Who Is Who?

                              Group Rights in Liberal Austria and the

                                Dilemma of Classificatory Procedure


                                          Jeremy King[1]


            In imperial Austria, the years between 1905 and 1914 saw half a dozen attempts at
compromise between the country's territorially intermingled and politically opposed peoples—its
Czechs and Germans, for example, and its Poles and Ukrainians. The compromises between these
"races" or "nations" amounted to a dramatic response to a worrisome challenge. In 1897,
long-building Czech-German conflict had erupted into demonstrations and riots, and brought down the
government. Over the following years, Austria's Parliament and some of the seventeen crownland
legislatures, or Diets, had become lamed by unstable majorities and obstructionist minorities. Now,
in the compromises, which took the form of crownland legislation, elites sought to reduce conflict
by separating the peoples—not territorially but politically, and not hierarchically but on the
basis of group equality of rights. At the same time, those elites tried to preserve the core,
liberal principles of Austria's constitution of 1867: individual equality of rights, and liberty.
They tried to reconcile legal equality for peoples with legal equality for persons, and both of
those equalities with racial or national difference. Austria undertook an experiment with what
might be called "separate but equal." Unlike the American experiment known by that name, however,
the Austrian experiment was in earnest. "Equal" was not a lie.

            Through the Moravian Compromise, in 1905, Czechs and Germans in the crownland of
Moravia were segregated into mutually exclusive school systems, on the basis of national equality
of rights. Czechs ran Czech schools for Czech children, and Germans ran German schools for German
children. The electorate for the Diet was also divided, such that Czechs voted for Czech
candidates, and Germans voted for German candidates. The number of representatives for each people
was set in advance, again on the basis of national equality of rights, and the contingent of German
representatives, now no longer in the majority, was equipped with strong minority protections. Next
came a compromise in the Bukovina. After 1909, elections to the Diet of that crownland were
conducted separately for Ukrainians, Romanians, Poles, and Germans. In Bohemia, Czech and German
schoolchildren had already been segregated, with careful attention to equality, before 1900. Now
separation was planned of Bohemia's electorate and of more as well. Negotiations over a Tyrolean
Compromise, separating Germans from Italians, were advanced, but stalled. A Galician Compromise,
dividing Polish voters from Ukrainian ones, became law in 1914. Implementation had to be postponed,
however, because of the outbreak of war.[2] When war ended, imperial Austria did too, and with it,
the compromises. If all of them had come into force, they would have directly affected almost
twenty million citizens.[3]

            This essay focuses on one of the dilemmas which resulted from group rights in a liberal
state: the dilemma of classificatory procedure. The group rights created in Austria after 1905
required formal classification. They required determining, in precise, legal fashion, who was Czech
and who was German, who was Polish and who was Ukrainian. But how can government mandate
classification into nations or other kinds of peoples, in support of group equality of rights,
without infringing on individual rights, especially the right of voluntary association? Or rather,
how can government mandate national, racial, or ethnic classification, in support of group equality
of rights, without infringing on individual rights any more than necessary? Classificatory
procedure proves to be not just a technical matter, but an arena where group rights collide with
individual ones. As Paul Starr has noted, "Although the conventional, hard-nosed view is that
politics is about 'who gets what,' the prior question is who 'who' is."[4] Different classificatory
procedures can produce different answers to "Who is who?," and can structure collisions of group
rights with individual ones in different ways. This essay approaches the dilemma of classificatory
procedure by digging into the Moravian Compromise of 1905. It closes by arguing that the Austrian
experiment with liberal group rights a century ago may contain problems and techniques relevant to
Affirmative Action today.


I. National Classification in 1905

            The Moravian Compromise, a set of laws passed by the Moravian Diet and sanctioned by
Emperor-King Francis Joseph of Habsburg in November 1905, established group rights above all in
representative government and in education. The Czech and German nations in Moravia were guaranteed
73 and 46 seats respectively in the Diet. They were also guaranteed representation in specific
ratios on Diet committees and on Diet-appointed boards of crownland institutions, and considerable
control over the Czech-language and German-language public school systems. This required a national
partitioning of several subsets of the citizenry. The adult male citizenry was separated into Czech
and German electorates for the Diet. In other words, it was divided between two sets of electoral
districts, with each set covering Moravia.[5] (A third, nonnational electorate contained 200-odd
great landowners, who had the right to vote in one of the national electorates as well. Its 30
seats, as well as 2 seats filled ex officio by Moravia's bishops, combined with the Czech and
German seats to yield a total of 151.) School-age children were separated into Czech and German
schools. Because the local, district, and crownland school boards that oversaw those schools were
separated at the same time, members of those boards were separated nationally, too. So were the
employees of nationally divided crownland institutions, which included public secondary schools.
Only adult females and children under the age of six—ineligible to vote, to serve on school boards,
or to attend primary or secondary school—were not separated officially into Czechs and Germans. The
rest of the 2.5 million Austrian citizens resident in Moravia were.

            The group rights created through the Moravian Compromise required formal
classification. But the Compromise specified classificatory procedures only incompletely. Indeed,
as a judge noted in 1906, "Regarding the procedure for determining the national belonging of
members of the school authorities, no particular provisions are contained in the law."[6] What,
then, did the Compromise specify about the "national belonging" or "nationality" of school board
members?[7] And what provisions did it contain concerning the national classification of
schoolchildren and of voters?


IA. Local School Board Members

            Beginning in 1869, long before the Moravian Compromise, Austrian law mandated that
municipal governments provide a free education for all resident children between the ages of six
and fourteen. At first, crownland school boards possessed full discretionary authority to decide in
which language that education would be provided in each municipality. (Thanks to national tensions,
above all between Czechs and Germans in the neighboring crownland of Bohemia, bilingual instruction
was prohibited by Article 19 of Fundamental Law 142 of 1867, whose final sentence read as follows:
"In those crownlands inhabited by more than one nation, public institutions of education shall
enable each of the nations to be educated in its language, without being compelled to learn a
second language of the land."[8]) From the 1870s, however, as will be explained in Section IB,
court rulings had set progressively stricter procedural standards regarding the determination of
the language of instruction in public elementary schools.[9]

            To cover the cost of the education, each municipal council levied property taxes—which
were paid only by a wealthier minority of male residents. By no coincidence, that minority made up
the electorate for the municipal council. And the municipal council appointed most members of the
local school board, which drew up the school budget. This arrangement kept property taxes down, by
preventing most parents from having a say in how much money was spent on the education of their
children. It protected individual property rights. But in binational municipalities, the system did
not prevent the national majority on the council and on the local school board from allocating a
disproportionate amount of money to those schools which used the majority's language. It did not
protect national rights. In Moravia until 1905, Czech-dominated municipalities often discriminated
against their German schools, and German-dominated municipalities often discriminated against their
Czech ones. Such were the realities of nineteenth-century Austrian liberalism.[10]

            It was in part to prevent such discrimination that the Moravian Compromise divided
school boards nationally. But only the Austrian Parliament could create a new funding mechanism for
schools, or a new set of appointment procedures for school boards. The Moravian Diet, rather than
waiting for Parliament to act, did what it could itself. In the Compromise, it wrapped new,
crownland provisions around the model laid down for all of Austria in 1869. Henceforth, in Moravia,
binational municipalities would have both a German and a Czech school board, and "The
representatives of the municipality for both local school boards must be taken from members of that
nation for which the school represented by the local school board is meant." A similar requirement
was added for some members of the district and crownland boards.[11] (For other members at
non-local levels, and for the employees of nationally divided crownland institutions, appointment
procedures were nationalized in a different way, discussed in Section IIIC.) Funding for all public
elementary schools continued to come from municipal property taxes. But the men who drew up the
budget for Czech schools had to be Czech, and the men who drew up the budget for German schools had
to be German.

            This approach turned out to have problems. To be sure, the local school board of the
minority could assert itself against the municipal council. To cite a ruling in 1911 by Austria's
Administrative Judicial Court, the local school board of a particular nation had the right "to
demand and to ensure, even in opposition to what may be divergent interests of the municipality
whose obligation it is to pay for the schools, that the schools of the nation represented by it
receive such resources as are required per legal directive, or that these schools are not stunted
through the withholding of any legally mandated resources."[12] But what if the national majority
on a municipal council appointed men to the local school board of the other nation who only claimed
to belong to it, while actually sympathizing with the other side? Or what if they did belong to it,
but cared less about investing in public education than about keeping taxes on private property
low?

            Such things came to pass after 1905, spawning disputes. Who was Czech, and who was
German? How Czech or German should a man be to count as a legitimate appointee to the relevant
school board? The Diet had provided no procedure in the Compromise for resolving such disputes.
Instead, it seems to have assumed that every person belonged to one nation, and that it was
self-evident to which one.[13] Preceding group rights had been what Rogers Brubaker has called
"groupism": a tendency to take "discrete, bounded groups as basic constituents of social life,
chief protagonists of social conflicts, and fundamental units of social analysis."[14] Members of
the Diet assumed the existence of national groups, and did not see that they were in fact helping
to create them. They based formal classification of local school board members into nations on
informal, national understandings of nationhood. And in some cases, those understandings proved to
be too vague or inconsistent for legal purposes.

            Groupism had led to the same problem before. The parts of the Moravian Compromise which
concerned local school boards were closely modeled on legislation, the first of its kind, enacted
in the neighboring crownland of Bohemia more than a generation previously. In 1873, in the minority
of Bohemian municipalities which were nationally mixed (German and Czech), public education had
been divided nationally. Municipal councils had been required to appoint to German and Czech local
school boards only men who were "members of that nation for which the school represented by the
local school board is meant." Procedures for determining national belonging had not been specified,
and disputes had started immediately.[15] In 1905, the Moravian Diet committee which drew up
legislation to divide many more municipalities (Czechs and Germans were considerably more
intermingled in Moravia) perhaps failed to learn about those disputes in Bohemia after 1873. But it
is unlikely that the same mistake was made by the legal advisers in government ministries who
helped draft legislative bills, and who made formal recommendations to the Emperor-King as to
whether He should sanction those bills which Austria's various legislatures passed. After all,
knowing precedents and foreseeing complications was their job. Their review of the Compromise,
however, had to be unusually hasty—because political pressures, including clashes in the streets,
dictated that certain parts of the Compromise be written, passed, and sanctioned within days.
Another, slim possibility is that the groupist approach taken in 1873 was repeated in 1905 on
purpose. Perhaps key players decided to avoid the risk that legislators would get mired in disputes
over the proper procedures for determining national belonging, and calculated that a later lawsuit
would trigger the application of procedures worked out over the intervening years in Bohemia and
other crownlands by the Administrative Judicial Court.

            Whichever of these explanations is correct, the outcome is clear. Classificatory
procedures developed elsewhere in Austria, by judges working to fill in holes in the law, did come
to apply in cases involving local school boards in Moravia after 1905. What were those procedures?
Who determined a person's national belonging? What definitions and rules of evidence were used? Who
had standing to challenge the determination? Which body ruled on the appeal? And what were the
effects on individual liberty and equality before the law?

            A landmark decision had come in 1881, in a dispute over the national belonging of a man
appointed to the German local school board in a Bohemian town by the Czech-dominated municipal
council. National belonging, the Administrative Judicial Court had ruled, was "essentially a matter
of consciousness and feeling." "Therefore, if in a concrete case the national belonging of an
individual is contested and external manifestations of his national convictions are lacking, it
will surely be necessary to question him about his national belonging, and to treat him as a member
of the nation in favor of which he makes a declaration."[16] The "declaration principle" contained
in this statement made liberal sense. By placing self-classification at the base of determinations
of national belonging, the Court saved at least some of the individual right of voluntary
association. On the other hand, a declaration could be false. The Court, with its mention of
"external manifestations" of national convictions, seemed to imply that if such manifestations
clearly contradicted a person's declaration, then perhaps the declaration could be overruled.
Self-classification perhaps could be subjected to verification.

