks for interpreting also otręčemь xoditъ in the Folia medicinalia in this sense. Since the presence of a (seemingly) comitative instrumental without a preposition is unexpected from the perspective of attested Slavic, the question arises whether it must indeed present an archaism or whether it can be explained as resulting from a secondary development. In Section 3.2.2 we present some ideas about the origin of the construction under scrutiny.

3.2.2 A genuine Slavic construction?

There are in principle two ways of explaining the expression otročęmь xoditi as genuine Slavic. First, it might contain an archaic comitative instrumental without a preposition. The meaning of the construction would thus originally have been ‘to walk with a child’, as we have assumed for the sake of simplicity in the previous sections. Second, the construction could also have originated as the result of ellipsis in a longer construction, in which case the semantics of the instrumental would not necessarily have to be comitative. In this section, we first of all intend to evaluate the plausibility of these two explanations. Following that, we explore the possibility of the construction being a loan translation from another language.

According to Luraghi (2001: 387), a “[p]rototypical Comitative involves an animate agent performing an action together with another animate individuated entity, conceived as performing the same action”.^[28] The expression otročęmь xoditi does not entirely fit this definition since the unborn child, while probably still conceived as animate,^[29] does not perform the same action as the mother. In the same way as we cannot paraphrase ‘Mary goes to school with her books’ as *‘Mary and her books go to school’ (Luraghi 2001: 387), we cannot paraphrase ‘The woman walks with an unborn child’ as *‘The woman and the unborn child walk’. Therefore, the OCS expression would belong to a group of less prototypical comitative constructions.^[30] The validity of the assumption of a comitative precursor for otročęmь xoditi ‘to be pregnant’ thus depends on whether it is plausible to posit bare instrumentals with the same semantics for Old Church Slavic or an earlier stage of Slavic.

As mentioned in Section 3.1, bare instrumentals with prototypical comitative semantics do not occur in contemporary Slavic as a productive formation. And even at older stages they are extremely rare. Of the 16 OCS examples for bare comitative instrumentals cited by Staniševa (1958: 42–43) and Vondrák (1928: 277), only five approximate the prototypical comitative.^[31] It should be noted, however, that four of these denote groups, sometimes metonymically (cf. 2× vьsěmъ domomъ ‘with the entire house’), rather than individuals, and in the single remaining example the omission of the preposition s(ъ) could be either phonologically conditioned or instigated as a scribal error by the initial letter s- of the following abbreviation. The situation is similar in Old East Slavic as well as in the older attestations of Polish, Czech, and BCMS, which leads Staniševa (1958: 43) to the conclusion that bare instrumentals with prototypical comitative semantics were ousted by constructions with the preposition sъ ‘with’ already in Proto-Slavic.^[32] Bare comitative instrumentals in the broader sense largely suffered the same fate in most Slavic languages.^[33] While BCMS deviates to some extent from the overall picture (cf. Staniševa 1958: 43–45; Nepokupnyj 1964: 75–76), the constructions attested there do not reflect the prototypical meaning either.^[34] Ultimately, if a bare comitative instrumental is assumed for otročęmь xoditi, it seems most convincing to explain it as an archaism preserved in an idiomatic expression.^[35]

The second possibility to account for otročęmь xoditi as a genuine Slavic construction is to assume that it is based on a construction which apart from a word for child and the verb xoditi contained a manner expression or a similar adjunct. As noted in Section 3.2.1, ‘to walk with a child’ is sometimes attested together with adjectives, like Middle Polish brzemienna ‘pregnant’ (originally: ‘burdened’), or adverbial prepositional phrases, like Russian dial. v položenii ‘pregnant’ (originally: ‘in condition’), cf., respectively, (płód) którym brzemienną chodźiłá ‘(the fetus) with which she was walking pregnant’ (SPXVI 2: 468); Ja xodila ej [dočkoj] v položenii ‘I was walking pregnant with her (a daughter)’ (SRNG 51: 78–79). Therefore, it is possible that expressions of the type ‘to walk with a child’ resulted from ellipsis in structures such as ‘to walk adj because of a child’, with the instrumental of cause.^[36]

