Search Menu


Issues list


15 Years of the Estonian Constitution

ISBN 978-9985-870-22-8

Cover image



Roman Law in the Baltic Private Law Act –the Triumph of Roman Law in the Baltic Sea Provinces?

A part of the Baltic countries belonged to the Holy Roman Empire from the 13th century. Roman Law was recepted in Europe and legal education was usually based on sources of Roman Law; the Baltic countries which belonged to the Holy Roman Empire of the German Nation also recepted Roman law, both as a part of canon and secular law. Roman Law remained in force even after the status of the Baltic states changed. *1 Until the entry into force of the Baltic Private Law Act *2 (BES) in 1865 Roman Law applied in the Baltic states subsidiarily with the European ius commune. In other words, it was applied only if local sources of law did not provide a solution to an issue. In the practice of not later than the 18th –19th centuries Roman Law was still applied quite extensively; it was preferred to local law and in many areas it was even transposed in its entirety. *3

1. Conflicting conclusions of earlier studies

Such a wide application of Roman law gave reason for criticism by proponents of local law. The author of the draft BES, Friedrich Georg von Bunge *4 was convinced that local practitioners rely too much on the principles of Roman law. *5 In his programmatic writing on the drafting of provincial law, Bunge claimed that Roman law should be avoided as much as possible in the preparation of the future law *6 , but at the same time he stated that Roman law was the common part of all provincial law and its omission from the Provincial Code would imply an incomplete approach to the local private law. However, the approach to local private laws should be “reliable and complete”. *7 In addition, there are gaps in provincial law where the principles of Roman law should be referred to and their applicability defined, but not more. *8

At the same time, researchers of the BES have often stated that Roman law was the main source of the BES. J. Jegorov finds that the reception of Roman law, especially its third part plays an important role in the codification of local laws. *9 A. Ylander has even cheered that in the form of the BES, Roman and canon law was able to celebrate its triumph in the Baltic Sea provinces. *10 The BES references and index of sources seem to refer to that, as many authors have concluded. *11 These suggest that despite Bunge’s efforts and claims, Roman law was indeed the main source for the BES. The reason why the sources were added was the requirement arising from the rules of the Russian imperial codification project that sources should be cited under each article. The requirement was based on the idea of gathering together all existing law and the same applied to other legislation of the Russian empire.As the Baltic Private Law Act was essentially a collection of various local and subsidiary laws *12 , each section was supplied with a reference to the source. Roman law (the codification composed in 529–533, Corpus iuris civilis – CIC)was cited as the source very often. *13

Hermann Blaese has claimed (as opposed, e.g., to Samson von Himmelstierna *14 , the author of the previous draft provincial law, whose CIC citations were not always valid, especially where the article was copied from the General Prussian Land Law *15 ) that when compiling the BES, Bunge personally checked “all the referenced citations from textbooks and only if they proved to be correct, he transposed the text and the latter [i.e., citations]”. *16 Such a multitude of references to Roman law and Bunge’s alleged diligence in checking them is somewhat out of line with Bunge’s words about the necessity to reduce the proportion of Roman law.

2. Objective and methodology of this paper

In this paper I have tried to identify whether and to what extent the references of the BES articles actually refer to the substance of the Roman law sources referred to. As Hermann Blaese has (probably based on A. E. Nolde) stated that “codifiers were able to find ready formulations from textbooks and it was easy to transpose those to the codification” *17 , I have viewed not only “pure” Roman law *18 but also the textbooks and manuals that Bunge could have used for drafting the BES, so as to identify whether they and which of them could serve as the basis for the sections and the references to the CIC. My choice of authors is based on Bunge’s own references, the study by A. E. Nolde *19 and contemporary standard literature.

