Volume 44, No.2 - Summer 1998
Editor of this issue: Robertas Vitas
ISSN 0024-5089
Copyright © 1998 LITUANUS Foundation, Inc.


University of Illinois at Chicago

The strange title of my paper is to be explained. A long time ago I noticed while working with ancient Lithuanian texts that we often deal with at least two sorts of errors: 1) the errors made by the author himself and 2) the errors made by an editor, publisher or some other intermediator. The first kind of error could be called rather charming and helpful. The second kind I would prefer to name annoying and even harmful.

1. ERRORS OF THE AUTHOR. Let us start from the more pleasant and interesting errors which I have called charming. I would like to concentrate upon a specific sort of Mažvydas' errors - on the errors of dialectal interference, i. e. the errors that he made trying to write in a dialect other than his own native one. This means that Mažvydas made certain erroneous phonetic or morphological reconstructions of the dialect he wanted to use in writing. We need to make very sure that these errors were not made by Mažvydas' first publisher in Prussia, Janas Weinreichas (16th century), since it is hardly possible that he knew Lithuanian at all.

Mažvydas brought his native dialect to Prussia where, at that time, highland Lithuanian was strongly dominant. For a long time there were many discussions about the native dialect of Mažvydas. At least two decades ago Prof. Zigmas Zinkevičius studied all of the Mažvydas texts very thoroughly (Zinkevičius, „Martyno Mažvydo raštų kalba", in: Baltistica, Vilnius, 1977-1979, vol. 13[2], p. 358-371, vol. 14[l-2], p. 38-14, 139-146, vol. 15[1], p.16-22) and concluded that Mažvydas' native dialect was the south lowland Lithuanian. Having enjoyed reading his study a decade ago, I expressed my reactions to another specialist of Mažvydas' language, Prof. Jonas Palionis. His answer was so unexpected that I have memorized it quite precisely: "Sure, Zinkevičius' study is very solid and interesting piece of research, but the main arguments to spot the native dialect of Mažvydas are based on Mažvydas' errors. So the conclusions of Zinkevičius' study are still problematic". Although then I did not dare to say anything to Prof. Palionis, certain fresh ideas struck my mind: could it really be possible that these important scholarly results are based on such miserable things as errors? And what place are we to assign to an error of the author in investigation in general: to ignore it or what? What are the arguments of Zinkevičius to maintain that errors are important and of Palionis to claim that they are not?

Since then I have been permanently interested in the errors which authors of old Lithuanian texts made. Although at that time I did not have any very clear position of my own about the errors that Zinkevičius described, recently I have gained more and more trust and confidence in Zinkevičius' ideas.

Zinkevičius has indicated at least two important kinds of phonetic erroneous reconstructions. a) Mažvydas' south lowland long ū often corresponds to highland uo. So Mažvydas, having noticed that highlanders pronounce uo (e. g. vanduo) in the position he pronounces ū (e. g. vandū, Mažv. wandu), forged the following rule for himself: where I feel a long ū, I should change it into highland uo. It is, however, important to emphasize that some highland words have long ū as well. Mažvydas very often did not recognize the difference in the two types. Knowing that his native dialect has ū in, for example, būsi 'you will be', Mažvydas transformed it to bůsi, where a letter ů gives the possibility of reading the word in a highland way *buosi , although no highland dialect ever had such a form. Zinkevičius counted over a hundred such forms in all of Mažvydas' texts and claimed that „only a south lowlander could multiply such forms" (Baltistica, vol. 13[2], p. 366). From our viewpoint it is an erroneous reconstruction of a phonetic feature of another dialect.

b) In a similar way Mažvydas coined a rule for another phonetic correspondence: where he did find his native ī (e. g. prī) he decided to transform it to highland ie (e. g. pre, prie or pre-, prie-), but sometimes in highland these morphemes are pronounced ī (or i) as well, so Mažvydas inevitably made forms that haven't ever existed in any dialect: priekelimas (highl. prikėlimas) 'resurrection', priekelk 'resurrect' etc. Again his rule became an instrument to produce errors. And that rule and those errors could only have been made by the representative of one particular concrete dialect.

Sociolinguistically those errors of dialectal interference are of much greater importance than correct forms. A correct phonetic form only informs us of its presence in general. Errors, however, involve a very important additional sociolinguistical parameter: the authority and prestige of a language variety. The erroneous phonetic reconstruction attests to the conscious intention of the author to switch a variety, to change linguistic habits and behavior in favor of the more attractive variety. Thus the highland dialect for Mažvydas was more authoritative to Mažvydas than his native one.

So first of all Mažvydas' artificial reconstruction of pseudo-highland forms enables us to verify his native dialect quite efficiently (only a speaker of this particular south lowland dialect could have produced the forms described above). Secondly, it attests to his conscious choice of the dialect (even though Mažvydas did not know highland dialect precisely). Correct forms would not have had such a power.