            The 1881 decision did not specify how verification should be carried out. It did
confirm who should do any verifying, though, by remanding the case to local administrative
authorities. It also made clear that appeal was possible, through the two upper tiers (the
crownland Governor's Office and the relevant Ministry) of the state administrative apparatus and
then on to the Administrative Judicial Court, as the final instance. Later decisions spelled out
more. Most important was that those decisions treated declarations as the decisive manifestation of
national convictions. If a declaration was suspected to be false, the person was to be
confronted—and taken at his word. In a Bohemian case in 1901, known in Austrian legal shorthand as
Budwiński 27A, the Court decided that "[I]f an individual's national belonging is in question in a
concrete case, then it seems that only his definite declaration, made in the matter in the course
of a formal hearing, can decide the issue." As a corollary, it disallowed verification of national
belonging by the authorities through inquiries into a person's public or private behavior. That
behavior, Budwiński 27A explained, often could not be read at face value, because of the pressures
that individuals faced to fit in or to get ahead:

the appeal assigns particular weight to the fact that in the most recent census, the four persons
elected [by the German-dominated municipal council to the Czech school board] entered 'German' as
their language of daily use. Against this, however, it must be remarked that, just as it cannot be
concluded that a person belongs to a race [Volksstamm] simply because he knows the language of that
race, so the language of daily use claimed by a person is not decisive for the question of his
national belonging, for national belonging as such is not even polled in the census. For that
matter, no more compelling conclusions regarding national belonging can be drawn from the fact that
the children of the person in question attend German schools, or from that person's membership or
non-membership in a voluntary association, or indeed from that person's participation in public
life—even if that participation should contradict his profession of national belonging. For all the
circumstances listed here can always have their roots in considerations of opportunity or of other
matters which are of concern to the person in question. They need not stand in any connection with
the question of national belonging.[17]

The procedures available to fill the hole in the Moravian Compromise regarding the national
classification of local school board members, then, were simple and clear. Self-classification,
carried out formally before administrative authorities, counted for quite a lot. Verification,
carried out  in quite passive fashion, counted for little.


IB. Schoolchildren

            Before 1905, the Administrative Judicial Court had also applied a variant of its
declaration principle to schoolchildren. In the 1870s and '80s, in several decisions, the Court had
restricted the discretionary authority of crownland school boards to decide which language would be
used in each public elementary school. The culmination, in 1886, had been a decision requiring the
creation of a school using the language of a particular nation if the municipality in question had
an average of forty schoolchildren belonging to that nation for five consecutive years. In 1896,
the Court had then ruled that in the determination of such children's national belonging, their
parents' declaration of national belonging was decisive. Parents exercised the right of
self-classification on behalf of their minor children. Another decision, in June 1905, had laid
down guidelines for orphans and for children in nationally mixed marriages.[18] Not only for school
board members but for schoolchildren as well, the Court had thus created safeguards for what it
could salvage, from nationally segregated education, of the individual right of voluntary
association. It had established an individual right of national self-classification, and checked it
only with a narrow procedure for bureaucratic verification.

            Now came the Moravian Compromise. The closest that it came to specifying a procedure
for determining the national belonging of schoolchildren was in a single sentence known as "Lex
Perek," or Perek's law, after Václav Perek, the Czech attorney who was its principal proponent in
the Diet. "As a rule," read Lex Perek, "only children proficient in the language of instruction may
be admitted to an elementary school." This had major implications. Since 1867, Article 18 of
Fundamental Law 142 had guaranteed individuals the right to educate themselves for an occupation as
and where they wished.[19] That had meant, among other things, that fathers could choose among
schools for their children. But now that right (the "parental right") was to be restricted.
Furthermore, the right of national self-classification established through rulings by the
Administrative Judicial Court was to be canceled for schoolchildren in Moravia altogether. Now not
fathers were to do the classifying but unspecified others, on the basis of language.

             Strangely enough, in November 1905, when the Emperor-King had to decide whether or not
to give His sanction to the Compromise, Austria's Minister of Religion and Education expressed only
mild concern about Lex Perek. "In and of itself," the Minister's report to His Majesty stated,
"this measure is not entirely beyond question, for it reduces the parental right to free
self-determination in this matter. Yet I do not believe myself obliged to object, given the
circumstance that this proposal was made by agreement of both nations of the crownland."[20] This
interpretation rested on groupist grounds, not on constitutional ones. The Minister justified
enactment of the Compromise by speaking as though it were already in force—as though members of the
Diet already represented one nation or the other, rather than all their constituents, regardless of
national belonging.

            Why did the Minister of Religion and Education seem to forget his responsibility to
protect the rights guaranteed by the constitutional laws of 1867? Despite what he said, it was not
in order to oblige two national movements when, for once, they could agree on something. Already in
1899, German members of the Diet, who were predominantly liberal, or "loyal to the constitution,"
had made clear that they would object strongly to infringement of the parental right. And after
1905, the German contingent would press often and hard for changes in how Lex Perek ended up being
implemented, arguing that the phrase "as a rule" in the law necessarily implied exceptions to its
language proficiency requirement for any schoolchildren whose parents so desired. Both the
Minister's words and the words of Lex Perek itself seem to have been intentionally confusing, so
that the Compromise could be concluded, but then turn out to mean not quite what some of its
authors had intended.[21]

            The fact is that Lex Perek was inserted into the Compromise at the insistence of Czech
members of the Diet, to compensate for certain national inequalities—not legal, but economic,
social, and demographic. Many of the other members of the Diet accepted it only reluctantly. Two or
three generations previously, in the 1840s, Czechs had tended to be considerably poorer and less
educated than Germans, and the Czech language had carried far less prestige than the German one.
(To repeat a point made in the Administrative Judicial Court's ruling from 1907 cited above,
speakers of Czech were not necessarily Czechs, nor were speakers of German necessarily Germans.
After all, many people were bilingual. For that matter, many people, whether bilingual or
monolingual, seem to have understood themselves simply as Austrians, or Moravians, or Catholics,
etc.) By the turn of the century, those gaps had narrowed. And they promised to narrow still more:
although the average German probably still outearned the average Czech, younger Czechs had pulled
even in education. In 1909, a Czech member of the Diet hurled at German members that "fables about
the tremendous value of German, about how a man can't become even a corporal without German, were
dismissed long ago, through bitter experience. Indeed, to the contrary, we [Czechs] may say that
without Czech, no one can become even the governor, or perhaps even a minister in the government.
For we have come so far that we will accept injustice from no one."[22]

            Another gap, however, remained stubbornly wide. The census (which was deaf to
bilingualism) might show speakers of Czech to outnumber speakers of German by roughly 72:28 in
Moravia and 63:37 in Bohemia. But in Austria as a whole, German-speakers outnumbered Czech-speakers
by 61:39. And in Central Europe more broadly, they outnumbered Czech-speakers by more than
10:1.[23] Speakers of Czech, by mastering German, could increase their employment opportunities
many times over. But the reverse did not hold true.

            These patterns help explain why, before 1905, many parents who spoke Czech and little
or no German had enrolled their children in German-language schools: to give them a better chance
in life. Lex Perek, despite its nationally neutral language, was meant by Czech leaders to give
them the right and the power to "reclaim" children for the nation—not only from the German nation
and its schools but from what those leaders understood as nationally apathetic, unprincipled, or
intimidated parents. "Czech leaders," to quote a German judge in 1912, have constructed for
themselves, quite literally, a 'right of the nation to its children,' which even the children's own
parents cannot deny to it." The Czech leaders saw matters differently. "We, too, assert," said one
of them in 1909, "that yes, parents have a right to determine how and where a child is to be
educated. But the bread giver, the factory owner, does not have a right to coerce, violate, and
terrorize the parents, such that they cannot do otherwise than to put their children in German
schools." Or as another Czech explained in the Diet, Lex Perek "is not a law by which the rights of
parents to their children are governed, but rather a sort of nationalities law. (Bravo! Cries: That
is true!) The object of the protection of this law is the nation, by which I mean here in Moravia
[u nás] both the Czech nation and the German nation."[24] German and Czech leaders both spoke a
liberal language of rights and freedom. But the Germans, understanding themselves as part of a
culturally and economically superior minority, tended to emphasize the individual, while the
Czechs, understanding themselves as part of an oppressed majority, tended to emphasize the
group.[25]

            Czechs in the Diet seem not to have anticipated that Lex Perek would generate disputes
over who belonged to which nation. They had anticipated disputes over schoolchildren, with parents.
And they had provided a procedure for resolution, to their advantage: excluding the parents, and
having national belonging determined on the basis of an "objective" criterion, proficiency in a
language. A clash between individual rights and group rights was to be resolved through a
particular classificatory procedure. When it came to resolving disputes between Czech leaders and
German leaders, however, that procedure had holes. To cite only the largest, what constitutes
proficiency for a six-year-old? Children cannot command a language as well as adults do. But what
children can do, and adults cannot, is acquire an age-appropriate and native command of a language
quickly and effortlessly. Depending in large part on who administered a language proficiency exam
to a child, how and when, the procedure could yield a determination of "Czech," or of "German."
Czech leaders, searching for an external manifestation of national belonging on which to base a
classificatory procedure which was objectively verifiable but also to their advantage, had turned
to a founding myth of their national movement: everyone who spoke Czech belonged to the Czech
nation, even if he or she did not yet realize it.[26] But without tightly woven classificatory
procedures, such groupism was no net with which to catch Czech-speaking children before they had a
chance to start spouting in German, too.

            As has already been mentioned, Lex Perek, by defining national belonging in terms of
language, contradicted the definition developed by the Administrative Judicial Court. The 1896
decision mentioned above had agreed with the Ministry of Religion and Education that, "in the
absence of legal criteria," the national belonging of children was determined on the basis of the
formal declaration of parents or guardians.[27] Now Lex Perek provided the legal criterion of
language. And in Austria, judges could only interpret laws, not rule them unconstitutional.[28]
(Judicial review would have meant the unthinkable: mere mortals deciding that the constitution or
one of the crownland legal orders, all gifts in the 1860s from Emperor-King Francis Joseph to His
Austrian subjects, nullified His sanctioning of new legislation.) Lex Perek thus cut Moravia's
schoolchildren loose from the Court's rules of classificatory procedure, without specifying new
ones. In addition to reducing the parental right, it exposed schoolchildren and their families to
the caprice of whoever ended up administering the language proficiency exams. Would it have the
effect of cutting loose school board members, too? Or would completely different classificatory
procedures exist side by side, one for fathers and the other for children? In a law journal in
1914, Rudolf Herrmann von Herrnritt, an expert in nationalities law who was a justice on the Court,
labeled Lex Perek "a fateful example of a slapdash and, frankly, irresponsible legal technique."
"This legal provision," he continued, "makes a drastic intervention in primary education, and
affects important interests of the population, yet was clearly inserted into the school law [of the
Compromise] in haste, and, it seems, not without a certain craftiness."[29]

            In May 1907, an implementation decree for Lex Perek was issued, after time-consuming
and fruitless efforts at finding a formulation satisfactory to both Czech and German leaders.
Through it, the new Minister of Religion and Education, a German liberal named Gustav Marchet,
mended many procedural holes, just in time for the start of the school year. The language
proficiency exam was to be administered by the director of the school in question. Parents, as well
as the local school board of the other nation, could appeal the director's decision, to higher
administrative authorities. "Proficient" was defined, very broadly: "A child is to be regarded as
proficient in the language of instruction if it commands that language well enough to follow
instruction in it." "As a rule" was exploited in order to allow many exceptions. Children could be
admitted to a school whose language they did not yet know if, for example, the parents provided
"convincing reasons."[30] In short, Marchet implemented the law in such a way as to let almost any
child through. Perek and other Czech leaders, outraged to see one of their key contributions to the
Compromise turned inside out, condemned the implementation decree as "unacceptable for the entire
Czech people." Challenges soon began to make their way toward the Administrative Judicial
Court.[31]


IC. Diet voters

            For Diet voters, the Moravian Compromise did specify a procedure for national
classification, in three steps. In every municipality, the municipal authorities, "on the basis of
personal relations known to them," had to assign each voter to one of the two newly constituted
national electorates. Any voters who wished to switch sides could then do so. Voters neither Czech
nor German were to be assigned to that Diet electorate which was larger in their municipality.
Again, though, they had the right to switch to the other.[32] These first two steps ensured that
every adult male not only had a right to self-classification but exercised it, at least passively.
They allowed the national division of the electorate to be completed without costing anyone his
right to vote.