The most widespread and possibly inherited adjectival terms for pregnancy occurring with an instrumental case and/or the verb ‘to go’ can be reconstructed as CSl. *bermen(ьn)a^[37] and *tęžьka^[38] (cf. SD 1: 160). Furthermore, the residual adjective *berdˊa (ÈSSJa 1: 188; SP 1: 202) should be mentioned, whose reference to women’s rather than animals’ pregnancy is largely limited to BCMS (cf., moreover, with further reservations concerning Croatian, RHSJ 1: 622) and the earliest occurrences in RuCS (SRJaXI-XVII 1: 330).^[39] The unabridged Common Slavic prototype might therefore be reconstructed as ins.sg + *berdˊa/bermen(ьn)a/tęžьka xoditi ‘to walk pregnant/burdened/heavy because of (a child)’. Obviously, the possible variants would differ in redundancy, e. g., *otročę(tь)mь berdˊa xoditъ ‘she walks pregnant because of a child’ vs. *synъmь bermen(ьn)a xodějaše ‘she was walking burdened because of the son’. In the former, the present tense must have ruled out the knowledge of the child’s sex, and since berd’a ‘pregnant’ was probably monosemic, the noun *otročę ‘child’ could not add to the meaning of the phrase. In the latter, the noun *synъ ‘son’ is less redundant, in the context of the past tense (imperfect), as it helps specify, first, the past pregnancy being referred to and, second, the meaning of the polysemic adjective *bermen(ьn)a. In our opinion, the phraseologism would be more likely to originate in the latter type of expressions.

It follows from the above remark on redundancy that if the meaning ‘pregnant’ of a given adjective has been conventionalized, the noun may not have been an obligatory part of the expression, e. g., Middle Polish chodziłá brzemienna ‘she was walking burdened/pregnant’ (SPXVI 2: 468; 3: 281). However, given that pregnancy is not infrequently tabooed and referred to euphemistically in the Slavic languages (cf. SD 1: 160), a Common Slavic expression of this type may have been subject to various modifications. And since expressive potential is inherent in qualitative adjectives, it is their omission (as well as the potentially redundant inclusion of nouns) that could serve to moderate the expressivity of the entire phraseme, e. g., Old Polish dzyathkamy chodzą ‘(they) walk with children’ (Łopaciński 1893: 712).

Last but not least, the verb of motion seems not to have been obligatory either, e. g., Old Czech dietětem těžka bieše ‘she was pregnant with a child’ (Vokabulář 2023), Middle Polish byłá brzemienna dwoygiem dzyeći ‘she was pregnant with two children’ (SPXVI 6: 239). While it is difficult to determine the original conditions for using or omitting the verb *xoditi, its inclusion could be conventional, helping to avoid unintended expressivity or literality. Note the bare reflexes of the verb as well as constructions including time expressions as denoting pregnancy in Czech, Polish, and Russian (Section 3.2.1).

Thus, the core element of the phraseme in question seems to be the verb *xoditi, whereas the other constituents, i. e., the noun in the instrumental and the adjective (denoting pregnancy independently), might have been optional. Incidentally, it cannot be ruled out that the supposedly unabridged reflexes of the phraseme resulted from contaminations of previously separate expressions, e. g., the bare verb *xoditi and syntagmas of the type *otročę(tь)mь berd’a/bermen(ьn)a/tęžьka byti. Be that as it may, the variant of the phraseme that includes the noun and omits the adjective seems to go back to the Common Slavic period.

Apart from treating otročęmь xoditi as a genuine Slavic construction, the possibility that it presents a calque from another language should be taken into account. Considering that the expression occurs in East, West, and South Slavic, the loan translation would have had to occur at an early stage. Since otročęmь xoditi can be treated either as a syntagma with a comitative instrumental or as resulting from ellipsis (see above), the construction in the source language could have been either of the type ‘to walk with a child’ or of the type ‘to walk (pregnant/)burdened/heavy because of a child’.

