In earlier posts, we reported on the decisions of the lower courts, and finally of the Federal Administrative Court (Bundesverwaltungsgericht), on the legality of the continued publication of a search notice on lostart.de concerning Man in an Oriental Costume, a painting attributed to Jouderville, a pupil of Rembrandt, even after the painting had been located. I had written that the judgment did not convince me legally, nor did I think it did the parties a service. The last post on the subject reported a case note of Boas Kümper who reached the same conclusion. In doing so, Kümper reviewed the decision in the light of the jurisdiction of the Federal Constitutional Court (Bundesverfassungsgericht) dealing with information provided and disseminated by state institutions (staatliches Informationshandeln).
The colourful background of the dispute between van Diemen & Co, the “Jewish” art trading company that auctioned the work off in 1935, and the “Jewish” bank, Jacquier & Securius, that had financed van Diemen’s partent company Margraf & Co is discussed in a 2011 BBC Feature.
Henning Kahmann, who acted for the subpoenaed participants (Beigeladene) which were ultimately was sucessful in these proceedings, had kindly offered to defend the Federal Administrative Court.* In addition, he provided me with an English translation of the judgment.
The requirement of specific enactment (Vorbehalt des Gesetzes) does not apply to all encroachments (Beeinträchtigungen) of basic rights. Only encroachments with an imperative character, i.e. “classic” interferences by state bodies (Eingriffe), must be based on a specific enactment. While an encroachment without an imperative character does not have to be based on a specific enactment it can be unlawful for other reasons than the lack of such an enactment.
This yardstick leads to the conclusion that the publication of a search notice concerning a work attributed to Jouderville in the internet database www.lostart.de by the Magdeburg Coordination Office (Koordinierungsstelle)** is lawful. It may disseminate the implied information that there is a looting suspicion concerning the work, i.e. that it is plausible that the work was lost by the searching party due to of Nazi persecution.
It may do so despite the fact that this information is detrimental to the plaintiff, and despite the fact that there is no specific enactment that authorizes the Coordination Office to disseminate the information. The reason is that the information about the looting suspicion is not a ban of the sale of the work and it also does not have the factual effect of such a ban. Whether the plaintiff is successful in its attempt to sell the work is not decided by the government but by potential buyers. The search notice does not imply that selling or buying the work was illegal, unlike in the e-cigarettes cases. There are no other reasons that would render the publication of the looting suspicion unlawful. In particular, the information is true and objective and it came from the competent government department.
The key legal question in the case is whether the information about the work provided by the government is subject to the constitutional requirement of specific enactment (Vorbehalt des Gesetzes). The Federal Administrative Court says it is not, Kümper say it is.
Information Disseminated by Lost Art
In order to decide this question one needs to know what information is provided in the internet database and what its effects are. On the face of it the search notice only says (as it did at the time of the judgment) that an undisclosed party is searching for the painting. However, it is undisputed that the search notice implies that the following information is being disseminated by the State of Sachsen-Anhalt. The State thinks it is plausible that the work was lost due to Nazi persecution, i.e. that there is the plausible suspicion that the work was looted (looting suspicion – Raubkunstverdacht, cf. Judgment, marginal numbers 4 and 7). This implies that the searching party may have a legitimate interest in some form of amends (Wiedergutmachung) under the Washington Principles on Nazi-Confiscated Art of 1998. It needs to be stressed, because it is often overlooked, that disseminating a search notice does not mean that the state asserts that the work actually was looted or that there is such a legitimate interest, let alone a valid restitution claim based on property rights.
Encroached Upon Rights
It is undisputed that this kind of information may deter many potential buyers from actually buying the work at stake. At the very least it reduces the value. However, there is at least one case where a search notice like the one in this case apparently had no effect on the marketability of the work. This work concerned is a drawing by Egon Schiele with the Lost Art ID 478799. It was sold at Sotheby’s in New York in November 2014 for more than USD 1,000,000. The Court could not consider this in the case, because it had to adjudicate on the basis of the facts that were found by the lower court (Section 137 para. 2 Code of Administrative Procedure (Verwaltungsgerichtsordnung, VwGO).