            The third step was bureaucratic verification. Every voter had the right to demand
official verification of a co-national voter's national belonging. A Czech voter could challenge
the Czechness of any Czech voter, and a German voter could challenge the Germanness of any German
voter. State administrative authorities at the local level, working together with municipal
authorities, were to make the decision, which could be appealed up the usual chain of
administrative command. This other-classification, or ascription, was meant to prevent voters from
making a false declaration—perhaps because their employer threatened to fire them, or because they
wished to stir up trouble within the electorate of the other nation. Through the three-step
classificatory procedure of assignment, self-classification, and bureaucratic verification, the
right of the Czech and German Diet electorates to what might be called non-territorial,
non-Wilsonian national self-determination was to be balanced with the same right of individual
voters.[33]

            A definition of national belonging was missing. The Compromise stated only that in the
third step, the authorities should rule on a voter's national belonging "on the basis of a thorough
examination of all conditions."[34] This contradicted the Administrative Judicial Court's
declaration principle. Under the Court's rules of classificatory procedure for school board members
and schoolchildren, developed since the 1870s, investigations into national belonging had stopped
at the declaration of the relevant adult. Where would they stop now for voters?

            Overall, the Moravian Compromise dramatically expanded the volume and importance of
national classification in Austria. Now not only a small minority of school board members and of
schoolchildren in a crownland had to be classified, as had been the case in Bohemia since 1873, but
all schoolchildren and all adult males. At the same time, the Compromise specified classificatory
procedures only incompletely. What procedures it did specify or imply, furthermore, were
inconsistent—with each other, and with the procedures developed before 1905 by the Administrative
Judicial Court. The effect was to expose individuals to capricious or unequal treatment, to a loss
of rights.

            But the Moravian Diet was unlikely to clean up the mess it had made. For decades before
1905, Czech leaders and their "conservative" allies among great landowners had been in the minority
in the Diet. They had gained seat after seat, though, and by 1903 or 1904, had high hopes of
becoming the majority in the next elections, due in 1906. German leaders and their "liberal" allies
among great landowners, preferring an organized retreat to a rout, agreed in the fall of 1905 to
the Compromise. Through it, their opponents gained permanent control over as many as 56 percent of
the Diet seats. But the new minority acquired something close to a right of veto, through a
provision requiring that amendments to the Compromise pass by more than 61 percent. That, given how
much Czech and German leaders disagreed with each other, blocked change. Thus in the fleshing out
of how, exactly, national belonging should be determined, legislators left the administrative
authorities of the executive branch, in Justice Herrnritt's words, "completely in the lurch."[35]
As a check on capricious or discriminatory action by those authorities, there remained only the
judiciary. Forced to sort things out were the courts.


II. Refinements to National Classification by the Courts, 1905-1914

            The first dispute over classificatory procedure in the Moravian Compromise to end with
a landmark court decision concerned voters. There the organization of justice in Austria further
complicated an already complicated matter. In disputes involving "political rights," which included
the right to vote and the parental right, the final arbiter was not the Administrative Judicial
Court. Rather, it was a parallel institution, the Supreme Court.[36] Before 1905, political rights
had not been tied to national belonging anywhere in Austria. Thus the Compromise, in addition to
counting as the first legislation to specify a procedure—indeed, multiple procedures—for
determining national belonging, broke new ground in a jurisdictional sense as well. The first high
court to confront the dilemma of classificatory procedure in the wake of the Moravian Compromise
was not the Administrative Judicial Court but the Supreme Court, which had confronted that dilemma
only rarely in the past.[37]

            The details of the case were as follows. In January 1907, an Austrian law applied that
part of the Moravian Compromise which separated Diet voters nationally to the same men again, now
in their capacity as voters in elections to the lower house of the Austrian Parliament. Before
parliamentary elections took place in May, and probably even before Diet elections took place the
previous autumn, a number of men in Moravia's capital, Brünn/Brno, succeeded in enrolling in the
German electorate, although they seem actually to have been Czechs—whatever that might have meant.
As members, they made extensive use of their right to challenge the national belonging of
co-national voters. They attempted to "reclaim" 3,000 German voters (of 18,000) for the Czech
electorate (which numbered less than 6,000). The administrative authority having jurisdiction, the
Moravian Governor's Office, did not undertake a "thorough examination of all conditions," as
stipulated by the Compromise. As a representative explained before long to a panel of Supreme Court
justices, the Governor's Office was overwhelmed by the number of challenges. But it "behaved
passively" for another reason as well: in order to abide by the declaration principle. It followed
the classificatory procedures laid down over the years by the Administrative Judicial Court, not
those newly established by the Compromise. All voters in question were mailed a request for a
declaration of national belonging. Then the Governor's Office denied not only the challenges made
against voters who declared themselves to be Germans, but also those made against 1,500 voters who
made no declaration at all.[38]

            Represented by a Czech attorney, three of the supposedly German reclaimers (named
Hausner, Diwišek, and Doležal), appealed the decisions of the Governor's Office. They demanded that
"not the declaration of the individual but objective characteristics be considered decisive for
national belonging." And in October 1907, the Supreme Court found in their favor. In Hye-Hugelmann
1531, as the decision was known, the Court ruled, on the basis of the Compromise, that "the
decision by the authorities may not rest solely on a declaration regarding national belonging
obtained from the persons whose enrollment is contested. Rather, the decision must rest first on
inquiries into objective characteristics determining national belonging."

            In the confidential deliberations which preceded this pronouncement, members of the
Supreme Court expressed deep reservations. Requiring inquiries into every case would impose an
impossible burden on the administrative authorities. The objective characteristics which determined
national belonging remained wholly unclear, especially for bilingual persons and for Moravia's
small Jewish minority. The "most holy right of the individual," to make free decisions, would be
violated. On the other hand, in the course of oral arguments, the attorney for the appellants
(speaking openly of the 3,000 challenges as having been organized by "the Czech party") had
threatened that if the declaration principle were upheld, the Czech majority in Moravia would
exploit it to invade the German electorate in certain districts, in large enough numbers to affect
elections.[39] At least three of the eleven justices found the law at fault, and in need of
revision. Only legislators and the Emperor-King, though, had the power to change existing law.
Meanwhile, that law was so clear that nine of the justices voted to interpret it as requiring
inquiries even when the person in question had made a clear declaration of national belonging. The
best they could do about their concerns was to give them tactful expression in Hye-Hugelmann 1531,
and to state that declarations continued to be "of the greatest importance," and could prove
"sufficient" for the determination of national belonging, provided that they were not contradicted
by notorious circumstances or by credible factual considerations stated in a challenge.[40]

            The appellants, in winning, sawed off the branch on which they were sitting. One
implication of Hye-Hugelmann 1531 was that they lost hope of using patently false declarations of
national belonging to prevent their own eviction from the German electorate. Their backers were
more interested in capturing "renegades" and "turn-coats" for the Czech electorate than in trying
to capture districts of the German electorate through Trojan horse tactics. A larger implication,
however, was that all voters in Moravia lost some hope of using genuine declarations to block being
transferred against their will from one national electorate to the other, on the basis of
"objective characteristics."

            What were those characteristics? In the confidential deliberations, the reporting
justice, Karl Ritter von Czyhlarz, met with no disagreement when he argued that the characteristics
raised in the Czech challenges (name, place of birth, schooling, language used at home, language
claimed in the census, membership in voluntary associations, schools chosen for one's children,
etc.) could not be considered decisive. Another characteristic, raised by the Czech attorney, was
dismissed outright: the accent with which a person spoke German. During oral arguments, the
representative of the administrative authorities had cited relevant excerpts from two recent
decisions by the Administrative Judicial Court (including those parts of Budwiński 27A cited in
Section IA of this essay). But Czyhlarz also read out loud, twice, the provision of the Moravian
Compromise specifying that the authorities should determine a voter's national belonging "on the
basis of a thorough examination of all conditions."[41] Hye-Hugelmann 1531 ruled no objective
characteristics out of bounds—not even those involving ancestry, which counted under the
circumstances as a particularly illiberal basis for other-classification.[42]

            To understand the perils of assigning individuals to peoples through investigations,
particularly ones involving ancestry, Austrians did not have to know about Jim Crow in America, or
anticipate Nazi Germany's separation of citizens into Aryans and Jews. In the fall of 1910, Edmund
Bernatzik, a justice on the Supreme Court, became the new Rector of the University of Vienna. In
his inaugural lecture, "On National Registries," he defended the declaration principle
passionately, thereby distancing himself from Hye-Hugelmann 1531 (in which he had not
participated). "From all manner of indications, including perhaps pub conversations, theater
visits, the reading of suspicious books, and who knows what other circumstances," Bernatzik warned,
"evidence for or against the national belonging is to be adduced. The 'genuine German man' will no
longer be able to enjoy the Frenchman's wine without running the danger of being officially branded
a 'traitor to the national cause.' There loom trials all too reminiscent of the tribunals of the
Inquisition. At stake here, after all, is the ascertainment of convictions! ...a person could be
sentenced by an authority or court to belong or not to belong to a particular nation. That is
absurd, and would be even more absurd if there were to be several mutually contradictory
judgments—which is entirely possible."[43]

            After 1907, the Supreme Court issued no major rulings regarding the national
classification of voters. Nor did it rule on the tension between Lex Perek and the parental right.
Only three weeks after Hye-Hugelmann 1531, however, the Administrative Judicial Court, ruling in a
Bohemian local school board case, made clear that it continued to stand by the declaration
principle. Overturning a decision by the administrative authorities, it noted that they were
"invoking circumstances on the basis of which a compelling conclusion regarding nationality cannot
be reached. The inquiries concerned the personal, home, and family relations of the elected
individuals. In particular, questions were asked about origin, nationality of parents reported in
the census, language of daily use of the elected individuals and their family members, national
upbringing and education of children, associational activity of the elected individuals, etc. None
of these circumstances, however, may be taken into consideration in answering the question." The
passage which followed comprised a slight modification of the passage already cited in this essay
from Budwiński 27A. Elsewhere, in another echo of Budwiński 27A, the decision stated that "[I]f an
individual's national belonging is in question in a concrete case, then it seems that only his
definite declaration, made before the authorities and thus presumably made with all earnestness and
consideration, can decide the issue."[44] This ruling made utterly clear to administrative
authorities that the investigations which were now required by the Supreme Court for determining
the national belonging of Moravian voters continued to be prohibited by the Administrative Judicial
Court in the case of Bohemian or Moravian local school board members. (All Moravian school board
members, of course, were also Moravian voters.)

            Three years later, however, the Administrative Judicial Court changed course. Budwiński
7846A, dated December 30, 1910, established a new procedure for verifying the national
self-classification of local school board members in Moravia. Despite the Compromise's lack of
clarity, the Court ruled, it had been the intent of the Diet "to give sufficient guarantee for the
actual appointment of nationally feeling men to local school boards." Thus enfranchised "members of
the nation for which the school represented by the local school board is meant have the right to
contest the appointment of any representative of the municipality for the relevant local school
board on the grounds that the person appointed does not belong to their nation." The verifying
authority of first instance was the district school board of the relevant nation. It had to base
its decision on "how the person in question is nationally active, and to which nation the person
feels that he belongs. If a dispute exists over the belonging of a person to one of the two
nations, or over the truthfulness and credibility of a person's declaration regarding his
belonging, then that belonging must be determined by graspable [faßbare] characteristics. For that
purpose, it is permissible to take into consideration actions from private, social, and public life
which seem credible and earnest manifestations of national belonging."[45] The Czech-dominated
municipal council of Trebitsch / Třebíč did not succeed in appointing its preferred candidates to
the German local school board.

            Budwiński 7846A marked the end of the declaration principle for school board members in
Moravia. Self-classification was now subject to verification far more invasive than what the Court
had allowed as recently as November 1907. Why this change in course? The justifications given in
the decision were uncharacterictically vague, confusing, or irrelevant—a smokescreen. A close
reading, however, undertaken over the next several pages, reveals a subtle effort by the
Administrative Judicial Court, not just in Budwiński 7846A, to unify classificatory procedures for
school board members, voters, and schoolchildren, and to do so around verification less active than
that specified for voters by the Supreme Court in Hye-Hugelmann 1531.