In the course of writing the paper, a question arose which is posed at the end of the paper: what was Bunge’s goal of adding the source references? It may have been a wish to simply present the sources, or he may have had broader objectives. For example, when drafting the Svod Zakonov, M. Speranski who was criticised for excessive reliance on the Western, especially French legislation, was ordered to supply references to Russian law to all the articles. For this task Speranski selected a famous specialist in Russian law, who performed the task “with great difficulties and often with extremely strained interpretation”. Nevertheless, the Svod Zakonov contains articles which are almost identical to the wording of the French Code civil and institutes that Russian law was formerly unfamiliar with. *20 One of Bunge’s objectives was certainly to lay down as many provisions applicable in the “prevailing practice” of the Baltic provinces, so as to compile a full set of laws. Or did Bunge, as already mentioned, wish to reduce the proportion of Roman law?

My analysis is based on the chapters of the BES governing the classification of things and servitudes. These parts were chosen because they are especially based on the ius commune tradition and are still largely based on Roman law (CIC) today. *21 The source references of the relevant articles of the BES also refer to Roman law and Bunge has stressed that only the general principles should be adopted from Roman law. *22

In identifying the origin of the text of the BES articles, I used first of all the statement by F. G. von Bunge in his autobiography that “when studying the sources of private law of all three provinces I took the edition of Dabelow’s general private law handbook (Handbuch des gemeinen Civilrechts) with the empty supplementary sheets and filled them in with citations from provincial laws. This became the basis of all my later works on this law”. *23 I, too, referred to Dabelow’s Handbuch…. *24 In addition, Bunge has provided indirect hints on the preparation of the BES, namely that the comments of Ottomar Meykow, Professor of Roman Law at the University of Tartu have special value for him so that he took them into account to a great extent. *25 According to Bunge, Meykow was the editor of the BES and had services in bringing its content into conformity with the newer doctrine. *26 Therefore I compared the final version of the BES also with the draft prepared by Bunge alone. *27 There were hardly any differences in the two chapters I reviewed. One article concerning the classification of things had been split into two articles in the final version. One article had been added to the servitudes chapter compared to the draft. One source reference had been added later and one had been deleted.

Since Meykow may have advised Bunge also on the contemporary standard works, I looked for the textbooks which Meykow himself used for teaching. In the university’s lecture plan, the main work listed under the literature of Meykow’s lectures was the pandect textbook by L. Arndts in various editions. *28 In addition I studied the pandect textbook by K. A. Vangerow, which was also a common approach and used by O. Meykow in his thesis for a candidate’s degree. *29 I also checked the pandect textbook by F. G. Puchta, one of the best-known jurists of the 19th century, which was considered a standard work at the time. *30

A. E. Nolde, Docent of the University of St. Petersburg has written a paper on which textbooks and sources the articles of BES were based on. Nolde has distinguished between the authors used in Bunge’s so-called original draft (C. F. Mühlenbruch, C. F. Glück, F. Mackeldey, K. A. D. Unterholzner, C. F. Koch, A. C. I. Schmid, K. A. Vangerow) *31 and those used in editing the draft, i.e., the “newer” pandect textbooks (C. F. F. Sintenis, L. Arndts, J. Weiske *32 ). Unfortunately, he has not explained his principles of selection; he has only noted that Bunge himself provided almost no references at all. Although Nolde claims that wherever an article of the BES originated from the works of different authors, he specified all of them, this does not always seem to be the case. However, he has expressly omitted only those articles which were not Roman law and those whose author he was unable to identify. *33 If we look which articles he has actually not included on the list, I believe there are more of them. I cannot say on which basis the remaining articles were omitted. In the chapters viewed in this paper, he referred to two works: C. F. Mühlenbruch’s Lehrbuch des Pandektenrechts *34 and C. F. Glück’s Ausführliche Erläuterung der Pandecten nach Hellfeld. *35 This is why I have also compared the sections of the BES with those. My choice and Nolde’s were similar to some extent, but I have also considered other works which Nolde did not regard, but which seemed important to me.