The Court held that the only basic rights that may possibly have been subject to an encroachment (Beeinträchtigung) as a result of the dissemination of information in the search notice on www.lostart.de were either the occupational freedom (Article 12 Basic Law, GG) or the general freedom of action (Article 2 GG) of the plaintiff (cf. Judgment, marginal number 34). Kümper also considers an encroachment upon property rights as protected under Article 14 GG, an argument which the Court examined but did not follow. Yet, this is not relevant for this discussion.
Requirement of Specific Enactment
The doctrine of the requirement of specific enactment means that certain actions of the executive branch of government which encroach upon a basic right as defined in the German Constitution, the Basic Law, are only lawful if the legislator has authorized the government to undertake such an encroachment. The doctrine is not explicitly spelled out in the Basic Law, but has been developed by the courts and legal scholars from the constitutional principles of democracy and the rule of law, as set out in Article 20 GG. Its goal is to make sure that the basic rights are protected and can only be curtailed within the boundaries that the legislator defines. In the absence of this requirement the government would be in a position to curtail basic rights without democratic legitimation. In order to reach that goal then, according to the jurisprudence of the Federal Constitutional Court, the more severe the encroachment is the more specific the enactment must be.
According to Kümper, the dissemination of the looting suspicion requires a specific law that allows the government to carry out that activity. To wit he argues that the doctrine of the requirement of specific enactment pertains to all encroachments on basic rights. This includes not only direct, regulatory (regelnd) encroachments, the so called classic interferences (klassische Eingriffe), but also indirect, factual encroachments. “Classic” interferences (cf. Judgment, marginal number 35) are orders given by a government agency, directly addressing an individual and telling that individual what to do or not to do. In contrast, where government measures lead to third party behavior that is detrimental to an individual one speaks of indirect, factual encroachments.
Kümper further cites two leading cases decided by the Federal Constitutional Court (one level higher than the Federal Administrative Court), the Glycol Wine case and the Osho case. In these cases the Federal Constitutional Court held that governmental warnings, which had a negative effect on a wine seller and a religious sect respectively, did not need to be based on a specific enactment. Kümper considers these principles, which he calls the “Glycol Wine and Osho doctrine”, as a “teleological reduction” of the doctrine of the requirement of specific enactment. He mentions that many jurists have criticized the two decisions and agrees with them.
Yet, his main point is that the Federal Administrative Court did not apply the glycol wine and Osho doctrine correctly, i.e. that the yardstick applied by the Federal Constitutional Court would have led to a contrary judgment. His reading of the Osho decision is that the Federal Constitutional Court saw it as decisive that government “action to disseminate information was often necessary in view of challenges that occur at short notice and in the event of crisis” and which are polymorphic and fluctuating and therefore not accessible to norm setting. Kümper says that the Federal Constitutional Court would only make an exemption to the doctrine of the requirement of specific enactment if this was the case. However, none of the conditions for the exemption were given. The database had not been set up to address specific individual challenges or crises or dangers arising at short notice. It had rather been institutionalized. Also the activity of the database was obviously accessible to norm setting, since the Coordination Office itself had set up abstract and general guidelines on when to include certain information in the database and when this was not the case, just like a legislator could do. He concludes that in this case there should not be an exemption to the doctrine of the requirement of specific enactment. Thus, he says, also merely indirect, factual encroachments of basic rights like keeping up the search notice need a legislatory basis. Kümper says that in this case this it was missing.
I doubt that Kümper chose the right point of departure. His claim that all encroachments (direct and indirect, factual and regulatory ones) require a specific enactment is far from undisputed.
Application to Case
The question of whether the doctrine applies must be assessed according to the purpose of the doctrine. As mentioned above it is used to ensure the effectiveness of the constitutional principles of democracy and rule of law. This is also the yardstick which the Federal Constitutional Court applies in its analysis of the scope of the doctrine of the requirement of specific enactment in the Osho decision. (It concluded there that it does not apply, cf. marginal number 94). The principles mentioned above do not necessarily back Kümper’s opinion that even indirect and factual encroachments must be justified by a special enactment.