            The Trebitsch / Třebíč municipal council itself, one of the parties in the new case,
had attempted to unify classificatory procedures for school board members and voters, in original
but unconvincing fashion. In the summer of 1907, the German district school board had rejected four
appointees of the municipal council to the German local school board, claiming on the basis of
inquiries that they "are not to be regarded as members of the German nation." Appealing that
rejection, the council had argued that "only the declared will of the person in question can be
decisive." It had also argued that the school authorities were not legally entitled "to call into
doubt the belonging of the elected persons to the German race, once it has been officially
confirmed"—and that such confirmation had occurred when, in the parliamentary elections held in
May, the administrative authorities had allowed the men to vote as Germans.[46] In other words, as
in Brünn/Brno, the authorities had behaved passively, going no farther than requesting a
declaration by the persons in question of their national belonging. The council preferred passive
verification, a corollary of the declaration principle, to active verification, because passive
verification of the men in their capacity as voters had provided evidence that they were Germans.
But of course, by the time the appeal reached the Administrative Judicial Court in 1910, the
Supreme Court, in Hye-Hugelmann 1531, had disallowed passive verification for Moravian voters, and
mandated active verification instead, i.e., inquiry into "objective characteristics."

            In Budwiński 7846A, the Administrative Judicial Court did two things. First, it
prevented inquiry into "objective characteristics" from seeping into its own jurisdiction. "No
legal provision exists according to which [electoral] lists might serve as evidence regarding a
person's national belonging more generally, in matters unrelated to electoral procedures, and thus
might serve as evidentiary documents in all cases where the national belonging of a person comes
into question. Rather, in the constituting of school authorities,…which nation a person belongs to
is to be determined quite independently."[47] Second, it replaced passive verification for local
school board members with active verification—but of a new sort. The Supreme Court, constrained by
the "thorough examination of all conditions" clause of the Compromise, had required inquiry into
"objective characteristics." But the Administrative Judicial Court, constrained by no such clause
and more experienced in managing the dilemma of classificatory procedure, permitted "consideration"
of "graspable characteristics," especially "how the person in question is nationally active" and
claimed to "feel," if they seemed "credible and earnest manifestations" of national belonging.

            Budwiński 7846A jettisoned the declaration principle for local school board members,
but established an active verification procedure in its place which was liberal, in the German
sense of emphasizing the individual rather than the group and choice rather than ancestry—what a
man did of his own free will, rather than who a man was. To which clubs did a man belong? Did he
prefer one language over the other? Where did he send his children to school? How consistent were
his choices, with each other and with his declaration of national belonging before the authorities?
An Austrian scholar in the 1930s labeled this subset "objectivizations": markers not of a coerced
or ancestral and involuntary national belonging but of an evolving, voluntary one.[48] Without
mentioning Hye-Hugelmann 1531, the Administrative Judicial Court followed the Supreme Court in
embracing active bureaucratic verification, but focused it on the "truthfulness and credibility" of
self-classification, on internal, subjective consistency rather than on an external, "objective"
definition. Local school board members still lost some privacy. But they retained more of their
right to voluntary association, more of their right to choose between the nations.

            Local school board members made up only a very small minority of the population. But on
the same day it published Budwiński 7846A, the Administrative Judicial Court published Budwiński
7843A, concerning Moravian schoolchildren. On the basis of Lex Perek, this ruling overturned the
Marchet decree. It redefined linguistic proficiency, fairly strictly. ("According to the natural
meaning of the words, 'to be proficient in the language' means to command the language such that
one can use it in intercourse as a means of communication, and express in it one's thoughts and
ideas."[49]) It replaced the school director as examiner with a commission drawn from both the
Czech and the German local school boards. It determined that "It is not left to the discretion of
the school authorities to determine permissible exceptions" to Lex Perek. And in what Herrnritt,
the reporting justice in the case, later labeled a "bold" step, Budwiński 7843A recognized the
boards as national organs, "authorized to exercise the legal claim of their nation so as to ensure
that the children belonging by law to the schools of that nation are not taken from it."[50] The
Court snubbed Marchet, restored the active and groupist, "objective" other-classification based on
language that Czech leaders had wanted, and fleshed out procedures for its implementation.

            In a deeper sense, however, the Court sided with Germans. Lex Perek, Budwiński 7843A
determined, could not be used to deny a child admission to a school "if the child is not
'proficient' in the language of his own nation, but wishes to attend the school of his nation by
virtue of national belonging"—which, "for school-aged children, is the same as that of their
parents." In other words, "knowledge of the language of instruction has been established as the
objective characteristic of national belonging (refutable through counterevidence)…in the field of
elementary education." Counterevidence, however, "could not be provided through a simple
declaration regarding the national belonging of the parents." Rather, "in a dispute regarding the
truthfulness and credibility of the [parents'] declaration, that objective characteristic would
need to be invalidated by objectively graspable characteristics of national belonging."[51] In
other words, the national belonging of parents had to be determined through something like the
cautious, individualized procedure prescribed by Budwiński 7846A for local school board members. To
the classification of Moravia's schoolchildren on the basis of language, the Court added
classification on the basis of parental self-classification, subject to objectivizing, bureaucratic
verification. Czech leaders had intended Lex Perek to replace the declaration principle, but
Budwiński 7843A turned other-classification from a replacement into a complement. To be sure, it
also diluted the declaration principle, by giving verification much more bite. But the net effect
was to protect, in a legally more defensible fashion than the Marchet decree had done, much of
self-classification and of the parental right. Through its pair of rulings in December 1910,
furthermore, the Administrative Judicial Court succeeded in establishing classificatory procedures
for Moravian local school board members and fathers which overlapped in many ways with the
procedure established by the Supreme Court for those many men in their capacity as voters, yet were
more detailed, and different in important ways.

            To some degree, Justice Bernatzik's prediction, made within weeks of the oral arguments
in these two landmark cases, proved right. After 1910, courts did sentence persons to membership in
one nation or the other. At the start of every school year, Czech boards sought to "reclaim"
thousands of children. I STOPPED HERE, 7/19/07. I NEED TO LOOK AT 1911 QUESTIONNAIRES FROM THE
ARCHIVES, AND ACCOUNT FOR THE PRESENCE OF ANCESTRY-ORIENTED QUESTIONS. THEN, IN THE SECTION ON
ESTOPPEL, I NEED TO PULL SOME OF THE FINAL FOOTNOTE INTO THE BODY OF THE ESSAY, CUT THE QUOTE THERE
FROM 7843A, AND POINT OUT THAT THE VGH'S COMPLICATED STRATEGY WITH 7843A AND 7846A WORKED AGAINST
WHAT HERRNRITT WANTED: ESTOPPEL ACROSS FIELDS. And in 1911, administrative authorities responsible
for adjudicating disputes over the national belonging of schoolchildren and voters began using a
questionnaire. The questions concerned national belonging, national belonging of parents, schools
attended, schools attended by children, language used at home and in social life, language of daily
use claimed in the most recent census, membership in national associations, and other public
engagement in national matters.[52] No doubt some cases of false self-classification were
uncovered, and others prevented. But verification must also have resulted in the reversal of
self-classifications which it would be a mistake to call false. For some citizens, national
belonging simply did not matter much, or manifest itself in the either-or ways emphasized by the
Compromise. Although the numbers of such citizens cannot be known, they were certainly a
significant minority. And verification procedures, no matter how little they were used or how
rarely they resulted in reversal, threatened the privacy of everyone.

            Then again, the complex system of national classification which had taken shape in
Moravia by 1911 was no floodlit cage. The questionnaire offset ancestry with actions in the
present—and compares favorably with what American authorities demand to know today in order to
verify the marriage to a citizen of an applicant for a "green card," or permission to take up
permanent residence. More broadly, the Administrative Judicial Court chose repeatedly to refine the
sloppy classificatory procedures of the Compromise in ways which, perhaps as much as possible,
resolved clashes between individual and group rights in favor of the former. The first of the two
decisions issued in December 1910, for example, passed up the chance to define proficiency in terms
of a mother or first language. If parents who spoke only Czech could make their child proficient in
German—perhaps by turning to a private, German-language kindergarden—then they had the right to
enrol the child for first grade in a German public elementary school.

            Saddled with a mess, confined to interpreting (rather than striking down) the laws that
Austria's legislatures and Emperor-King had seen fit to enact, and forced to share the Moravian
Compromise with the Supreme Court, the Administrative Judicial Court had struggled to defend
individual rights. It had not attempted to keep alive the declaration principle, because the
Compromise contradicted that principle flatly, as well as pushed many citizens to make insincere,
inconsistent, or simply instrumental declarations of national belonging. But the Court had
standardized procedures for the national classification of school board members and schoolchildren,
and based them on some form of self-classification subject to objectivizing, bureaucratic
verification. Those procedures, furthermore, were consistent enough with the procedures mandated
for voters to be implemented through use of the same questionnaire. Informally at least, that
allowed the Court's objectivizing approach to discourage emphasis on ancestry in the verification
of voter self-classification.

            But had the Court done the best possible job? Was Josef Lukas, an Austrian professor of
law, right when he suggested in 1908 that self-classification subject to verification through
"investigation into political-national convictions by administrative organs" was "the lesser evil,"
the least illiberal approach possible, under the legislated circumstances, to the dilemma of
classificatory procedure?[53]


III. Alternative Approaches

IIIA. Estoppel for Self-Classification

            Herrnritt argued in 1914 that his court indeed had done the best possible job, given
the laws it had been dealt. Nonetheless, the outcome struck him as deeply flawed. Already in 1899,
he had argued that the declaration principle had the effect of replacing national belonging, which
he considered an "objective quality" of individuals, with their "subjective will" or "caprice" as
the foundation for certain legal claims. After 1905, to be sure, the declaration principle yielded
in Moravia to objectivizing, bureaucratic verification of self-classification. Yet the number of
individuals making declarations, as well as the incentives for making instrumental ones, increased
greatly. As Herrnritt wrote in 1914, "intentional deception regarding a person's national belonging
occurs quite often…." He warned that unless further legislative action were taken, "the whole of
nationalities law will be of problematic value, a mixture of right and caprice which, despite all
the 'compromise laws,' will hold national conflict in mixed-language crownlands in a state of
permanence." Badly needed, he argued, was a law that specified "who makes a declaration, to whom,
and in what form; whether and how long it is binding; whether it can be changed; and which persons
it concerns, i.e., whose declaration is decisive for the national belonging of children."[54]

            In Moravia after 1905, the Administrative Judicial Court had succeeded in establishing
formal procedures for bureaucratic verification, a form of other-classification. But it did not
establish formal procedures for the self-classification which preceded verification, apparently
because here the law offered no foothold. Thus Herrnritt appealed publicly to legislators, not only
in 1914 but in 1899 and in 1906 as well. And because legislators had formulated the Compromise
sloppily, he made detailed suggestions about what a new law should contain. His preference was for
a law which required everyone to make a declaration of national belonging once every ten years, in
the census, and which used those declarations to create "a sort of registry of the nations...to
which the administration would then refer in cases where the national belonging of the individual
was in question." Self-classification needed to be regulated in such a way, he argued, that it
determined a person's national belonging "for the entire field of administration for a longer
period of time, in advance." If a person called himself German for voting purposes in 1910, then he
should count as German also for schooling purposes in 1915—rather than being able to self-classify
differently, depending on the moment and the situation. (In Austrian law, to quote Lukas, "the
declaration of nationality serves always as the foundation only for an isolated legal claim of the
person who makes the declaration. Thus the possibility of abuse, of political maneuver, is much
greater."[55]) Should such a strict solution prove unacceptable to the legislature, Herrnritt
indicated that he would consider a less strict one better than doing nothing at all. Key was that
legislators give self-classification at least some estoppel value. Or to use Herrnritt's own, less
technical term, he wanted them to make self-classification more "binding."[56]

            Extensive estoppel for self-classification, Herrnritt's preference, would have
infringed on individual rights more, not less, than objectivizing, bureaucratic verification. By
defining almost any attempt at changing national belonging as individual caprice, it would have
reduced the right of the individual to self-classification. It would have expanded a right of the
nation to its members. In 1914, however, it became clear that extensive estoppel had little chance
of being enacted. In February of that year, fifteen Czech members of the Moravian Diet presented a
bill concerning the national classification of children. Modeled, like the more open-ended of
Herrnritt's two proposals, on an 1868 Austrian law regulating religious affiliation, the draft
legislation would have assigned everyone to a Czech or German registry at birth, according to the
belonging of the parents. Transfer would not have been allowed until the age of fourteen, and
national belonging would have decided which national school system the child entered. German
legislators, however, torpedoed the bill.[57] Even if it had somehow cleared the 61 percent hurdle
for amendments to the Compromise, sanction by the Emperor-King would have been far from assured.

