Tilburg University
Counterfactuals and the invention of religious traditions
Sarot, Marcel
Published in:
Religious identity and the invention of tradition
Publication date:
2001
Link to publication
Citation for published version (APA):
Sarot, M. (2001). Counterfactuals and the invention of religious traditions. In J. W. van Henten, & A. W. J.
Houtepen (Eds.), Religious identity and the invention of tradition: Papers read at a NOSTER Conference in
Soesterberg, January 4-6, 1999 (pp. 21-40). (Studies in theology and religion; No. 3). Assen: Koninklijke Van
Gorcum BV.
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners
and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
- Users may download and print one copy of any publication from the public portal for the purpose of private study or research
- You may not further distribute the material or use it for any profit-making activity or commercial gain
- You may freely distribute the URL identifying the publication in the public portal
Take down policy
If you believe that this document breaches copyright, please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
Download date: 18. jun. 2017
Counterfactuals and the Invention of Religious Traditions
Marcel Sarot
1
Introduction
It is often assumed that the concept of ‘invention of tradition’ originates with Eric
Hobsbawm.1 True as it may be that he is the one who gave rise to its present popularity, this
concept certainly has more ancient roots. To give one example, Karl Popper used it in a
lecture that he gave in 1948 at the Third Annual Conference of the Rationalist Press
Association at Magdalen College, Oxford.2 In this lecture, Popper addressed several of the
questions that have come into prominence as a result of Hobsbawm and Ranger’s edited
volume The Invention of Tradition. That is interesting, since from the literature on the
invention of traditions one often gets the impression that the concept in itself is a crystal-clear,
ready-made tool to be employed in dealing with various historical phenomena. It seems to be
assumed that not very much can be gained from looking into other – not to mention older –
literature. The existing literature applies the concept3 rather than that it analyses it.4
In this paper, I will concentrate on analysis rather than on application. I will contrast
Hobsbawm’s views with those of Popper. Such a comparison will enable me to tease out the
finer shades of meaning of ‘invention of tradition,’ and to introduce some of the questions and
problems connected with it. After that, I will inquire into the value of this concept for the
study of religious traditions. There, I will argue that the value of the concept is limited to the
historical study of religious traditions; for the systematic disciplines within theology, it is only
indirectly relevant.
2
Popper and Hobsbawm on the Invention of Tradition
What do Popper and Hobsbawm mean by ‘tradition’? As for Popper, in his talk he does not
give a proper definition of tradition. He does make clear what he is speaking about, however,
by providing examples. One of these is that of a musical tradition:
When I was in New Zealand I got hold of a set of American records of Mozart’s
‘Requiem.’ When I had played these records I knew what the lack of a musical
tradition meant. This set of records had been made in America, under the directorship
of a musician who obviously was untouched by the tradition, which has come down
from Mozart. The result was devastating.5
Besides musical traditions, Popper mentions scientific traditions and research traditions.6
About one of these, the rationalist tradition, he remarks:
A part of the rationalist tradition is ... the metaphysical idea of determinism. People
who do not agree with determinism are usually viewed with suspicion by the
rationalists. They are afraid that somehow there must be the Holy Ghost coming round
the corner – that is to say, the Free Will and the Divine Grace. ... Another element in
the rationalist tradition which we should question is the idea of observationalism – that
is to say, the idea that we know about the world because we look round, open our eyes
and open our ears, take down what we see, hear, and so on; and that this is what
constitutes the contents of our knowledge.7
So much for Popper. Let’s see whether this is more or less what Hobsbawm has in mind as
well. Hobsbawm does give a proper definition of ‘tradition’:
1
a set of practices, normally governed by overtly or tacitly accepted rules and of a ritual
or symbolic nature, which seek to inculcate certain values and norms of behaviour by
repetition, which automatically implies continuity with the past.8
Further on, Hobsbawm clarifies his definition of ‘tradition’ further by distinguishing between
traditions and customs.
