A воз и ныне-ка там. К истории частицы КА в русском языке [And the things are still the same. On the evolution of the particle KA in Russian]

In this talk, given at the 2025 Memorial Zaliznjak's Readings, I discuss the evolution of the particle KA in Russian. This enclitic particle expresses the meaning of soft encouragement in standard Modern Russian and attaches to imperative forms or ininitial proclitics like NU that can are semantically equivalent to imperatives. A close semantic parallel is provided by the Danish imperative particle LIGE that conveys a similar meaning. However, in the Russian dialects and in the Russian language of the XIX century, the particle KA is also used as a proximal deictic marker 'here and now' or 'towards the speaker' and attaches to other clitic hosts, e. g. personal pronouns (MNE-KA), wh-words (GDE-KA), etc. Max Vasmer in his Etymological Dictionary (vol. 1, 1964) takes for granted that the deictic KA (which has Slavic parallels) and imperative KA represent the same element. However, Vladimir Orel (2007) in his Etymological Dictionary accepts Igor S. Vahros's hypothesis (1984) that the imperative KA is a recent Russian innovation rooted in rapid colloquial speech and based on the allegro-variant of some focus particle going back to Proto-Slavic *toliko. He claims furthermore that the initial form is neither -KA, nor -KO, but -TKO ( < TOKO < TOL'KO < TOLIKO) and assumes that the imperative -KA /-KO /-TKO was absent from Old Russian. This theory, attractive as it is, must be rejected ipso facto since the imperative KA is safely documented by the Old Novgorod Birch bark letters № 109 and 854 dating with XII century, while deictic KA is attested in the same epoch by the Birch Bark letter 652. Moreover, the Birch bark letter № 1002 (XII century) already contains the compound deictic particle KO-VO, known from Northern Russian dialects. Andrej A. Zaliznjak (1993, 2004, 2008) correctly classified Old Russian KA as a 2P particle conforming to Wackernagel's law, though he stated that the data is too scanty to predict the place of KA in Old Russian clitic clusters. Regarding the Birch bark letter № 109, where KA attaches to the clause-initial adverbial NYNĔ, Zaliznjak initially (1989) assumed that NYNĔKA is a single lexical element characteristic of South-West Russia, though in 1993 he converted to the syntactic analysis with Wackernagel's law. Zaliznjak's first hypothesis was later supported by Polekhov & Krysko (2021) and Gippius et. al. (2023) provided that the author of the Birch bark letter № 109, certain Zhiznomir, indeed had clear Old South Russian features in his idiolect. However, the probable South Russian origin of Zhiznomir does not undermine the syntactic analysis of NYNĔ=KA since the context is unequivocally imperative. Comments. The Acknowledgments, incl. the gratitudes to the conference participants are on slide 29 of the revised presentation. All disclaimers follow. On the reasons of space, I don't discuss the important contribution of Pertsov (1996) claiming a morphological approach to the imperative KA in Modern Russian. This paper is listed in the References. Neither do I discuss Levontina's claim (2019) that the particle KA tends to expand its syntactics in Modern Russian. This paper is listed in the References, too. The checking of these claims requires an analysis of corpus data, which is beyond the reach of this paper. Andrej A. Zaliznjak's views of the Old Russian particle KA and its occurrences in the epigraphic documents have changed substantially in the period from 1989 to 1995 ― a detail which might be of interest for his biographers. In 1989, see [НГБ VIII] in the References, he clined towards a morphological analysis of NYNĔKA in the Birch bark letter № 109 (early XII century) but mentioned the cryptic spelling [T]OUTE KA <wrote X> 'X wrote right here' in