            In theory, more limited estoppel for self-classification, if introduced as a
replacement for bureaucratic verification, could have undermined individual rights less than it
shored them up. It would have reduced the right of self-classification (and hence of voluntary
association) somewhat, but restored to citizens their right to privacy. The elimination of some
individual caprice from the exercise of legal claims based on national belonging might also have
reduced national conflict. Theory was one thing, though, and practice another. The legislative
process values consistency and restraint far less than does the judicial process, as Herrnritt
himself knew all too well. By still calling for legislation which established some degree of
estoppel for national self-classification, he signaled that he was not only liberal but national,
not only individualist but groupist in his convictions.[58]


IIIB. Balancing Rights with Responsibilities

            The Moravian Compromise established national belonging as a legal status, but did not
recognize as legal entities the national groups implied thereby. Capable of filing suits and
otherwise defending the rights of a nation were only its individual members, each of whom acted as
"the occasional representative of the anarchic whole left unorganized by the state."[59] From a
legal perspective, Moravia's nations were only loose and unlinked lists of schoolchildren and
voters, without executive organs capable of expressing a national will. To be sure, the
Administrative Judicial Court's "bold" ruling regarding Lex Perek made the local school boards into
such organs, for limited purposes. But group rights still amounted mostly to rights exercised by
individuals. That approach greatly limited the rights that each nation could have, and how well
those rights could be defended.[60]

            Just before the Moravian Compromise, in the aftermath of the Czech and German riots
that had shaken Austria in 1897, Social Democratic legal theorists had begun to advocate a far more
radical approach to group rights. Austria's nations, Karl Renner had argued, had to be constituted
as legal entities. Every citizen had to belong to one and only one of them, and they had to be
"endowed in public and private law with executive and legislative capacity" in schooling, taxation,
and additional fields. The state had to turn nations, "communities of persons," into organizations
of persons—into actual groups, not just imagined ones. It had to give nations rights and
responsibilities vis-à-vis each other, their members, and the state in all national matters which
were compatible with a non-territorial organizational form. The result of such "national autonomy,"
supposedly, would be peace among citizens.[61]

            Social Democratic theorists considered only one classificatory procedure compatible
with national autonomy. In Renner's words, "Regarding national belonging, nothing can decide the
issue but the free declaration of national belonging by the individual before the competent
authority. This right to self-determination of the individual forms a corollary to the right to
self-determination of the nation." Self-classification would be almost pure; there would be
estoppel across fields (voting, schooling, etc.), but none across time, and no verification. Yet
false declarations of national belonging would be very rare. That was because national autonomy
would balance national rights of the individual with responsibilities—the responsibility to pay
national taxes, to obey separate bodies of national law, and so on. Each national right of the
individual would also count as a group responsibility, and each group right as an individual
responsibility. What estoppel could do only mechanically, national autonomy would do organically:
bind individuals to a nation. "If the declaration of national belonging is not something one does
for amusement over a mug of beer on Sunday," argued Renner, "but rather a legal institution, a
right and a responsibility, then that declaration demands serious reflection. …. Because Herrmann
von Herrnritt does not attach any important legal consequences to the declaration of national
belonging, the freedom to change national belonging at any time, at will, strikes him as
alarming."[62]

            James Madison, in 1788, had proposed to ensure accurate reporting from the American
federal states regarding the number of their residents, without verification, by tying that number
not only to a right (seats in the House of Representatives) but to a responsibility (contributions
to the federal budget). Now, by similar means, Renner proposed to encourage accurate reporting from
individuals regarding their membership in non-territorial federal units, i.e., in intermingled
nations functioning as "states within the state." This was revolutionary. It was also utopian.
Justice Bernatzik, in 1912, dismissed Renner's proposal as a "novel about the state."[63]

            Even if national autonomy could have been implemented, it would have infringed on
equality before the law, by making individuals in Moravia equal before separate laws. It also would
have confirmed and even exacerbated social and economic inequalities. Czechs, because economically
inferior to Germans, would have had worse schools, higher tax rates, or both. Otto Bauer, a Social
Democrat who embraced Renner's theories, admitted the German bias of national autonomy openly.
"Without doubt," he wrote in 1907, "under the rule of national autonomy in its most complete form,
the old historical nations [including the German one] would retain a certain superiority; without
doubt, through the dazzling development of their cultural institutions, through the lower taxation
burden on their members, they would also exert a powerful force of attraction on the members of
other peoples under this constitution and would thereby be able to achieve national conquests in a
peaceful way."[64] Austrian Social Democrats would have "solved" the dilemma of classificatory
procedure—and, supposedly, reduced national conflict—by pushing Czechs to choose between a right to
be disadvantaged and a right to become Germans.


IIIC. Verification by National Jury?

            Another approach to the dilemma of classificatory procedure was tucked into a corner of
the Compromise. What might be called verification by national jury came in two versions. The first
was used for Czech and German members of the Diet. The Compromise did not stipulate that Diet
members had to belong to the electorate which elected them. But it did partition the Diet into
three bodies, or "curias"—great landowning, Czech, and German—so that each could appoint men to
Diet committees and to the boards of crownland institutions. Here, complementing the local school
boards, was another national organ, also with narrowly defined powers. New members of the Diet were
assigned to a national curia "according to the nationality of the electoral district in which they
were elected."[65] This procedure might seem to have suppressed self-classification. But each man
had classified himself by courting voters of a particular national belonging. Verification,
meanwhile, had taken the form of those voters, a very large jury of his national peers, electing
him. Not administrative authorities but national will, as expressed at the ballot box, prevented a
Czech from infiltrating the German curia, or vice versa.

            The second version of verification by national jury was embedded in the procedures for
appointing some members of nationally divided school boards at the district and crownland levels,
as well as for hiring personnel at nationally divided crownland institutions, such as public
secondary schools. Those procedures varied in their details, but all involved two steps. The
relevant national curia or the relevant national members of the Crownland Committee nominated three
times as many men as there were positions to fill. (The Crownland Committee, which oversaw the
affairs of the Diet, consisted of nine members of the legislature. Four were chosen by the Czech
curia, two by the German curia, two by the great landowning curia, and one, the chair, by the
Emperor-King.) Then either the Crownland Committee as a whole or the Emperor-King, bound by that
list, appointed the members. The first step amounted to a national jury drawing up a short list of
qualified candidates, presumably with national belonging as one of the criteria. By nominating a
person, the jury verified the national self-classification that his willingness to serve implied.
The second step amounted to a binational and/or nonnational agent appointing those candidates which
it considered most qualified. Here, as in the appointing of members to local school boards by
municipal councils, one of the criteria was how much the men who would guide or staff a public
institution which served only one section of the public enjoyed the confidence of all sections, and
especially of their taxpaying elites—which footed the bill.[66]

            Christopher Ford, an American jurist with expertise in race-conscious law, has noted
that for liberals, other-classification into races creates some need for a classifying organ which
gives no reasons for its decisions. Building on Tragic Choices (1978), a study by Guido Calabresi
and Philip Bobbitt,[67] Ford argues that "As with the Calabresian example of decisions allocating
the scarce resource of artificial kidney machinery, race classification is a type of necessary
decision-making that is most conveniently done in the shadows…. [Consigning such power] to the
hallowed secrecy of the jury room can 'make the grounds for decisions less direct and perhaps even
less obvious, while at the same time trying to make sure that the decisions are based on broadly
held social values.'" These last words, taken from Tragic Choices, are followed with another
quotation: "'Juries apply societal standards without ever telling us what those standards are, or
even that they exist. This is especially important in those situations in which the statement of
standards would be terribly destructive.'" In contrast to an 'aresponsible,' jury approach, Ford
writes, a bureaucratic approach "invites appeal from adverse decisions, thus exposing the
process."[68] The relevance of his analysis—which inspired the terms "bureaucratic verification"
and "verification by national jury" in this essay—should be clear. Better than bureaucratic
verification, verification by national jury skirted the pitfalls of definition, explanation, and
appeal, as well as inquisitorial invasions of privacy. It produced clear decisions, but also held
out hope to the disappointed that next time, things might turn out differently.

            And yet, verification by national jury could not have replaced bureaucratic
verification in Moravia. First, local school board members could not have been elected by a local
subset of the relevant national Diet electorate, because the Austrian school law of 1869 mandated
their appointment by the municipal council. In theory, they could have been nominated to that
council by a suitable pair of national organs. But no such pair existed at the local level.
Creating one would have required dividing more of politics nationally—municipal councils, for
example. And that expansion of national rights, like Renner's more radical one, inevitably would
have caused a reduction in individual ones. In any event, German leaders would not hear of such a
reform, because they feared that certain elements of Moravia's municipal electoral order which
favored them (such as public voting and weighting of votes by wealth) would not survive the
process. That and the 61 percent hurdle for amendments to the Compromise made such a change
extremely unlikely.

            Second and third, children had a right to attend school in one of the two national
systems, while men had a right to belong to one of the two national electorates. But verification
by national jury could work only in connection with national privileges, not with national rights.
A Czech jury could verify that certain individuals eager to serve the Czech nation in one position
or another were Czech. And a German jury could verify that certain individuals eager to serve the
German nation in one position or another were German. But when the goal was not to honor a few
individuals with positions but rather, to divide many thousands of individuals between Czech and
German electorates or school systems, the dilemma of classificatory procedure reasserted itself.
Who would decide which of the two juries had jurisdiction in any given case? If the individuals
themselves, or their legal guardians, that would reduce the procedure to farce. Parents intent on
enrolling their children in German schools would simply choose the German jury. If someone else,
then who, and on what basis? Verification by national jury could serve to confirm that a person was
Czech, or that a person was German. But persons who declined nomination to a national position by
no means classified themselves implicitly as belonging to the other nation. Verification by
national jury could not serve to determine where the line between the nations ran.

            Verification by national jury was not a viable approach to Austria's dilemma of
classificatory procedure. It was an attempt at verifying national self-classification in national
fashion, and thus an exercise in groupism. To cite Ford again, "The problem with group
classifications arises where the categories employed are neither genuinely 'objective' (i.e.,
scientifically determinable) nor susceptible to legitimation by reference to a decision-making
entity (such as a Native American 'tribe') whose authority to make such decisions is unquestioned."
In such cases, other-classification cannot work "without presuming the very thing formal
classification seeks to accomplish: the division of the population into discrete groups of
clearly-differentiated individuals. Where there is no identifiable 'base population' with a formal
structure permitting collective in-group decision-making, the Native American approach to
classification by member-reference will be of no use."[69]

            Renner's and Herrnritt's approaches were exercises in groupism, too. To Renner, nations
were "communities of persons." To Herrnritt, national belonging was an "objective quality." Both
men were confronted with strong evidence to the contrary. Yet, in a testimony less to personal
failings than to the power of groupism in late imperial Austria, they failed to see beyond groups,
to the dilemma of classificatory procedure.

            In 1914, Herrnritt expressed deep misgivings about the compromises. "The road taken…,"
he wrote, "must lead to a gradual alienation of the nations from one another, and, worse still,
from the idea of a unitary state." Given that legislators and the Emperor-King had taken that road,
however, Herrnritt urged that they travel farther along it, and mandate tighter classificatory
procedures. "As long as Austrian legislators lack the courage, through appropriate regulations…to
organize the nations of the state and to determine and secure reliably who their members are, every
attempt at an ordering of the legal relations of the nations and their members will remain only a
half-measure, sure to lead to continually renewed struggle."[70] Less individual caprice and more
group rights, he seems to have thought, would help Austria reduce national conflict—even though
thus far, such shifts in balance probably had made national conflict worse, and certainly had made
individuals less free.