3. Roman law provisions in the BES classification of things

Firstly, I analysed thoroughly the first chapter of the first title, “Corporeal and non-corporeal, movable and immovable things” of the second book of the BES (property law). It contains 10 articles, nine of which are referenced to the CIC. In addition, four articles are supplied with references to local sources of law such as knighthood and municipal laws, etc. Out of the nine articles with references to Roman law, only three and a half had a direct link to the sources referred to. In three cases out of ten, the concept contained in the article was present in the Roman law text, but in a completely different context. The remaining two and a half articles have no relation whatsoever to the sources referred to. I will not give a full account of the analysis of each article; below are some more typical examples.

For example, there is a direct link with the references in the BES basic classification of things, in which things were divided into corporeal and non-corporeal (articles 529 and 535). As opposed to the following examples, article 529 was more or less in conformity with the Roman law referred to. *36 The texts of the BES and CIC are not entirely similar. While in the Institutions *37 (Inst. 2.2.1) it is written that “corporeal things can actually be touched” (corporales hae sunt, quae sui natura tangi possunt), BES article 529 is about general perception: “things are corporeal or non-corporeal depending on whether or not they are perceivable by the external senses” (die durch die äusseren Sinne wahrnehmbar sind). The BES definition is thus much more equivocal and broader. Perception by the senses covers also sight, smell, hearing, etc., and the line between corporeal and non-corporeal things is completely different than in the CIC.

If we search C. C. Dabelow’s Handbuch… for corporeal and non-corporeal things, we will find the original Roman form with the following extension: “All things which can be touched or otherwise perceived by the senses are corporeal things.” *38 It seems that in BES article 529, Bunge may have used Dabelow’s handbook and the definition contained in it. At the same time, it is not the “more Roman” part of the definition that has been adopted, but only its second half. *39 Bunge may have used Mühlenbruch’s textbook *40 , although the text of the BES is slightly different also from Mühlenbruch’s.

The following example is of how contemporary civil law theory on the concept of fungible things influenced the drafting of the BES articles and their references. It also resulted in the fact that most of the references under the articles concerning fungible and consumable things are inaccurate. Namely, in article 532 of the BES, things are divided into fungible and non-fungible. The source of Roman law referred to under this article (D. was, however, the direct source for only the last sentence of the article concerning fungible things; the rest of the article was added by jurists of later centuries. In Bunge’s draft BES, article 532 (article 733 part (a) in the draft) and article 534 (article 733 part (b) in the draft) were contained in a single article –– article 733.

Article 733 part (b) of the draft BES later became BES article 534 and it defined consumable things, but only one of the three sources of Roman law referred to concern consumable things (D. 7.5.1). The other sources referred to (D.; D. concern fungible things. Therefore, the latter references are not relevant to this article, but should belong to article 532. The probable reason for the inaccurate references was the difference between the final version and the draft of the article. According to the draft, all the three sources referred to should have belonged to the same article (article 733) and would have been correct there, but when the articles were changed, the sources were probably misplaced. Another thing is that these references would not have added anything new even if placed correctly, as their content was the same as regards fungible things. *41

Nolde’s claim that the texts of articles 532–534 originate from Mühlenbruch proved to be true; most of the references are also the same, but Mühlenbruch does not have all the references: namely, he does not have reference to D. 7.5.1 concerning consumable things. However, Mühlenbruch has expressly cited in his textbook only that Digesta text which Bunge referred to under article 532. *42

I compared the articles of the BES also with other textbooks and identified certain similarities with Arndts. The text of article 532 of the BES does not entirely overlap with the text from Arndts’ book, although it may have been used. Of the references of article 534 concerning fungible things, the text of one (D. is cited in Arndts’ textbook in Latin. It has not been used in the text concerning consumable things, but the reference to consumable things under the text of the article is the same. *43