One point which in my view weighs particularly heavily against Kümper is the following consequence of his opinion: It would allow any individual a defense claim against any information issued by the government that is detrimental to him or her as long as there is no law “specifically” allowing the government to do that. Why would the rule of law or the principle of democracy warrant such a defense claim? Why should it be in the interest of society if information that is relevant for, e.g. for the participants of the art market, is withheld simply because there is no enactment that entitles a government authority to provide that information? Such a defense claim would also make it impossible for a state-run university or museum to publish research findings that are negative for, say the general freedom of action (Article 2 GG) of an individual, unless the government was specifically allowed to provide this kind of information. This would unnecessarily hamper the efficiency of government action and ultimately be detrimental to democratic society in general. How could such a defense claim be reconciled with the many obligations the government has to disseminate information, e.g. to the press or in connection with the various laws on access to information? As these laws show, the legislator regards access to information as helpful for democratic principles. This is emphasized as well in the glycol wine decision (cf. marginal number 53). Thus the constitutional democracy principle does not seem to back Kümper’s conclusions.
If the Court had followed Kümper’s view it would have also stipulated an obligation of the government to withhold information from the public that supports the Wiedergutmachung interests of the subpoenaed participants. Why would the principles of democracy and the rule of law demand such a result? It would mean that the interests of the subpoenaed participants weighed less heavily than those of the plaintiff. This is especially questionable because any seller of such a work would have to disclose the looting suspicion regardless of the search notice. The reason is that, at least in Germany, such a suspicion is a negative characteristic of the work that is customarily regarded as essential (Section 199 para. 2 German Civil Code, BGB). Withholding information would also be questionable, because the plaintiff would not have a defense claim of this kind against a private individual who disseminates the same looting suspicion. For example, the plaintiff would not have such a claim against the Central Registry of Information on Looted Cultural Property 1933 – 1945, or against an individual publishing an article mentioning the looting suspicion. Similarly, the plaintiff could not forbid anyone from reporting that the Federal Administrative Court held that the State of Sachsen-Anhalt found that the suspicion that the work was looted is plausible. Such a defense claim would, however, be conceivable, if the information about the suspicion were wrong.
Other Criteria than Doctrine
Not applying the doctrine of specific enactment does not mean that the government can disseminate any kind of information detrimental to the basic rights of individuals. The limits derive from other legal sources than the doctrine. One may say the limits come from a level above, namely the principle that the government must not act contrary to the law. This is a more general expression of the principle of the rule of law than the doctrine. It was from this general principle that the Federal Constitutional Court deduced the following criteria for the legality of disseminating information in the Glycol Wine and Osho decisions. They are applied in the Judgment of the Federal Administrative Court as well.
According to the Judgment disseminating information must be a government task and it must come from a government body that is competent for this task (Judgment, marginal number 35). The competence to issue information in this regard is an annex to the competence for the task itself, as fulfilling the task (often) makes it necessary to disseminate information. As most governmental tasks are assigned to the executive branch by the legislator, in my opinion this criterion resembles the doctrine of the requirement of specific enactment, only that the law on which the activity of the government is based is not very specific. This is why the Court says that the doctrine of the requirement of specific enactment did not demand an authorization for action to disseminate information, which goes beyond the statutory assignment of the task (cf. Judgment, marginal number 35). One could say that the doctrine (or rather democracy and the rule of law) do not require more than a merely general enactment if there are merely indirect, factual encroachments at stake. This can be reconciled with the principle mentioned above that the more severe the encroachment is the more specific the enactment must be.
It would not only be unlawful for an agency to handle a task which the legislator assigned to another agency or which is not a governmental task at all. It would also be contrary to the rule of law if the information were not objective and correct (Judgment, marginal number 38) or if it violated the constitutional principle of proportionality (cf. Judgment, marginal number 39). If these general legal boundaries are violated, then the encroachment upon basic rights also constitutes a violation of those rights. Whether the encroachment is also a “classic” interference is irrelevant, because this result implies that there was no special enactment which justified the encroachment.