            In 1994, Ford saw the threat of a similar hardening of difference in the United States,
as a consequence of Affirmative Action. For him, though, classificatory procedure counted not as a
solution but as one of the dilemmas of liberal group rights: "a coherent system of group
benefit-allocation requires more attention to the process of classification than has hitherto been
given, but this attention may solidify the very social divisions beyond which preferential programs
were ostensibly designed to move us. Many will thus doubtless suggest that there is something
profoundly wrong with an…ethic which calls forth such jurisprudential segregation and brands badges
of racial identity onto the face of public life, insisting that the real problem lies deeper in the
group-essentialist paradigm rather than merely at the level of procedural propriety."[71] The
constitutional order prevented Herrnritt and other Austrians with "courage" from putting their
solutions into effect. To the extent that liberal principles are worth defending, that was probably
for the best.


IV. Conclusions

            What conclusions can be drawn from the dilemma of classificatory procedure, as it was
manifested in one of the Austrian compromises between 1905 and 1914? First, nations, races, and
other kinds of peoples are not best understood as groups. Peoples are above all cognitive
categories, widely shared views on the world which work in many ways, including through "systems of
classification, categorization and identification, formal and informal." So writes Rogers Brubaker
in Ethnicity without Groups (2004).[72] This essay complements and confirms his arguments by
showing in a specific instance that classification did not simply articulate peoples which were
already "out there" as groups. Rather, classification rested on an understanding of peoples as
groups, and thus contributed to constituting them more as such.

            Second, the Moravian Compromise encourages pessimism, if not despair, about how well
group rights can be reconciled with liberalism. Will Kymlicka, in his influential study,
Multicultural Citizenship (1995), discusses the decision of the U.S. Supreme Court in Brown v.
Board of Education (1954), that "Separate educational facilities [for Negro and white children] are
inherently unequal." Invoking it, American and even Canadian courts have struck down legislation
providing separate institutions for Native Americans and other "national minorities"—which Kymlicka
distinguishes from racial ones. This "overgeneralization of Brown," he finds, is "unfortunate, and
unjust." "Nothing in the judgement warrants the claim that national rights are incompatible with
liberal equality."[73] In part, the Moravian Compromise confirms this argument. After 1905, the
equality before the law guaranteed to citizens by the Austrian constitution suffered little
damage.[74] Unlike Jim Crow, the Compromise was a genuine attempt at "separate but equal." But as
has already been mentioned, bulking up Austrian group rights past the quite modest levels of the
Compromise, in the direction of Renner's national autonomy, would have entailed considerable
infringement on equality before the law, because it would have made citizens equal before separate
bodies of national law. Furthermore, liberalism is not only about equality. It is also about
liberty. At modest levels, group rights may not imply legal inequality between individuals. But
they must imply classification into peoples. And that raises the dilemma of classificatory
procedure, which Kymlicka and others today appreciate perhaps less than did Herrnritt.[75] Who is
who? How to classify individuals racially or nationally without violating their rights, their
liberty? How to establish what Ford has termed "classificatory due process," while being ever
mindful to keep asking "what 'process,' in this respect, we are all 'due'"?[76] It is no solution
simply to assume, as a Czech saying goes, that the wolf will eat its fill and the goat will remain
whole.

            Third and last, the Austrian compromises are relevant to Affirmative Action in America
today. Contrasts, to be sure, are stark. Affirmative Action rests on an "implicit normative theory
of proportional representation in the workforce,"[77] while the Moravian Compromise rested on an
explicit normative theory of separation in politics. Blackness in America has a historical link
with slavery, while neither Czechness nor Germanness had any particular link with a significant
legal status (although since the "ethnic cleansing" of the 1940s, they have been linked with
different citizenships). In America, the list of "official minorities" has expanded since the 1960s
to include a majority of the population. In the process, women, the disabled, and others have been
"racialized." In Austria, the number of peoples remained constant. An attempt by some Jewish
leaders at adding a Jewish people was denied by the Supreme Court in 1909.[78]

            And yet, America today and Austria back then have much in common: a liberal
constitutional order, multiple peoples, and considerable social and economic inequality among them.
Affirmative Action, like most of the Austrian experiment with "separate but equal," addresses such
inequality through group rights exercised by individuals. America and Austria thus share the
dilemma of classificatory procedure. India, with its liberal constitution and preferential policies
for "backward castes," shares it with them. Indian classificatory procedures today are strikingly
similar to those developed in Moravia after 1905.[79] Compelling comparisons have been made of
India's preferential policies with America's Affirmative Action, including their classificatory
approaches.[80] No one, however, has compared either America's or India's approaches with
Austria's—probably because Austria's is so little known in the English-speaking world.

            Such comparison might not prove a merely academic exercise. Affirmative Action, in
contrast with Jim Crow, has survived legal challenge on the grounds that it infringes on equality
before the law. Challenge to Affirmative Action's racial classification on the grounds that it
infringes on due process, on the other hand, seems not to have been attempted. That is probably
because the procedures for racial classification used by liberal America today are quite different
from those used by illiberal America not so long ago.[81] The procedures for Affirmative Action
have employed more self-classification and less verification, have been much more ad hoc, and have
been administered very inclusively, if not consistently. Yet those procedures, like the older ones,
have rested in large measure on folk definitions of race as natural and self-evident, often
manifested through "objective markers" such as skin color. And such definitions are now eroding,
under the impact of immigration and intermarriage, as well as scholarship on the social
construction of race and ethnicity. That, although perhaps good news, is helping to make
instrumental self-classification, or "identity fraud," a growing problem. Who is black and who is
white? Who is Hispanic, who is Asian, and who is Native American? What is the connection between
those categories and the disadvantage that Affirmative Action is supposed to counter? Like Ford,
other experts see building pressure for different and more formal classificatory procedures.[82]
Austria, other than India, seems the only liberal country which has experimented with formal
procedures for classifying citizens into peoples to any great degree. Its complex experiences might
prove instructive, and cautionary, to Americans as they begin to confront their own version of the
dilemma of classificatory procedure.
________________________________

[1] For criticisms that helped me improve this essay, I thank Peter Bugge, Daniel Gordon, Pieter
Judson, and Tara Zahra.

[2] Bosnia-Herzegovina should also be mentioned. Annexed from the Ottoman Empire in 1908, that
territory became a part not of Austria but of the larger entity of Austria-Hungary. Thus the legal
order imposed on it in 1910 was parallel, not subordinate, to the Austrian constitution. But that
order, like several Austrian crownland ones amended between 1905 and 1914, separated inhabitants
for elections to the Diet—into Catholics, Moslems, members of the Serbian Orthodox Church, and
Jews. See Julius Giegl, ed., Verfassungsgesetze der Länder Bosniens und Herzegovina (Vienna: Manz,
1910).

[3] There exists no study centered on the "little compromises" of 1905-1914 as a group. Gerald
Stourzh is the master of the topic, however, having explained them as a culmination of long, broad
trends in Die Gleichberechtigung der Nationalitäten in der Verfassung und Verwaltung Österreichs,
1848-1918 (Vienna: Verlag der österreichischen Akademie der Wissenschaften, 1985). He has also
examined individual cases and aspects in several essays, cited in the footnotes which follow.

[4] Paul Starr, "Social Categories and Claims in the Liberal State," in Social Research, vol. 59,
no. 2 (summer 1992), 289-295.

[5] Landesgesetz- und Verordnungsblatt für die Markgrafschaft Mähren (Brünn: Rohrer, 1906), 2-7
(Nr. 1, §3-15), 10-17 (Nr. 2, §1-15). For a summary of the Moravian Compromise and of the
historiography concerning it, see Jiří Malíř, "Der mährische Ausgleich als Vorbild für die Lösung
der Nationalitätenfragen?," in Thomas Winkelbauer, ed., Kontakte und Konflikte. Böhmen, Mähren und
Österreich: Aspekte eines Jahrtausends gemeinsamer Geschichte (Horn-Waidhofen an der Thaya:
Waldviertler Heimatbund, 1993), 337-46. Why it was decided to partition the Diet electorate, rather
than to introduce proportional voting, as well as why seats were allocated in a ratio of 73:46 (or
to be more precise, 73:40:6—see note 65), are questions for another essay. Suffice it to say here
that through partitioning, the Diet ratio could be set roughly halfway between the demographic
ratio of Czechs to Germans in Moravia (understood, on the basis of language data in the census, to
be about 72:28) and the per capita ratio of Czech to German tax payments (understood to be about
1:2). Czech leaders embraced the former, quantitative understanding of equality, while German
leaders embraced the latter, qualitative one. See Jeremy King, "Group Rights in Liberal Austria:
The Dilemma of Equality in Proportional Representation," in Lukáš Fasora, Jiří Hanuš, and Jiří
Malíř, eds., Moravské vyrovnání z roku 1905/Der mährische Ausgleich von 1905 (Brno: Matice
moravská, 2006), 27-42.

[6] Rudolf von Herrnritt, "Die mährischen Ausgleichsgesetze und das Nationalitätsrecht," in
Österreichische Rundschau, vol. 6 (1906), 171.

[7] Regarding the inconsistent terminology used in Austrian jurisprudence, see Josef Lukas,
"Territorialitäts- und Personalitätsprinzip im österreichischen Nationalitätenrecht," in Jahrbuch
des öffentlichen Rechts der Gegenwart, vol. 2 (1908), 398.

[8] Edmund Bernatzik, Die österreichischen Verfassungsgesetze mit Erläuterungen, second edition
(Vienna: Manz, 1911), 426-27.

[9] Public elementary schools [Volksschulen/obecné školy] are the focus in this essay, rather than
public secondary schools of the Bürgerschule/měšťanská škola, Realschule/reálná škola, or
Gymnasium/gymnasium types, because schoolchildren were divided nationally above all upon entering
first grade.

[10] Graf Anton Pace, ed., Ernst Mayrhofer's Handbuch für den politischen Verwaltungsdienst in den
im Reichsrathe vertretenen Königreichen und Ländern, fifth edition, vol. 4 (Vienna: Manz, 1898),
500-13, 565-67, 734-43; and Karl Lamp, "Volksschulen," in Ernst Mischler and Josef Ulbrich, eds.,
Österreichisches Staatswörterbuch. Handbuch des gesamten österreichischen öffentlichen Rechtes,
vol. 4 (Vienna: Holder, 1909), 829-43. See also Pieter Judson, Exclusive Revolutionaries. Liberal
Politics, Social Experience, and National Identity in the Austrian Empire, 1848-1914 (Ann Arbor:
University of Michigan Press, 1996), 51-58.

[11] Landesgesetz- und Verordnungsblatt für…Mähren (1906), 44-48 (Nr. 4, Abt. 1, §8, 20-37).

[12] Administrative Judicial Court, Budwiński 8377A (5 July 1911, Moravia), in August Ritter von
Popelka, ed., Budwińskis Sammlung der Erkenntnisse des k.k. Verwaltungsgerichtshofes,
Administrativrechtlicher Teil, vol. 35 (Vienna: Manz, 1912), 1260. See also Landesgesetz- und
Verordnungsblatt für…Mähren (1906), 51-53 (No. 4, §41-43)

[13] Alfred Freiherr von Skene, a member of the Diet who played an important role in the
long-lasting negotiations which preceded the Compromise, barely mentions questions of
classification in his account. Nor do the Diet protocols and committee reports from 1905. See
Skene, Der nationale Ausgleich in Mähren 1905 (Vienna: Konegen, 1910); and Landtagsblatt über die
Sitzungen des mit dem Allerhöchsten Patente vom 29. September 1905 einberufenen mährischen
Landtages (Nach stenographischen Aufzeichnungen) (Brünn: 1905).

[14] Rogers Brubaker, Ethnicity without Groups (Cambridge: Harvard University Press, 2004), 8.