The confusion with the concepts of fungible and consumable things and the references is probably due to the change in relevant legal theory.B. Windscheid has said that it was quite common earlier to confuse between consumable and fungible things, but at Windscheid’s times (the second half of the 19th century) the difference need not be stressed anymore, as it is already known. *44 The two authors that Bunge could have used in this respect did not consider the difference between these classes of things so self-evident. Namely, in his handbook Dabelow provided consumable things with the Latin equivalent of fungible things, res fungibiles. Dabelow has not separately written about fungible things. *45 Mühlenbruch considered consumable things a subcategory of fungible things and has mentioned that there was no legal difference between fungible and non-fungible things. As far as wording goes, the text of these articles of the BES bears more similarity to Mühlenbruch *46 than to Dabelow. A reconstruction of the origin of BES articles 532–534 may be proposed: when Bunge wrote the draft BES, he used Dabelow and Mühlenbruch on the premise that those were the same things. The editors of the BES (e.g., Meykow) drew his attention to the fact that those were not the same things and should be split into different articles. Once Bunge did it, the references of the articles were not changed.

The last example is of two mutually related articles of the BES, in which case the only link between the articles and the referred sources is that the texts contain the same concepts, but the text of the article does not arise from the referred source. Namely, article 536 of the BES concerns property law and the classification of things into movable and immovable property. *47 The sources referred to under the article (D.; 50.17.15; Inst. 4.6.1) indeed concern movable things and res mobiles and ownership actions and actio in rem, but the content of the article cannot even remotely be derived from these sources. Those were completely different texts, whose only common feature is the same concepts. Of the chosen authors, only Glück discussed the classification of things into movables and immovables, but not exactly in the same way as in article 536 of the BES. Glück generally discussed real right claims and claims under the law of obligations together. However, he provided, e.g., an examples of servitudes, which are considered to be immovable property. *48

Article 537 which elaborates on article 536 of the BES provides: Persönliche und Forderungsrechte, wenn letztere auch auf die Erlangung einer unbeweglichen Sache gerichtet sein sollten, gehören zu dem beweglichen Vermögen.

L.15 § 4 D. qui satisdare coguntur (II,8), vgl. mit § 1 I. de action. (IV,6). *49

However, the first of the two provisions of Roman law referred to (D. *50 ) states: It is a different case with one who has a personal claim to land. And the second one (Inst. 4.6.1 *51 ) reads:“The main classification is into two:every action which takes an issue between parties to a trial before a judge or arbiter is either real or personal.A plaintiff may sue a defendant who is under an obligation to him, from contract or from wrongdoing.The personal actions lie for these claims.In them the plaintiff says that the defendant ought to give him something, or ought to give or and do something. […]”

It is difficult to derive the text of the BES even indirectly from these two citations. The only similarity is that the text of the BES mentions movable property and the Digesta mentions a parcel of land –– both are properties, but of different classes. Also, both the BES and Digesta mention personal claims (Persönliche und Forderungsrechte and petitio personalis ja actio in personam *52 ). However, these sources of Roman law do not indicate in which case movables are concerned or are not concerned. It cannot be identified from which textbooks the references to sources originate from, as the only author who has discussed this subject was Glück, and he did not refer to these sources.

Nolde believes that the text of the article originates from Glück, but this is not exactly the truth: Namely, Glück states the opposite: “rights, rights of claim, and claims are classified as immovables if their objects are immovables [...].” *53 Why Bunge worded this in the opposite way is not clear.

4. Roman law provisions in the BES general principles of servitudes

The first title “General Provisions” of Title 4 “Servitudes” of property law contains 14 articles, two of which make reference to not only Roman law, but also other laws (e.g., the statutes of Courland and Lübeck law). The coincidence of the text of the articles and the sources referred to is greater in the title on servitudes than in the classification of things. There are few completely erroneous or irrelevant references; however, one-third of the references require a great degree of generalisation or all the references of an article regard only one point of the article, while there are no references concerning the rest of the article. There are many articles that are completely correspond to the provisions of Roman law (BES articles 1094, 1095, 1096, 1097, the first half of articles 1098, 1099) *54 , i.e., six out of fourteen articles. The texts of different authors often coincide in their approach to servitudes. Although Nolde has mainly referred to Mühlenbruch, other authors (e.g., Arndts) often use exactly the same wording. Some typical examples are provided below.