So far, we have only been able to detect possible examples for the loan in relatively recent German. An expression of the type ‘to walk with a child’ meaning ‘to be pregnant’ is attested in the 15^th-c. pharmacopoeia of Erhart Hesel, cf. so get si volkenlich mit ainem kind 3 r (Haage 1972: 40; Leidig 2004: 404), and a construction in which the instrumental could be interpreted as an instrumental of cause is reported from Middle High German, cf. Sich dîn mûme Elizabêth / ouch grôz mit eime kinde gêt which can be translated literally as ‘go big with a child’ (Erlös 2782) (14^th c.) (Jesko 2006: 244). Since this evidence is too scarce to convincingly argue for a calque, we, however, believe that at least in the current state of research an explanation that interprets otročęmь xoditi as a genuine Slavic construction is preferable. Of the two alternatives discussed above, the one assuming the ellipsis of an adjective seems more justified.

3.2.3 Conclusion on otročęmь xoditъ

In this section we have brought forward hitherto unnoticed evidence for constructions consisting of bare instrumentals and the verb ‘walk’. Examples from East and West Slavic clearly suggest that the expression otročęmь xoditi is to be translated as ‘to be pregnant’. Thus, it confirms Rosenschon’s (1991; 1993; 1994) original reading and refutes the alternative interpretation by Miklas et al. (2021) (cf. Section 3.1).

As regards the function of the instrumental, we suggested that it may either present an archaic bare instrumental with comitative meaning – the semantic shift would then have been from ‘to walk with a child’ to ‘to be pregnant’ – or that it was originally an instrumental of cause and that the expression resulted from ellipsis in a construction such as ‘to walk heavy because of a child’. The latter hypothesis was based on the occurrence of similar structures in Slavic. While we cannot rule out any of these hypotheses, we believe that the second may be more convincing considering that the type of ellipsis in question seems to be reflected in historical languages whereas the evidence for prepositionless prototypical comitative instrumentals is very scarce.

Finally, we also considered the possibility that otročęmь xoditi presents a calque from another language. Since we have so far only found similar expressions in German (starting from Middle High German), we, however, concluded that there is currently not enough evidence to substantiate such a claim. Therefore, we believe that in the current state of research it is most convincing to treat otročęmь xoditi as a genuine Slavic construction.

Having determined the meaning of otročęmь xoditi in this section, in the next section we explore whether there are alternative readings for the remaining parts of the prescription. As it turns out there are at least two possible readings for ne puštaetъ sę kry.

3.3.1 Blood release

Rosenschon (1993: 132; 1994: 307) glosses the clause to ne puštaetъ sę kry in German as so lässt sich das Blut nicht los ‘then blood does not release’ and further explains it as kommt es nicht zu einer Blutung ‘bleeding does not happen’. Thus, she analyzes the verb as grammatically middle. Considering her interpretation of the prescription as providing a treatment for abortus imminens, i. e., threatened miscarriage associated with bleeding during pregnancy (Rosenschon 1993: 144; 1994: 319–320), the entire clause would therefore merely imply the possible health problem (as though “blood may release”) by stating what is expected (“blood does not release [during pregnancy]”).

Indeed, the semantics of the verb assumed by Rosenschon is corroborated within the Old Church Slavic canon, e. g., <puštei potъ svoi na zemljǫ aky krъvь> ‘releasing his sweat onto the ground like blood’ (Euch 47a21). Another parallel, moreover, involving a reflexive form of the verb, comes from 15^th-c. Old Czech medical texts, e. g., tehdy se jie pustí krev z nosa ‘then blood will release from her nose’.^[40] The fact that these examples could present calques from a foreign original does not seem to impair Rosenschon’s interpretation – at the linguistic level, it seems fully acceptable. As regards another, more specific type of non-intentional blood release, assumed by Miklas et al. (2021: 105–106, 110–111), i. e., menstruation, the only alleged example of pustiti kry referring to it that we have come across is based on a questionable interpretation of a passage in the Old Serbian Hodoški miscellany by Angusheva (2005: 13), on which see Section 3.3.2.