[15] Landtagsblatt, Druckvorlage 222/212 (Brünn: 1905), 1-4; and Bernatzik, Verfassungsgesetze,
989-91. See also Max Menger, Der böhmische Ausgleich (Stuttgart: Cotta, 1891), 62-70; Karl
Gottfried Hugelmann, ed., Das Nationalitätenrecht des alten Österreich (Vienna: Braumüller, 1934),
234-35; and Horst Glassl, Der mährische Ausgleich (Munich: Fides, 1967), 217-24. A similar approach
was used for local school boards in the crownland of Tyrol from 1892. See Bernatzik,
Verfassungsgesetze, 1001.

[16] Cited in Stourzh, Gleichberechtigung, 205.

[17] Administrative Judicial Court, Budwiński 27A (12 January 1901, Bohemia), in Rudolf Alter, ed.,
Budwińskis Sammlung der Erkenntnisse des k.k. Verwaltungsgerichtshofes, Administrativrechtlicher
Teil, vol. 25 (Vienna: Manz, 1901), 48-49. For similar pronouncements by the Court, see Budwiński
14024 (5 April 1900, Bohemia), in Alter, Budwińskis Sammlung, vol. 24 (1901), 378-79; Budwiński
3634A (14 June 1905, Moravia), in Alter, Budwińskis Sammlung, vol. 29 (1905), 726; and Budwiński
4738A (6 November 1906, Bohemia), in Alter, Budwińskis Sammlung, vol. 30 (1906), 1148. See also
Lukas, "Personalitätsprinzip," 401; Stourzh, Gleichberechtigung, 205-8; and Wolfgang Steinacker,
Der Begriff der Volkszugehörigkeit und die Praxis der Volkszugehörigkeitsbestimmung im
altösterreichischen Nationalitätenrecht (Innsbruck: Wagner, 1932), 17-24; as well as Budwiński
5492A (14 November 1907, Bohemia), which is discussed in Section II.

[18] Administrative Judicial Court, Budwiński 9708 (3 June 1896, Moravia), in Adam Freiherr von
Budwiński, ed., Erkenntnisse des k.k. Verwaltungsgerichtshofes, vol. 20 (Vienna: Manz, 1896),
925-27; Budwiński 3634A, 722-34; Pace, Mayrhofer's Handbuch, 576-7, 672-73; Stourzh,
Gleichberechtigung, 169-74; Steinacker, Volkszugehörigkeit, 7-9, 28-29; Hannelore Burger,
Sprachenrecht und Sprachgerechtigkeit im österreichischen Unterrichtswesen 1867-1918 (Vienna:
Österreichische Akademie der Wissenschaften, 1995), 64, 96-102, 199; and Rudolf Herrmann von
Herrnritt, Nationalität und Recht (Vienna: Manz, 1899), 78-79.

[19] Bernatzik, Verfassungsgesetze, 1000, 426. Because Lex Perek was embedded in a section of the
Compromise concerning public elementary schools, private ones were not affected. Administrative
Judicial Court, Budwiński 8925A (4 May 1912, Moravia), in Popelka, Budwińskis Sammlung, vol. 36
(1912), 724-25.

[20] Stourzh, Gleichberechtigung, 221, 315.

[21] Stourzh, Gleichberechtigung, 216-17, 315; Landtagsblatt über die Sitzungen des…mährischen
Landtages (Nach stenographischen Aufzeichnungen) (Brünn: 1909), 3510-11, 3523-30 (d'Elvert,
Fischel); and Moravské listy, 31 March 1914, 1, 3.

[22] Landtagsblatt (1909), 3519 (Kadlčák). For an example of partisan statistics, see Hugo Herz,
Der nationale Besitzstand und die Bevölkerungsbewegung in Mähren und österreichisch Schlesien
(Brünn: 1909).

[23] Emil Brix, Die Umgangssprache in Altösterreich zwischen Agitation und Assimilation (Vienna:
Böhlau, 1982), 436-37, 447; and Jeremy King, Budweisers into Czechs and Germans: A Local History of
Bohemian Politics, 1848-1948 (Princeton: Princeton University Press, 2002), 58-60.

[24] Edmund Bernatzik, Die Ausgestaltung des Nationalgefühls im 19. Jahrhundert. Rechtsstaat und
Kulturstaat (Hannover: Helwing, 1912), 43; and Landtagsblatt (1909), 3520-21 (Kadlčák), 3533
(Stránský). See also Stourzh, Gleichberechtigung, 311-16; Steinacker, Volkszugehörigkeit, 55-57;
Toshiaki Kyogoku, "Národní agitace a obecní školství na Moravě na přelomu 19. a 20. století. Boj o
české dítě," in Harald Binder, Barbora Křivohlavá, and Luboš Velek, eds., Místo národních jazyků ve
výchově, školství a vědě v habsburské monarchii 1867-1918 (Praha: Výzkumné centrum pro dějiny vědy,
2003), 563-77; and Tara Zahra, "Reclaiming Children for the Nation: Germanization, National
Ascription, and Democracy in the Bohemian Lands, 1900-1945," in Central European History, vol. 37,
no 4 (2004), 499-541.

[25] Peter Bugge, "Contradictions in Terms? Czech and Austrian-German Liberalism in the Late
Habsburg Empire," article manuscript, 2005.

[26] Jeremy King, "The Nationalization of East Central Europe: Ethnicism, Ethnicity, and Beyond,"
in Nancy Wingfield and Maria Bucur, eds., Staging the Past: The Politics of Commemoration in
Habsburg Central Europe, 1848 to the Present (West Lafayette: Purdue University Press, 2001),
112-52.

[27] Budwiński 9708, 925.

[28] Hence the importance of recommendations by ministerial experts as to whether His Majesty
should sanction legislation; these executive-branch advisory opinions were the major line of
defense against unconstitutional laws. Bernatzik, Verfassungsgesetze, 482, 488-90, 900-901, 937;
Erwin Melichar, "Die Rechtslage der Nationalitäten in Zisleithanien nach der Dezember-Verfassung
1867 im Lichte der Judikatur des Reichsgerichtes (1869-1918)," in Ludovit Holotik, ed., Der
österreichisch-ungarische Ausgleich 1867 (Bratislava: Verlag der slowakischen Akademie der
Wissenschaften, 1971), 474; and Stourzh, Gleichberechtigung, 63-65.

[29] Rudolf Herrmann von Herrnritt, "Die Ausgestaltung des österreichischen Nationalitätenrechtes
durch den Ausgleich in Mähren und in der Bukowina," in Österreichische Zeitschrift für öffentliches
Recht, vol. 1 (1914), 609-10. See also the caustic comments of Herrnritt's colleague on the Court,
Friedrich Tezner, cited in Stourzh, Gleichberechtigung, 216, 315.

[30] In August 1910, state administrative authorities throughout Moravia were informed that "the
Ministry of Religion and Education has declared…, in the course of adjudicating an appeal, that the
reason adduced by the appellant—that he wishes to send his sons to the German school so that he
might eventually make better use of them in his shop, in which there is an active intercourse with
German tradesmen and customers—must be considered convincing." Z. 4966 Praes., Moravský zemský
archiv, B13 Carton 2039.

[31] Burger, Sprachenrecht, 192-98; and Stourzh, Gleichberechtigung, 310. For the text of the
Marchet decree, see Landesgesetz- und Verordnungsblatt für…Mähren, No. 11 (1907), 84; and Alfred
Fischel, Das österreichische Sprachenrecht. Eine Quellensammlung, second edition (Brünn: Irrgang,
1910), 314-16. Fischel omits the sentence defining "proficient."

[32] Landesgesetz- und Verordnungsblatt für…Mähren (1906), 23-24 (No. 2, §30-32), 38-40 (No. 2,
§66-72).

[33] Karl Renner ["Rudolf Springer"], Der Kampf der österreichischen Nationen um den Staat (Leipzig
and Vienna: Deuticke, 1902), 65.

[34] Landesgesetz- und Verordnungsblatt für…Mähren (1906), 40 (No. 2, §72).

[35] Robert Luft, "Die Mittelpartei des mährischen Grossgrundbesitzes 1879 bis 1918. Zur
Problematik des Ausgleichs in Mähren und Böhmen," in Ferdinand Seibt, ed., Die Chance der
Verständigung: Absichten und Ansätze zu übernationaler Zusammenarbeit in den böhmischen Ländern
1848‑1918 (Munich: Oldenbourg, 1987), 205; Glassl, Ausgleich, 142-48, 154-55, 168-69, 185-86;
Landesgesetz- und Verordnungsblatt für…Mähren (1906), 8-9 (Nr. 1, §38); Herrnritt, Nationalität,
78, 148; and Herrnritt, "Ausgestaltung," 600.

[36] Bernatzik, Verfassungsgesetze, 482; Erwin Melichar, "Die Freiheitsrechte der
Dezember-Verfassung 1867 und ihre Entwicklung in der reichsgerichtlichen Judikatur," in
Österreichische Zeitschrift für öffentliches Recht, vol. 16 (1966), 256-90; and Theodor Dantscher
von Kollesberg, "Politische Rechte," in Mischler and Ulbrich, Österreichisches Staatswörterbuch,
vol. 3 (1907), 927-32. I translate Reichsgericht as "Supreme Court," rather than "Imperial Court,"
in order to avoid confusion of the institution with the Emperor's household and retinue, or Hof.

[37] In 1880, in a case concerning whether Jews were Germans, the Supreme Court had made a ruling
which anticipated the declaration principle laid down by the Administrative Judicial Court the
following year. Supreme Court, Hye 219 (12 July 1880, Galicia), in Anton Hye, ed., Sammlung der
nach gepflogener mündlicher Verhandlung geschöpften Erkenntnisse des k.k. Reichsgerichtes, vol. 5
(Vienna: Manz, 1881), 1003. See Stourzh, Gleichberechtigung, 74-77.

[38] Bernatzik, Verfassungsgesetze, 762; Stourzh, Gleichberechtigung, 226-28; and Austrian State
Archive, Allgemeines Verwaltungsarchiv (AVA), k.k. Reichsgericht, Sitzungsprotokolle, 15 October
1907, 50 (Wacha).

[39] See in this regard the observation by Adolf Fischhof already in 1871 that an organ formed on
the basis of national separation "is the fortress within which the national minority may defend
itself successfully against the attacks of the national majority, as long as the garrison is an
unmixed national one, a reliable one. If there are joined to this national garrison alien or
doubtful elements, however, let alone hostile ones, then the fortress becomes a trap." Cited in
Gerald Stourzh, "Ethnic Attribution in Late Imperial Austria: Good Intentions, Evil Consequences,"
in Ritchie Robertson and Edward Timms, eds., The Habsburg Legacy: National Identity in Historical
Perspective (Edinburgh: Edinburgh University Press, 1994), 69.

[40] AVA, k.k. Reichsgericht, Sitzungsprotokolle, 15 and 16 October 1907, 47-66, 81-108 ("most holy
right of the individual," Piniński's words, may be found on page 105); and Supreme Court,
Hye-Hugelmann 1531 (17 October 1907), in Karl Hugelmann, ed., Sammlung der nach gepflogener
mündlicher Verhandlung geschöpften Erkenntnisse des k.k. Reichsgerichtes, vol. 14 (Vienna: k.k.
Hof- und Staatsdruckerei, 1912), 433-40.

[41] AVA, k.k. Reichsgericht, Sitzungsprotokolle, 15 and 16 October 1907, 50 (Wacha), 55-57
(Pluhař), 65-66 (excerpts from Budwiński 27A and 14024), 82-83, 87 (Czyhlarz).

[42] Starr, "Social Categories," 290.

[43] Edmund Bernatzik, Über nationale Matriken (Vienna: Manz, 1910), 29-30.

[44] Budwiński 5492A (14 November 1907, Bohemia), in Alter, Budwińskis Sammlung, vol. 31 (1907),
1064-66.

[45] Administrative Judicial Court, Budwiński 7846A (30 December 1910, Moravia), in Popelka,
Budwińskis Sammlung, vol. 34 (1911), 1752.

[46] Budwiński 7846A, 1748-50.