In the first example, the references under the BES article are partly relevant, while others could be replaced with other references:

BES article 1090:Betrifft die Servitut den Vortheil einer bestimmten physischen oder juristischen Person, so heisst sie Personalservitut; bezweckt dieselbe dagegen den Vortheil eines bestimmten Grundstücks, so dass dieser also von dem jedesmaligen Eigenthümer des Grundstücks beansprucht werden kann, so wird sie Real- oder Prädialservitut genannt.

L.1. L.15 D. de servitut. (VIII,1). § 2 et 3 I. De rebus incorporal. (II,2). *55

This article provides the main classification of servitudes and defines the concepts of personal and real servitudes. The first source referred to, D. 8.1.1, also provides a classification of servitudes: Servitudes attach either to persons, as in the case of the right to use and usufruct, or to things, as in the case of rustic and urban praedict servitudes. *56 The second reference, D. 8.1.15, is not quite exact, as the 15th fragment is divided into principium and section 1. However, references are usually made with the precision of a section.D. contains a few examples of invalid servitudes. *57 D. defines the nature of servitudes, which lies not in doing something, but in tolerance and inactivity. *58 Neither of these is related to the text of the article. *59 The third and fourth references (Inst. 2.2.2 and 3) are to the chapter on non-corporeal things and it notes that, e.g., the right of usufruct is also a non-corporeal thing. “The rights which belong to urban and rustic estates also come under this heading.These are also called servitudes.” *60

In conclusion, although these references are about servitudes and at least some of them mention the classification of servitudes, none of them define the nature of personal and real servitudes. They can serve as the basis for this article only very remotely. However, the article should have a reference, e.g., to Inst. 2.2.3 a few fragments below, which clearly states that “the reason these rights are called servitudes belonging to land is that they cannot exist independently of land”, which would be much closer to the text of the article. Other more suitable sources could be referred to.

As the sources of Roman law which are referred to do not provide for such a definition, and it is hardly likely that Bunge referred to the incorrect sources when knowing the correct ones, I tried to find a similar definition from textbooks. Arndts defined them not word for word, but still in a very similar way: “Rights of use are [...] whether personal, servitutes personarum, personal servitudes, or rights of use of plots of land, servitutes rerum, iura praediorum, praedial or real servitudes depending on whether the right has been established for the benefit of a certain person or a certain immovable, meaning the actual owner of the immovable.” *61 Arndts also provides the first reference mentioned by Bunge, D. 8.1.1; his other references differ from those of the BES. Dabelow’s wording is a lot more different from the BES, although he uses the same first source reference, but not the others. *62 Puchta’s wording also differs from the BES, although he maintains the same principle. At the same time, stressing the need of each owner of a servitude to gain benefits, Puchta refers to another source, D., which is referred to in BES article 1090, noting that the impulse for creating the principle of the need of a benefit arose from D. 8.1.19 *63 ; he has not referred to the other sources mentioned in the BES. *64 The common part of these sources which the BES refers to is that the above textbooks contain not only the references to Roman law, but also the Latin text of the provisions. Although the textbooks contained other references to sources, the BES mentions those whose text was provided in the textbooks and which were therefore easy to check. It seems that Bunge has taken the text of this and other articles almost or completely word for word from a textbook and added some references to Roman law on the basis of the citations made in the books without delving into the substance of the citations.

As regards servitudes, there were many cases (4.5 *65 ) (BES articles 1090, 1092, 1098 point d, 1099, 1102) where one of the sources referred to in the text reflects the Roman law source of the BES text, while the other references made in the BES are irrelevant. It also happens that various references to a source repeat the same idea, while the repeated part makes up only a half of the BES article and the other half has no reference (BES article 1101).

It is quite common that a source contains a specific example or case, from which the BES draws a general rule (BES articles 1092, 1093, 1102). In such case, secondary literature has done the generalisation work. Some sources may have served merely as inspiration for the text of the BES (BES articles 1089, 1090, 1091, 1101). Among the general principles of servitudes, none of the articles have only completely irrelevant references to sources.