The problem with reading (to) ne puštaetъ sę kry as ‘blood does not release’ in Rosenschon’s interpretation is that this clause, first, does not directly state the relevant health problem (rather, the reader has to infer that the unwanted condition consists in the opposite of what is stated, which seems unusually complicated when compared with other prescriptions) and second, does not fit the general structure of the prescriptions in the Folia medicinalia. The conjunction to, if present, introduces curative instructions recommended to the reader, usually rendered by active verbs denoting intentional actions. The objection may be raised, in order to defend Rosenschon’s interpretation, that to in the relevant part of the text is not preceded by a colon, whereas it is so elsewhere in the manuscript when introducing instructions (cf. the transcription in Miklas et al. 2021: 481); accordingly, the conjunction might have some other function here. However, if one consults the facsimile edition (Miklas et al. 2012: 298) and especially the digital images appended to the critical edition (Miklas et al. 2021), one recognizes a colon, albeit somewhat fainter, also before this to (cf. Fig. 2). A reading of ne puštaetъ sę kry amounting to an instruction is therefore to be preferred.

Figure 2

Colon in 141ar14 compared to that in 141ar15

The structure of the prescription can be explained more plausibly if the middle construction ne puštaetъ sę kry is read as referring to lack of menstruation rather than a pathological blood release. Crucially, in the Middle Ages, menstruation was considered purgative (cf. the use of Latin purgatio ‘purgation’ with the meaning of menstruation, Green 2005: 52; Hewera 2012),^[41] as reflected, for example, in Trotula’s (11^th–12^th c.) influential works (Green 2001: 73). Now since it is unavailable and undesired during pregnancy, it might possibly be substituted, for example, by administering a cathartic.^[42] The prescription might therefore be read as follows: “if a woman is pregnant, blood does not release [she has no purging menstruation]; [to salvage this,] give her a cathartic potion to drink”. Although, in this interpretation, to still remains not followed by an instruction proper directly, at least the health problem (i. e., lack of menstruation) being addressed in a more explicit way arguably makes the connection between the two parts of the prescription more comprehensible.

One possible explanation for why the health problem is not stated explicitly is that it is implicitly expressed in the first two clauses of the prescription. The conditions that come to mind here are the typical attendant symptoms of pregnancy. According to humoral theory, molimina gravidarum – especially, nausea, vomiting, and appetite – are a direct consequence of the relocation and qualitative change of bodily fluids, in particular of blood (Diepgen 1937: 158). Referring to the lack of menstruation during pregnancy could thus be an efficient way to address these symptoms. Recipe No 5 could then be interpreted as providing a treatment for pregnancy-related discomforts, possibly, but not necessarily, through purgation: “if a woman is pregnant, blood does not release [which causes molimina gravidarum]; [to treat these,] give her (...)”.

One problem with this proposal is that it assumes a fairly specific meaning of the collocation ne puštaetъ sę kry, which makes its corroboration, e. g., with Slavic cognates, more difficult (cf. above). We will address this issue in Subsection 3.3.2, where we suggest an alternative account that refers the phrase to bloodletting.

3.3.2 Bloodletting

In Section 3.3.1, we have tentatively rejected Angusheva’s (2005: 13) reading of krъvъ pustitъ in an Old Serbian manuscript as ‘menstruates’ but did not rule out that this collocation might shed light on the meaning of the clause ne puštaetъ sę kry occurring in prescription No 5. In this section, we will argue that the Old Serbian phrase should be read as ‘to let blood, phlebotomize’ and that this is also a possible interpretation for the Old Church Slavic expression in question. Paired with the doubtful segmentation of <žena> as že na (see Section 3.1), this reading appears already in Dobrev’s (1989–1990: 169) translation of the Folia medicinalia (“do not let blood”). To our knowledge, it has, however, never been substantiated.