[47] Budwiński 7846A, 1752-53.

[48] Steinacker, Volkszugehörigkeit, 40-41, 48. See also Heinz Kloss, Grundfragen der Ethnopolitik
im 20. Jahrhundert. Die Sprachgemeinschaften zwischen Recht und Gewalt (Vienna: Braumüller, 1969),
231; and Ariela Gross, "Litigating Whiteness: Trials of Racial Determination in the
Nineteenth-Century South," in Yale Law Journal, vol. 108 (October 1998), 109-88. Regarding
"objective" markers, see Christopher Ford, "Administering Identity: The Determination of 'Race' in
Race-Conscious Law," in California Law Review, vol. 82 (1994), 1237-38.

[49] Administrative Judicial Court, Budwiński 7843A (11 December 1910, published 30 December 1910,
Moravia), in Popelka, Budwińskis Sammlung, vol. 34 (1911), 1741. See also Administrative Judicial
Court, Budwiński 7844A (Moravia), issued on the same day. In the case at hand, stated this
decision, it was demonstrated not that 33 schoolchildren were proficient in the language of
instruction, "but rather that they command the German language of instruction 'well enough to
follow instruction in it' (§2, final sentence, Ministerial Decree of 14 May 1907, Crownland Legal
Gazette Nr. 52). This, however, in the view of the Judicial Court, does not suffice for the
admission of the children to the German elementary school." Popelka, Budwińskis Sammlung, vol. 34
(1911), 1747.

[50] Budwiński 7843A, 1734-35; Herrnritt, "Ausgestaltung," 601; and Stourzh, Gleichberechtigung,
220.

[51] Budwiński 7843A, 1740-42.

[52] Tara Zahra, Your Child Belongs to the Nation: Nationalization, Germanization, and Democracy in
the Bohemian Lands, 1900-1945 (Ann Arbor: University of Michigan doctoral dissertation, 2005),
125-29; and Stourzh, "Ethnic Attribution," 75.

[53] Lukas, "Personalitätsprinzip," 400.

[54] Herrnritt, Nationalität, 79-80; and Herrnritt, "Ausgestaltung," 600-602. See also Steinacker,
Volkszugehörigkeit, 32-33; and Christopher Ford, "Challenges and Dilemmas of Racial and Ethnic
Identity in American and Post-Apartheid South African Affirmative Action," in UCLA Law Review, vol.
43 (August 1996), 1954, 2006.

[55] Lukas, "Personalitätsprinzip," 400.

[56] Herrnritt, Nationalität, 77-82, 138; Herrnritt, "Ausgleichsgesetze," 171-73; and Herrnritt,
"Ausgestaltung," 600-2, 615. Bernatzik made similar proposals in 1910 and 1912: see Matriken,
31-32; and Ausgestaltung, 31-33.

[57] Bernatzik, Verfassungsgesetze, 464-68; Landtagsblatt über die Sitzungen des mährischen
Landtages in der Zeit vom 16. bis 18. Juli 1913 und vom 3. bis 28. Februar 1914 (Brünn: 1914),
appendix 283; and Moravské listy, 31 March 1914, 1.

[58] To my knowledge, no law in Austria established any estoppel for self-classification. The
compromise for the Bohemian town of Budweis/Budějovice, however, which was on the verge of being
finalized when war broke out in 1914, would have established extensive estoppel for national
self-classification had it been enacted. It would also have scaled back bureaucratic verification
to some degree. See King, Budweisers, 141-42; and Emil Brix, "Der böhmische Ausgleich in Budweis,"
in Österreichische Osthefte, 24, no. 2 (1982), 236. On the other hand, the Administrative Judicial
Court did establish some estoppel for other-classification. In 1913, for example, it prohibited the
admission of children to a school if they had been denied admission to a school of the same
language, on grounds of insufficient proficiency, the previous year. In 1906, Herrnritt had hoped
for more: "One question which will need to be answered is whether a legally binding decision
regarding registration in the electoral rolls of one of the national groups should have estoppel
value [präjudiziell sein] for other administrative questions of nationalities law—particularly in
connection with the assertion of demands for the erection of national schools and for national
representation in the school supervisory authorities." Herrnritt, "Ausgestaltung," 611; and
Herrnritt, "Ausgleichsgesetze," 173. The answer given by the Court, however, was "no." The final
sentence of Budwiński 7846A, in December 1910, blocked the use of national electoral rolls to
settle disputes over the national belonging of school board members. "Rather, in the constituting
of school authorities…, which nation a person belongs to is to be determined quite independently."

[59] Szema Jona Wyszewianski, Über die formalrechtliche Behandlung der Nationalitäten in der
modernen Gesetzgebung (Mannheim: Hensle, 1909), 16-23.

[60] On group rights and who exercises them, see Karl Renner ["Synopticus"], Staat und Nation.
Staatsrechtliche Untersuchung über die möglichen Principien einer Lösung und die juristischen
Voraussetzungen eines Nationalitätengesetzes (Vienna: Dietl, 1899), 3-11; Stourzh,
Gleichberechtigung, 189-96; and Peter Schuck, Diversity in America (Cambridge: Belknap, 2003),
36-39.

[61] Karl Renner ["Rudolf Springer"], Der Kampf der österreichischen Nationen um den Staat (Leipzig
and Vienna: Deuticke, 1902), 37, 53, 87-90; and Otto Bauer, The Question of Nationalities and
Social Democracy, edited by Ephraim Nimni and translated by Joseph O'Donnell (Minneapolis:
University of Minnesota Press, 2000), 222, 281-87. See also Brubaker, Ethnicity without Groups,
especially Chapter 1.

[62] Renner, Kampf, 65-68. See also Karl Renner, Das Selbstbestimmungsrecht der Nationen in
besonderer Anwendung auf Österreich (Vienna: Deuticke, 1918), 113-14; Bauer, Question of
Nationalities, 281-83; and Lukas, "Personalitätsprinzip," 400.

[63] James Madison, "The Apportionment of Members among the States," in Clinton Rossiter, ed., The
Federalist Papers (New York: Signet, 2003), 338; Paul Starr, "The Sociology of Official
Statistics," in William Alonso and Paul Starr, eds., The Politics of Numbers (New York: Russell
Sage Foundation, 1987), 18, 33; Bernatzik, Ausgestaltung, 31, 40; Renner, Selbstbestimmungsrecht,
69; and Karl Renner, Grundlagen und Entwicklungsziele der österreichisch-ungarischen Monarchie
(Vienna: Deuticke, 1906), 247-48.

[64] Lukas, "Personalitätsprinzip," 359-62; Renner, Staat und Nation, 30-31; Renner, Kampf, 69; and
Bauer, Question of Nationalities, 283-291. The compromise for Budweis/Budějovice would have divided
school taxes nationally, and prevented inequalities by providing Czech schools in the town with a
subsidy from the Bohemian Diet—i.e., from other towns. That solution could not have worked for an
entire crownland. King, Budweisers, 144.

[65] Landesgesetz- und Verordnungsblatt für…Mähren (1906), 3 (Nr. 1, §10a). The six members of the
Diet who were elected by Moravia's Chambers of Commerce and Trade could choose between the national
curias. After the elections of 1906 and 1913, all six, as expected, chose the German one. Any
member of the Diet could choose to join no curia, at the high price of exclusion from the Diet's
many staffing decisions.

[66] A variant of this procedure reversed the roles. A binational and/or nonnational agent
nominated, and a national jury appointed. Here, however, the Compromise required that each person
appointed belong to the relevant nation. Administrative authorities, not national juries, did the
real work of verification. Landesgesetz- und Verordnungsblatt für…Mähren (1906), 8 (Nr. 1, §32a),
44-48 (Nr. 4, Abt. 1, §20, 22, 35, 36, 37).

[67] Guido Calabresi and Philip Bobbitt, Tragic Choices (New York: Norton, 1978).

[68] Ford, "Administering Identity," 1283-84, citing Calabresi and Bobbitt, Tragic Choices, 57, 71.

[69] Ford, "Administering Identity," 1238, 1267. See also 1272-73.

[70] Herrnritt, "Ausgestaltung," 615. See also Lukas, "Personalitätsprinzip," 405.

[71] Ford, "Administering Identity," 1285.

[72] Brubaker, Ethnicity without Groups, 17.

[73] Will Kymlicka, Multicultural Citizenship: A Liberal Theory of Minority Rights (Oxford:
Clarendon, 1995), 37, 58-60.

[74] The Czech electoral districts created through the Moravian Compromise contained more voters
than did the German ones. A Czech and a German whose votes had carried equal weight before 1905
were thus rendered unequal by law. See Landtagsblatt (1905), 309-10 (Svozil), 536 (Fux);
Wyszewianski, Behandlung, 74; and Suzanne Konirsch, "Constitutional Aspects of the Struggle between
Germans and Czechs in the Austro-Hungarian Monarchy," in Journal of Modern History, 27, no. 3
(September 1955), 232. A similar inequality holds in the United States to this day, in elections to
the Senate—and was the norm in state legislatures until Reynolds v. Sims (1964), 377 U.S. 533. See
Harlan dissent, at 589.

[75] See, for example, Iris Marion Young, Justice and the Politics of Difference (Princeton:
Princeton University Press, 1990); as well as Brubaker's critique of Young, in Ethnicity without
Groups, 58-60. Also relevant is Zahra's critique of Kymlicka, in "Your Child," 34. For an approach
to Affirmative Action which perhaps errs in the opposite direction, by rejecting racial
classification altogether, see Andrew Kull, The Color-Blind Constitution (Cambridge: Harvard
University Press, 1992).

[76] Ford, "Administering Identity," 1282-1285.

[77] Hugh Davis Graham, The Civil Rights Era: Origins and Development of National Policy (New York:
Oxford University Press, 1990), 120, 321, 383.

[78] See John Skrentny, The Minority Rights Revolution (Cambridge: Belknap Press, 2002); and Gerald
Stourzh, "Galten die Juden als Nationalität Altösterreichs?," in Anna Maria Drabek, ed., Prag,
Czernowitz, Jerusalem: der österreichische Staat und die Juden vom Zeitalter des Absolutismus bis
zum Ende der Monarchie (Eisenstadt: Edition Roetzer, 1984), 73-117.

[79] Regarding preferential policies and caste classification in India, see Marc Galanter,
Competing Equalities. Law and the Backward Classes in India (Berkeley: University of California
Press, 1984); Ford, "Administering Identity;" and Laura Dudley Jenkins, Identity and Identification
in India: Defining the Disadvantaged (New York: Routledge Curzon, 2003).

[80] See the works cited in the preceding note, as well as Clark Cunningham, Glenn Loury, and John
Skrentny, "Passing Strict Scrutiny: Using Social Science to Design Affirmative Action Programs," in
Georgetown Law Journal, vol. 90 (2002), 836-82; Clark Cunningham, "After Grutter Things Get
Interesting! The American Debate over Affirmative Action is Finally Ready for Some Fresh Ideas from
Abroad," in Connecticut Law Review, vol. 36, no. 3 (Spring 2004), 665-77; and Laura Dudley Jenkins,
"Race, Caste and Justice: Social Science Categories and Antidiscrimination Policies in India and
the United States," in Connecticut Law Review, vol. 36, no. 3 (Spring 2004), 747-85.

[81] Recent scholarship on pre-civil rights era racial classification includes Gross, "Litigating
Whiteness"; and Daniel Sharfstein, "The Secret History of Race in the United States," in Yale Law
Journal, vol. 112 (April 2003), 1473-1509.

[82] Lawrence Wright, "One Drop of Blood," in The New Yorker, 25 July 1994, 46-55; Cunningham et
al, "Passing Strict Scrutiny"; Skrentny, Minority Rights Revolution, especially v-vi, 355; and Hugh
Davis Graham, Collision Course: The Strange Convergence of Affirmative Action and Immigration
Policy in America (New York: Oxford University Press, 2002). For an alarming perspective, see
Luther Wright, "Who's Black, Who's White, and Who Cares: Reconceptualizing the United States's
Definition of Race and Racial Classifications," in Vanderbilt Law Review, vol. 48 (March 1995),
515-69.