I would like to give an example of the source reference of an article, in which case the general principle is presented more specifically as a principle of the right of servitude and others need to be generalised or modified in order to understand the connection with the text of the BES article:

BES article 1092:Ist der Umfang einer Servitut zweifelhaft, so spricht die Vermuthung für den geringsten Umfang des Servitutenrechtes.

L.20 § 4 u. 5 D. De servitut.praed.urb. (VIII,2); L.20 D. De servitut.praed. rust. (VIII,3); L.9 regulis iuris (L,17). *66

The third reference of this article (D. 50.17.9) is to the principle of Roman private law serving as the basis for the article –– Semper in obscuris quod minimum est sequimur (In matters that are obscure we always adopt the least difficult view), which has been transposed to servitudes.

D. gives a specific example of the servitude of stillicide (stillicidium): If rainwater was originally discharged from tiles, it is not permissible subsequently to discharge it from broad-work or any other material. *67 D. goes further with the servitude of stillicide and describes other possibilities or acquiring it and the establishment of such a servitude. It has been stated amongst other things that “a servitude can be rendered lighter, but not heavier”. *68 While the remaining text described various cases, this part of a sentence contained a more general rule, which, however, requires further generalisation and a quite different approach for the text of BES article 1092. The reference to D. 8.3.20 is again inaccurate, as it consists of four fragments; however, if all four were to be true, we have another four inappropriate references. Namely, they all describe methods of establishing other servitudes in addition to the right of servitude or the extinguishment of such servitudes, but that the rights of the owner of the dominant parcel of land correspond to the applicable servitudes.

The content of BES article 1092 has thus been provided in a generalised form in only one source. In addition, another source contains a rule that needs to be modified. The remaining sources describe in smaller or greater detail specific servitudes, but they can only serve as inspiration for article 1092 at best. In addition to that, they could have served as inspiration to many other articles to at least an equivalent extent.

Such a principle of the right of servitude, although not in quite the same wording, has been mentioned only by Mühlenbruch, from whom the first two sources also originate. *69 None of the other authors (Dabelow, Arndts, Vangerow and Puchta) discuss such a principle of the right of servitude and neither do they refer to these sources.

Compared to all textbooks, the regulation of the BES is extremely detailed and casuistic *70 ; article 1092 is a good example of this. Namely, essentially the same (the fact that the right of servitude is applicable in an as small as possible scope) can be concluded from BES article 1101. Also in the event of references to sources, Bunge has remained true to the principle of great detail and casuistry –– he has added as many references as possible that repeat each other, are partly irrelevant, or simply decorative. For example, article 1093 has seven references, all of which contain more specific, more or less relevant examples that support the general principle of the article. As many of them are only very remotely related to the text of the BES article, at least some of them could have been omitted, especially considering that Bunge wanted to reduce the role of Roman law. Considering all the incorrect or inaccurate references, he would have had plenty of opportunities to reduce the number of references to Roman law.

To summaries to results of this random analysis, all completely irrelevant references were found in the part concerning types of things; there were none in the servitudes part. Thus, of the 23 articles of the BES analysed, 3.5 articles contained only incorrect references; 6 articles referred to the same concept in a different context, and the remaining 13.5 articles had at least one source each that served as the basis for the article, even when it had to be generalised.

5. Purpose of the BES references to Roman law sources

Although Bunge himself claimed that the proportion of Roman law in the BES should have been reduced, this wish did not apply to adding references to Roman law sources. *71 More rather than less of such references have been added to the parts of the BES studied in this paper. Quite a few of the references are irrelevant. As a rule, the references have been drawn from textbooks, in which they were relevant (although perhaps in a different context), but they are irrelevant to the specific article of the BES. Considering that most articles were still supplied with at least one relevant source, it may be presumed that Bunge did not choose the references quite randomly, but made a certain choice, probably choosing references from textbooks and preferring those which provided also the text of the source and not only a reference. Bunge’s own words suggest that he wished to reduce the role of Roman law. However, this does not seem to be the case if we consider the multitude of references to Roman law sources. Although the references are not completely random, he has chosen to add a larger rather than smaller number of them. This raises the question of what role these references to Roman law played. Was the instruction to supply all articles with references to applicable law the only reason why Bunge added such a great number of references to Roman law?