Angusheva refers to the following passage from the Old Serbian miscellany (15th c.) stored in the National Museum in Prague (IXF10): <Reče bo Ipokrat(ь): žena imušti vъ črěvě ašte krьv(ь) pustit(ь), iznemagaet(ь), vědušti jako krьv(ь) es(tь) pišta mladěncu i pače ašte budet(ь) velěi ml(a)d(ě)n(ь)c(ь)> (IXF10: 147b) (Katić 1990: 60) ‘as Hippocrates says: a woman having in the belly – if she lets blood – falls ill, considering that blood is food for a baby, and all the more so if the baby is big’. In our view, contrary to Angusheva,^[43] it echoes Hippocrates’s aphorism V.31: Γυνὴ ἐν γαστρὶ ἔχουσα, φλεβοτομηθεῖσα, ἐκτιτρώσκει· καὶ μᾶλλον ᾖσι μεῖζον τὸ ἔμβρυον ‘A woman with child, if bled, miscarries; the larger the embryo the greater the risk’ (LCL 150: 166–167). The similarity of the parts referring to the child is especially straightforward. There are, however, some obvious differences as well, cf. the active, finite form pustit(ь) as opposed to the feminine passive participle φλεβοτομηθεῖσα ‘to bleed sb’, the unspecific meaning of prs.3sgiznemagaet(ь) ‘falls ill’ as opposed to prs.3sg ἐκτιτρώσκει ‘bring forth untimely’, and the inserted clause vědušti jako krьv(ь) es(tь) pišta mladěncu ‘considering that blood is food for a baby’. Nevertheless, analyzing the collocation krьv(ь) pustit(ь) as denoting bloodletting (phlebotomy, bleeding) seems fully justified by ten other occurrences of related expressions (including the imperfective verb puštati and the compound noun krъvopuštenie) in the same manuscript, relating to the intervention in question (cf. Katić 1990: 65, 69, 7l–72, 76).

The above reading of the Old Serbian collocation is, moreover, paralleled in 15^th-c. manuscripts written in other Slavic languages, cf. Old Czech krev púštěti,^[44] Old Polish prs.3plpusczayą krewy,^[45] and Old Russian krъvь puštati.^[46] All these facts suggest that Old Church Slavic ne puštaetъ sę kry might refer to bloodletting. If so, the first part of prescription No 5 could be analyzed in the spirit of Hippocrates’s aphorism V.31: ‘if a woman is pregnant, blood is not let’.

This interpretation clearly implies the passive value of the verb ne puštaetъ sę. While passives in sę do occur in Old Church Slavic writings,^[47] they are distributionally marked, being one of two options for translating Greek synthetic passives (cf. Večerka 1996: 219–220).^[48] This suggests that we may be dealing with a translation – and possibly a calque – from a language that uses the latter type of formations. Expressions corresponding to (ne) puštaetъ sę kry are attested already in ancient treatises written in both Greek and Latin, cf. Greek αἷμα ἀφίεται/ἀφαιρεῖσθαι ‘blood is let’ in Hippocrates (Ermerins 1859: 582; LCL 477: 232–233) and Latin sanguis (e)mittitur ‘blood is let’ in Celsus (Daremberg 1891: 52, 54, 233, 251; LCL 292: 154–165). Note also that both Greek ἀφιέναι ‘send forth, let go’ and Latin (e)mittere ‘send (forth), let go’ could be translated with OCS pustiti (SJS 3: 510). And since “[t]he Byzantine and Latin medieval codices concerning women’s health drew mainly on the Hippocratic medical tradition, which was later reworked by the late Hellenistic medical practitioners” (Angusheva 2005: 8), both Greek and Latin seem possible original languages of prescription No 5.

The question now is how the new reading can be linked to the remainder of the prescription’s text. If we consider that, according to the interpretation discussed here, the phraseme ne puštaetъ sę is to be read as ‘blood is not let’, the prescription can only be understood as providing a treatment for health problems that would otherwise be treated by bloodletting. In this regard, it is interesting to note that Hippocrates in his aphorisms IV.1 and V.29 prescribes purging as a therapy that, with some restrictions, could be prescribed during pregnancy: Τὰς κυούσας φαρμακεύειν, ἢν ὀργᾷ, τετράμηνα καὶ ἄχρι ἑπτὰ μηνῶν, ἧσσον δὲ ταύτας· τὰ δὲ νήπια καὶ τὰ πρεσβύτερα εὐλαβεῖσθαι χρή ‘Purge pregnant women, should there be orgasm, from the fourth to the seventh month, but these last less freely; the unborn child, in the first and last stages of pregnancy, should be treated very cautiously’ (LCL 150: 134–135); Τὰς κυούσας φαρμακεύειν, ἢν ὀργᾷ, τετράμηνα, καὶ ἄχρι ἑπτὰ μηνῶν ἧσσον· τὰ δὲ νήπια καὶ πρεσβύτερα εὐλαβεῖσθαι ‘Purge pregnant women, if there be orgasm, from the fourth month to the seventh, but less in the latter case; care is needed when the unborn child is of less than four months or of more than seven’ (LCL 150: 165–168). Thus, while bloodletting was to be avoided during pregnancy, purging could be performed with certain restrictions.