Court: Boundaries not Overstepped
The Federal Administrative Court held that these criteria were met by the State of Sachsen-Anhalt. The action of the Coordination Office was governmental, because it was assigned to the state of Sachsen-Anhalt under an agreement of 2009 between the German states and the Federal Republic (cf. Judgment, marginal number 36). In addition, the Court argues that disseminating information about a looting suspicion concerning works of art stems from the governments’ competence to guide general public policy (Staatsleitung) and that this involved the competence to engage in publicity work. This in turn, the Court goes on, included the competence to inform about matters that are important to the public. Given Germany’s historic responsibility, information on possibly looted art is regarded as important enough to be part of the governments’ (that includes the government of the State of Sachsen-Anhalt) publicity work.
I would like to add that all the German states have the task of handling cultural affairs, which is a little more specific than their very general task of guiding general public policy. For the State of Sachsen-Anhalt this is stipulated in Article 36 State Constitution. According to the Court, the information that there is a looting suspicion is correct and objective (cf. Judgment, marginal number 38) and the consequences for the plaintiff are considered not to be disproportionate for the reasons discussed in marginal number 39.)
Equivalent of “Classic” Interference?
The courts, Kümper and I agree that there is another limit to what kind of detrimental information the government may disseminate. That limit is reached if providing the information is the equivalent of a direct and regulatory interference (Judgment, marginal number 41). In that case there must be a specific enactment allowing the government to provide this kind of information. The reason is obvious: otherwise the government could undermine basic rights without the consent of the legislator just by putting an order or a ban into the form of information.
An example of such a circumvention of the protection of basic rights by the requirement of specific enactment was given in a recent case where the health authorities of another state did not forbid e-cigarettes but instead issued a “warning” that e-cigarettes may well fall under the German Pharmaceuticals Act and that selling pharmaceuticals without the appropriate license may be illegal and result in criminal persecution by the state. The Federal Administrative Court had held in another decision that this warning constituted the functional equivalent of a direct and regulatory interference, namely a practical ban on the sale of e-cigarettes, for which the government was not authorized by the legislator. By the way: One can see the doctrine of the requirement of specific enactment at work in the recent passing of a federal law banning the sale of e-cigarettes to minors.
In the Lost Art Database case the Court discussed the circumvention problem as well. It cited an earlier e-cigarettes judgment of a lower court in North Rhine Westphalia. The Court held that keeping up the search notice did not result in a practical ban on the sale of the artwork. More specifically, keeping up the search notice did not imply the notion that selling the work was illegal, did not involve a threat of prosecution and had no effect on the question of who owned the work or whether anyone was entitled to restitution.
Kümper argues that the Court’s understanding of interference was too narrow. Generally, he says, an interference should already be assumed if the informing authority intended certain steering effects (finality), and here, Kümper correctly says, the state had downright intended to impair the marketability of the work by informing about the search notice. However, Kümper’s view that the government’s intention to influence the market makes its action an interference and therefore subject to the requirement of specific enactment is not generally held. Again the relevant question seems to be whether the principles of democracy and the rule of law require a specific enactment. Given that the Court concluded in the judgment that this was not the case it is consistent that it did not consider the intended steering effects to be decisive.
The arguments for Kümper’s understanding of the equivalent of a “classic” interference seem to be the same as those that weigh in favor of his broader understanding of the doctrine of the requirement of specific enactment and so are the counter-arguments. Because of this I have developed sympathies for the idea that there should be only one criterion for both, the interference and the requirement of specific enactment. This would result in the abandonment of the differentiation between “classic” and other infringements: If government action has an imperative character, because it compels someone to do something or it forbids something, there should be a specific enactment by the legislator so as to ensure that the effective protection of basic rights. Any encroachment without an imperative character should not require a specific enactment.
Relevance of Accessibility to Norm Setting?
This line of arguments in a way bypasses Kümper’s argument that the glycol wine and Osho doctrine only allowed government action to disseminate information, if the task of the government was not accessible to norm setting (and if the encroachment is the equivalent of a “classic” interference, which will be discussed below). The Court could do so because in its correct view there were other constitutional reasons why government action to disseminate information does not have to be based on a specific enactment.