The multitude of references to sources (based only on the index of sources) is one of the reasons why the BES embodies the triumph of Roman law. Could Bunge have wished the BES to strike as mainly the outcome of Roman law, even if it was not entirely so? Did the great number of references to Roman law serve to legitimise Bunge’s undertaking despite his intention to reduce the use of Roman law? On the one hand, it may have been necessary to legitimise the BES for the local provincial practitioners who had been constantly using Roman law in their practice and wished to continue doing so. It should be admitted that provincial legal science was still in a much poorer state compared to Roman law research. This way, practitioners would have been left with the impression of “soft landing” and the hope that former Roman law practice would still be usable.

On the other hand, times had changed also in terms of imperial governance: a former toleration for the special regulations applicable in various parts of Russia was being replaced at the beginning of the 1860s with the Russian central government’s wish to harmonise Russian legislation *72 , and in such case, a great number of local sources would not have been recepted so well. In addition, the Russian government was at that time fascinated by everything originating from Rome *73 –– it was certainly more acceptable for the central government for the BES as a local special law to rely on Roman rather than the local law. Therefore, it was perhaps possible to legitimise the BES this way for both local practitioners and the Russian government, while preserving many aspects of local laws.

Certainly the references may have been of help for users in interpreting the provisions of the BES; some examples of this are known. For example, Erdmann used the references of BES article 529 in his commentary to the article. *74 However, the references to the BES article served a minor role in his argumentation and were not among the main arguments. Judicial practice has not been studied in this context; this is a subject for further research. The highest court of the Russian Empire, the Senate, has used a source reference of the BES in its reasoning in at least one case. A Senate decision of 1878 relies amongst other things on a reference to Roman law provided under BES article 1602, which was cited first in Latin and then in a translation into Russian. Luckily, the source of that article was in line with the content of the article. *75 Although Roman law no longer served as a subsidiary law, the source references in the BES and their texts could thus be regarded as a part of law. How often this was done and to what extent the references could be used to extend the wording of the law and not just serve as supporting arguments, is not yet known.

6. Conclusions

Naturally this small excerpt cannot serve as a basis for very far-reaching conclusions. Still, it may be said that when the classification of things gave the impression that most references had no connection or had only a very indirect connection with the sources of Roman law; this was not the case with servitudes. However, it is the classification of things that would not be expected to deviate so much from Roman law. Servitudes are an area where later theory may have much to say and change. The result is therefore surprising.

When writing the BES articles that originate from Roman law, both Bunge and the editors of his draft used the textbook by L. Arndts “Lehrbuch des Pandektenrechts”, which Professor of Roman law O. Meykow of the University of Tartu used in his lectures. The wording of several sections of servitudes, but not so much of the classification of things, corresponds to the wording of Arndts’ textbook. Although Nolde refers almost only to Mühlenbruch and Glück in this chapter, they do not seem to be the only authors who have been used in the part discussed in this paper. It is possible that the works of various authors (e.g., Arndts and Mühlenbruch) were used in several articles of the BES. Bunge has also used Dabelow’s Handbuch… in the wording of several articles.

Most of the additions in the wording of the articles certainly originate from Arndts; some corrections seem to have been guided by Dabelow’s approach. Whether these were done by Meykow or someone else (Bunge himself) is not clear. In any case, the references in the observed chapters were added by Bunge himself in 95% or more of the cases. The question of whether the coincidence between the references of article 1090 and the Roman law sources cited in textbooks is accidental or intentional needs further investigation. I cannot currently confirm or refute whether any other textbooks were used. The reason for the large number of references to Roman law requires further research, but probably it arises from the wish to legitimise the BES in the eyes of local provincial practitioners as well as the Russian central government. However, it may be said that the multitude of references to Roman law did not imply an equally extensive use of Roman law, and we cannot speak about the triumph of Roman law that has been claimed so far.