The more I think about it the more I believe in the effectiveness and plausibility of error interpretation. My own experience working with Lithuanian texts of the 18th and 19th century proved to me very clearly that errors of this sort are of great importance for the reconstruction of the linguistic intention of an author. This can be said about the language of Jurgis Ambraziejus Pabrėža, Bonaventūra Gailevičius, Dionizas Poška, Antanas Savickis, Motiejus Valančius etc. Their erroneous reconstructions attest not only to priorities of dialectal varieties but also to social priorities in general. So I regard such errors as being socially charged. I have no doubt about the correctness of Zinkevičius' interpretation.

2. ERRORS OF AN INTERMEDIATOR. Let me continue with an intermediator's errors, the kind that do not cause a scholar so much optimism.

The errors of an editor or anyone other than the author made in the period when the text was written are not always easy to distinguish. Most often they are deducible only by frequency: if the same error or an error of that type is repeated quite often, it can be claimed to be an error of the author, but not of the editor. And vice versa, if the frequency is low - the probability of the author's error becomes less.

I will not talk about the errors made by 16th century editors. I want to draw attention only to the mistakes of recent, i. e. 19th-20th c. editors who made them while republishing the books which originally were published ca. four hundred years ago.

Prof. Pietro Dini, the Italian Balticist, recently did the research which proved that certain mistakes there were made even in photo-graphically reprinted books. He prepared and published a book in Italy: L'Inno di S. Ambrogio di Martynas Mažvydas, Roma: La Fenice Edizioni, 1994, which is a study of the second Lithuanian book, written by Martynas Mažvydas in 1549, and a new facsimile of the book itself.

The second Lithuanian book and at the same time the second book of Mažvydas includes only three hymns and consists of only 13 pages of text, including the notes for the hymns. Dini did determine that even in such a small book the first photo reprint by Celichowski in 1897 (Poznań, Poland) and the following one by Jurgis Gerulis in 1922 (Gerullis, Mažvydas, Seniausieji lietuvių kalbos paminklai iki 1570 metams, Kaunas, Lithuania) and in 1923 (Heidelberg, Germany) contain evident errors (Gerulis did republish Celichowski's reprint with all the errors). And these are the publications that served scholars for up to one hundred years!

Here are some photo-errors discovered by Dini:

wietoye welikas pauta dowanoiju (2) (C wietoye welikas panta dowanoiju)

ßmogu, neƒƒibayƒeijey (7) (C ßmoguzneƒƒibayƒeijey)

anta amßiu amßia (9) (C anta amßia amßia)

keles iƒchgi graba (12) (C keles iichgr graba)

There are some other small errors, but I do not intend to get into all the details. What is important is that Celichowski retouched the photocopies he made, but he could not reproduce them exactly. And the scholars of the twentieth century were naive enough to believe that photocopies do not lie! This means that none of the republishers before Dini had even cast a glimpse at the only existing original of Mažvydas' second book in the Kórnik library (Poland). This also means, that some strange, inconsistent forms that for one hundred years were thought to be the fault of the author or of the ancient publisher, appear to be the errors of the modern publishers! Today the publisher of the 1549 edition Janas Weinreichas after the centenary may acquire a higher reputation! After Dini looked through the book with special attention to the photo-errors, the texts of Mažvydas became more „friendly" to a reader; some „strange" places in the text do not seem to be confusing any more.

The truth ad fontes is never out of fashion.

The other recent publication was published by Baltos lankos in 1993: Martynas Mažvydas, Katekizmas ir kiti raštai, Vilnius: Baltos lankos, 1993. This publication was practically parallel to Dini's, and much more ample, but on the other hand, no investigation or research has accompanied it. Thus it is only a bare photo-reprint of all Mažvydas' books and the transliterated texts of his known letters. I took part in preparation of this publication myself and my task was to try to find any known originals, to make microfilms of them and to try to avoid all the dangers of retouching anything that could distort the texts. So this book is an attempt at a more precise facsimile publication, although this does not mean it is perfect. We could not find two of Mažvydas' books out of the known six, so only four of them are reprinted from originals in the book: Catechismvsa praƒty szadei, 1547; Gieƒme S. Ambraßejaus, 1549 (parallel republication by Dini); Forma chrikƒtima, 1559; Paraphrasis, 1589. The originals are in Poland (Kórnik and Toruń), Sweden (Uppsala) and Lithuania (Vilnius). Two others, the largest ones, we were forced to republish from Gerulis' photocopies (Geƒmes Chrikƒczoniskas, vol. 1-2). So any errors he might have made are errors that we could not avoid either, but we still are not in any position to trace them.

The proverb says: "Let us learn from mistakes". It is also said that fools alone learn from their own mistakes, but sages learn from the mistakes made by others. So I urge everybody to learn from the errors of others too: those of Mažvydas and of his editors.