the graffito [Медынцева 1978: 154, № 212]. If he assumed a syntactic approach in 1989, he should treat NYNĔ=KA 'right now' on a par with TOUTE=KA 'right here'. In 1993, see [НГБ IX: 285 ] in the References, he already mentions that Birk barch letters 109 and 682 contain the enclitic particle KA but do not mention [Медынцева 1978, № 212]. In the first edition of "Old Novgorod Dialect" (1995), Zaliznjak does not treat -KA in NYNĔKA from the Birch bark letter № 109 as a suffix but do not comment its status as a clitic either. This gap is filled in the 2nd edition of "Old Novgorod Dialect" (2004: 259), where NYNĔKA from № 109 is explicitly analyzed as NYNĔ=KA, i. e. as a clitic host NYNĔ + enclitic =KA. The same syntactic interpretation is taken in the book [Zaliznjak 2008]. Finally, in 2018, Savva M. Mikheev (see the 'Related Works' section) showed the reading [T]OUTE KA <wrote X> 'X wrote right here' in the graffito [Медынцева 1978, № 212] is a phantom. Therefore, this Novgorod graffito must be excluded from the list of Old Russian documents containing the particle =KA. In the latest updated version (25.05.25), I fixed the bug with the name of Igor Vahros's paper (1984. The correct title is 'Происхождение приимперативной частицы - ка в русском языке' in the file and in the 'References' section on this website. I also corrected the mistake with misclassifying the example from Nikolai Leskov on slide 12.本次演讲发表于2025年扎利兹尼亚克纪念读书会（Memorial Zaliznjak's Readings），文中探讨俄语助词KA的演化脉络。 该附着助词（enclitic particle）在标准现代俄语中表达温和鼓励的语义，可附着于命令式形式，或语义上等同于命令式的前置助词（proclitic）如NU。丹麦语命令式助词LIGE所表达的语义与其极为相近，可视为平行案例。 但在俄语方言及19世纪俄语中，助词KA还可用作近指指示标记（deictic marker），表达“此时此地”或“朝向说话者”的语义，可附着于其他附着助词宿主，例如人称代词（如MNE-KA）、疑问词（如GDE-KA）等。 马克斯·瓦斯纳（Max Vasmer）在其《词源词典》（Etymological Dictionary，第1卷，1964年）中默认，近指用法的KA（在斯拉夫语中存在平行用法）与命令式用法的KA为同一语素。然而，弗拉基米尔·奥列尔（Vladimir Orel）在其《词源词典》（Etymological Dictionary，2007年）中采纳了伊戈尔·S. 瓦赫罗斯（Igor S. Vahros）1984年提出的假说：命令式KA是晚近的俄语创新形式，源于快速口语，其基础为源自原始斯拉夫语（Proto-Slavic）*toliko的某一焦点助词的快速变体。他进一步提出，KA的初始形式既非-KA也非-KO，而是-TKO（演化路径为-TKO < TOKO < TOL'KO < TOLIKO），并认为古俄语中不存在-KA/-KO/-TKO这类命令式助词。 尽管该理论颇具吸引力，但实则应被推翻：命令式KA有明确的文献证据，即12世纪的旧诺夫哥罗德桦树皮文书（Old Novgorod Birch bark letters）第109号与第854号；同期的桦树皮文书第652号则证实了近指用法的KA的存在。此外，12世纪的桦树皮文书第1002号中已出现复合指示助词KO-VO，该形式在北俄方言中仍有留存。 安德烈·A. 扎利兹尼亚克（Andrej A. Zaliznjak）在1993、2004、2008年的研究中，将古俄语KA正确归类为符合瓦克纳格尔定律（Wackernagel's law）的第二位置附着助词，不过他也指出，现有文献数据过于匮乏，无法确定KA在古俄语附着助词群中的位置。 针对桦树皮文书第109号中KA附着于句首副词NYNĔ的现象，扎利兹尼亚克最初（1989年）认为NYNĔKA是一个单一的词汇单元，是俄罗斯西南地区的特色用法，但在1993年他转而采用基于瓦克纳格尔定律的句法分析路径。扎利兹尼亚克的初始假说随后得到了波列霍夫与克里斯科（Polekhov & Krysko，2021）的支持；吉皮乌斯等人（Gippius et. al.，2023）的研究也证实，桦树皮文书第109号的作者季兹诺米尔（Zhiznomir）的个人语域（idiolect）确实带有鲜明的古南俄方言特征。但即便季兹诺米尔确实来自南俄地区，这也不会动摇NYNĔ=KA的句法分析结论，因为文书上下文明确为命令式语境。 注释 本演讲修订版演示文稿的第29页包含致谢内容，其中包括对参会学者的感谢，随后是所有免责声明。 受篇幅限制，本文未讨论佩尔佐夫（Pertsov，1996）的重要研究成果——其提出现代俄语命令式KA的形态学分析路径，该论文已列入参考文献。本文也未涉及列万季娜（Levontina，2019）的观点：她认为助词KA在现代俄语中呈现出句法范围扩张的趋势，该论文同样被列入参考文献。验证上述两项假说需要对语料库数据（corpus data）进行分析，而这超出了本文的研究范畴。 安德烈·A. 扎利兹尼亚克关于古俄语助词KA及其在铭文文献（epigraphic documents）中出现情况的观点，在1989年至1995年间发生了重大转变——这一细节或许对其传记作者具有研究价值。1989年，参阅参考文献中的[НГБ VIII]，他倾向于对12世纪早期桦树皮文书第109号中的NYNĔKA进行形态学分析，但同时提及了题刻[Медынцева 1978: 154, № 212]中难以解读的拼写形式[T]OUTE KA，意为“X在此书写”。若他在1989年已采用句法分析路径，则应将NYNĔ=KA（意为“即刻、此时”）与TOUTE=KA（意为“在此处”）视为同类结构。1993年，参阅参考文献中的[НГБ IX: 285]，他已提及桦树皮文书第109号与第682号中存在附着助词KA，但未提及[Медынцева 1978, № 212]。在《旧诺夫哥罗德方言》第一版（1995年）中，扎利兹尼亚克未将桦树皮文书第109号中NYNĔKA的-KA视为词缀，也未对其附着助词的属性作出评论。这一空白在《旧诺夫哥罗德方言》第二版（2004年，第259页）中得到填补：该版明确将第109号文书中的NYNĔKA分析为NYNĔ=KA，即附着助词宿主NYNĔ与附着助词=KA的组合。扎利兹尼亚克在2008年的著作中也采用了相同的句法解读。最终，在2018年，萨瓦·M. 米赫耶夫（Savva M. Mikheev，参阅“相关研究”部分）证实，题刻[Медынцева 1978, № 212]中[T]OUTE KA（意为“X在此书写”）的释读实为误读。因此，该诺夫哥罗德题刻不应再被纳入收录助词=KA的古俄语文献清单。 在最新的2025年5月25日更新版本中，我修正了伊戈尔·瓦赫罗斯1984年论文名称的错误：该论文的正确标题为《俄语命令前助词-ка的起源》，原文俄语为Происхождение приимперативной частицы - ка в русском языке，此标题已在本文档及本网站的“参考文献”部分更新。同时，我也纠正了第12页幻灯片中尼古拉·列斯科夫（Nikolai Leskov）例句分类错误的问题。