Now, considering that according to humoral theory, bloodletting and purging presented the usual therapy for treating the misbalance of humors responsible for an illness (Magner 2005: 101, 188–189; Stanford 2015 a: 723), it seems possible that prescription No 5 provides a cathartic potion that could be used as an alternative to bloodletting during pregnancy. It could then be applied in case of a health problem that would under normal circumstances be treated by bloodletting to avoid a miscarriage.

The custom of treating illnesses caused by an imbalance of humors by both bloodletting and purging could also be responsible for an interesting distortion of Hippocrates’s aphorisms IV.1 and V.29 (see above) spreading in medieval medical texts. For instance, in Trotula’s influential works a corresponding passage refers not only to purgation but also to phlebotomy (Green 2001: 99), and this modification is also attested in a 15^th-c. Old Czech manuscript, cf. Die Ipokras, jest li ženě třeba počiščenie bráti aneb krev púštěti, nemá před čtyřmi měsieci púščeti ani bráti počištěnie a takéž po sedmi neděléch nemá toho učiniti^[49] ‘Hippocrates says: if a woman needs to take a purge or let blood she is not to let (it) nor take a purge before four months and neither is she to do that after seven weeks(sic!)’. We believe that the inclusion of phlebotomy into the passage could have been facilitated by the common practice of prescribing the two treatments as a means to reestablish eucrasia, i. e., a balance of the basic humours.

According to this interpretation, the first part of the prescription can therefore be read as follows: ‘if a woman is pregnant, blood is not let; [instead,] give her a laxative potion to drink’. This reading seems supported by collocations of the type *krъvь pustiti ‘let blood’ occurring in medieval medical texts translated to other early Slavic languages and justified by Old Church Slavic passives in sę (as a feasible translation strategy). As regards the narrative structure of the Folia medicinalia, a prohibitive, habitual passive construction following the conjunction to seems to be in line with the composition of the prescriptions.

However, it should be mentioned that in all other instances where to occurs in the manuscript, it is followed by an instruction containing an infinitive (or an imperative, unless the lack of <t> in <nasipai> 141bv14 in prescription No 15 is due to a scribal error), not by a present tense form. Moreover, the prescription remains vague in addressing the relevant health problem in this reading. Despite these issues, we believe that ‘blood is not let’ presents a plausible reading of the clause ne puštaetъ sę kry.

3.3.3 Conclusion on ne puštaetъ sę kry

In this section, we have examined various readings of the clause ne puštaetъ sę kry. In Section 3.3.1, we have concluded that Rosenschon’s ‘middle’ interpretation (‘blood does not release’) is plausible in linguistic terms but lacks motivation unless one specifies the meaning as ‘menstruation does not occur’. The prescription could then be interpreted as providing a treatment of the discomforts accompanying pregnancy due to the lack of the cleansing effect of menstruation. In Section 3.3.2, it has been argued that the alternative, ‘passive’ interpretation (‘blood is not let’) is reliable as well. While not allowing for an identification of the health problem addressed in prescription No 5, the reading of the verb puštati as referring to bloodletting can be substantiated by parallel examples in Slavic medical texts from later periods. Moreover, it seems to match the narrative structure of the Folia Medicanalia better. Thus, both readings have their advantages and disadvantages and there seems to be no clear reason for favoring one of them over the other.

Regarding the purpose of the prescription, we have suggested that in both cases it could be related to purgation. In Section 4 we will now address the problem of the ingredients of prescription No 5.