This is revealed by a careful reading of the Osho decision. Here the Federal Constitutional Court held that the government could call the so-called Osho movement a “sect” and a “youth religion” (cf. Osho decision, marginal numbers 56 and 94). The relevant basic right at stake was the freedom of religion as protected in Article 4 GG . The court started its analysis of whether calling the Osho movement a “sect” required a specific enactment by saying that informing the public about the Osho movement – which was at the height of its influence at the time of issue of the government information which was subject of the dispute – was the task of the government. The government had the task of providing information to the public “in order to facilitate the solutions of conflicts in state and society”, in particular in cases where there was a lack of available impartial information. It deduces this task from the democracy principle of the constitution (cf. Osho decision, marginal number 74, 75). Only as an additional reason the Federal Constitutional Court used the argument that a specific enactment cannot be required if the way the task is to be carried out is not accessible to norm Setting. At the same time it stressed that the relevant question is this: can a specific enactment contribute to the fulfillment of the aims of the doctrine, which lay in the democracy principle and the rule of law (cf. idem., marginal number 78)? While it is true that the Federal Constitutional Court then bases its judgment on the notion that government action to disseminate information was supposedly not accessible to norm setting (cf. idem, marginal numbers 79 – 82), one cannot say that this was the court’s only argument, because the question of whether or not the democracy principle or the rule of law require a specific enactment cannot only be answered by the question of whether it is possible to regulate a task or not. Therefore, it is not enough to say that the task of the Coordination Office can be regulated by an enactment in order to conclude that specific enactment is required. The question is whether the constitutional principles mentioned above demands an enactment.
Glycol Wine Decision
The glycol wine decision is not based at all on the argument that the task was not accessible to norm setting. This shows again that there are arguments for not applying the doctrine of the requirement of specific enactment other than the question if the task of running the database can be regulated. In the glycol wine decision the Federal Constitutional Court said that truthful and objective information about a product (in that case that certain wine had a negative characteristic) did not affect the occupational freedom of the wine seller at all (cf. marginal numbers 48 – 50). In short: it held that whether they like it or not every market participant has to live with information from a number of market agents all the time, just as long as it is truthful. The court did not see why the democracy principle or the rule of law demanded more restraint from the government than from other agents who are allowed to influence the market. Thus, only untruthful information would constitute an encroachment on occupational freedom, and then one which would right away be unlawful (cf. idem, marginal number 63). Arguably the government has to fulfil a higher standard of accuracy and objectiveness than others, but it does not have to remain quiet altogether.
So from the point of departure that led the court to conclude in Osho that an encroachment by information was lawful, it concluded in Glycol Wine that there was no encroachment at all. I wonder if this differentiation is necessary. It makes things more complicated, and it would have been possible to come to the same result if the Federal Constitutional Court had concluded that the information about glycol wine was a lawful encroachment. However, I have not explored what this argument would do to the legal edifice around occupational freedom.
The Federal Constitutional Court’s conclusion applies as well in the case discussed here: The plaintiff is a limited liability company (GmbH) liquidating its former art trading business. Its main purpose is to participate in the art market, e.g. by selling art. It has an interest in doing so without information that is negative for the value of the works it intends to sell. But as long as the information disseminated about the work is correct and objective this interest is not constitutionally protected. So in this case there is not only no need for a special enactment that allows an encroachment on a basic right. There is not even an encroachment (or at least a lawful one).
Note from the Editor: Even if you prefer Henning Kahmann’s approach over Boas Kümper’s analysis, it would only follow that specific legislation is not required from a constitutional perspective. Of course, it would be possible to implement legislation. From a policy perspective, it would in my perspective be preferable to do so and to put the lostart.de database on a sound legal footing. This could be on a stand-alone basis for the database, or as part of a more comprehensive piece of legislation, perhaps even a Restitution Act.
* In addition to the case notes already mentioned in earlier posts, the decision has been commented upon by the presiding judge of the case, Berlitt in Steinkühler, jurisPR-BVerwG 10/2015 Anm. 5; by Döring, a judge in the case, in juris – Die Monatszeitschrift 2015, 252; as well as by Bischof in Kunst und Recht 2015, 14 – 15; Elmenhorst in Handelsblatt February 24, 2015 “Lost Art. Grundsätzliche Fragen an die Datenbank”
** Sachsen-Anhalt used to run the Lost Art Database through the Magdeburg Coordination Office (Koordinierungsstelle) at the relevant time. Today, the Stiftung Deutsches Zentrum Kulturgutverluste is operating the database.
Source of the photo: BBC via Wikimedia.