The ... characteristic of ‘traditions’ ... is invariance. The past ... to which they refer
imposes fixed (normally formalized) practices, such as repetition. ‘Custom’ ... does not
preclude innovation and change up to a point...9
For Hobsbawm, then, traditions are sets of practices; this seems to match Popper’s examples
of musical and research traditions. These sets of practices are rule-governed; here again,
Popper seems to agree; the conductor he mentions, failed to know certain rules about playing
Mozart, and scientists have to know certain rules about how to acquire knowledge. Whereas
Hobsbawm seems to restrict traditions to the field of rituals and symbols, however, Popper
also allows for traditions in other fields, such as traditions of thought. Here, I think, we have
every reason to follow Popper: religious traditions, for example, go beyond symbols and
rituals, and also include thought, morality, etc. In another respect, Popper also broadens
Hobsbawm’s definition, namely by mentioning metaphysical beliefs. These are beliefs of a
factual nature (e.g., that all our acts are determined, and that as a consequence, freedom is an
illusion), though they do not admit of empirical falsification.10 It seems that Popper has a
point here, and that factual beliefs normally form part of a tradition. I say ‘factual beliefs’
rather than ‘metaphysical beliefs’ because it seems to me that non-metaphysical factual beliefs
can also form part of a tradition (e.g., ‘This Requiem was written by Mozart’ and ‘Jesus lived
on earth,’ which in principle both admit of empirical verification or falsification). Moreover, I
assert that they normally form part of a tradition because rules always presuppose a factual
context.11 Thus, rules for the performance of Mozart’s music suppose the existence of
Mozart’s compositions, the existence of certain music instruments, etc. etc. These are factual
suppositions that are part of this musical tradition.
What about the invariability of traditions? Popper’s examples suffice to show that
absolute invariability – which seems to be required by Hobsbawm – is not required. As the
establishment of the Netherlands School for Theology and Religious Studies NOSTER shows,
research traditions may change, and as anyone who compares a 50-year-old performance of
Mozart’s Requiem with a contemporary recording will notice, musical traditions can change
as well. This is not to say that Hobsbawm is entirely mistaken in requiring invariability:
whenever a complete change of a set of practices takes place all at once, we will no longer be
inclined to speak of the same tradition. A tradition seems to require at least a certain degree of
continuity over time; and we need not determine here, how much continuity is required to
justify speaking about the same tradition.12
What is the function13 of a tradition? As we have seen, Hobsbawm suggests that
traditions serve “to inculcate certain values and norms of behaviour by repetition.’ He
explains that these “values and norms” may include (1) “social cohesion or the membership of
groups, real and artificial communities”, (2) “institutions, status or relations of authority”, and
(3) “beliefs, value systems and conventions of behaviour”.14 Popper also discusses the
function of tradition, and attaches much value to this element of his argument: “(w)hat is the
function of traditions in social life? … Can we analyse the functions of traditions? That is
perhaps the main task of a theory of tradition”.15 Popper specifies four functions of traditions.
2
First, traditions help us to perceive order in an environment that might otherwise appear
chaotic to us:
(w)henever we happen to be surrounded by either a natural environment or a social
environment of which we do not know anything, and where we consequently cannot
predict what will happen, then we all become very anxious and very terrified. We are
terrified and anxious, because if there is no possibility for us to predict what will
happen in our environment – for example, how people will behave – then there is no
possibility for us to act rationally.16
Here, the importance of the factual assumptions of traditions, to which Popper drew our
attention earlier on, becomes apparent. By means of these factual assumptions, traditions help
us to interpret the (natural and social) environment in which we live, “to bring some order in
the chaos in which we live”,17 so that we can act rationally within it.
A second function of traditions is, that they give us something “by which to go on” and
tell us “how to proceed”.18 Thus, the function of traditions is not merely that they specify the
factual context within which we must act. As Popper’s example of the Mozart recordings
shows, traditions typically include guidelines for acting or conventions of behaviour as well.
Our actions may include the revision and adaptation of traditions; like scientific
theories, traditions may need correction. “(T)raditions have the important double function that
they not only create a certain order or something like a social structure, but that they also give
us something upon which we can operate: something that we can criticize and change”.19
Thus, the third function of traditions is that they provide the stable background that enables us
to make meaningful changes. In this connection, Popper argues against “canvas-cleaning”: the
sweeping away of existing traditions in order to create new ones. He explains the problem
with canvas-cleaning by again referring to the analogy with science: “In science it would be a
tremendous loss if we were to say: “We do not make very much progress. Let us sweep away
all science and start afresh.” The rational procedure is to correct it and to revolutionize it, but
not to sweep it away. You may create a new theory, but the new theory is created in order to
solve those problems which the old theory did not solve”. Similarly, “traditions … have no
meaning except on the basis of traditions and institutions – such as myths, poetry and values –
which all emerge from the social world in which we live”.20 Innovations are meaningful only
against the background of an existing tradition. The way in which the Dutch painter Vincent
van Gogh reconceived the art of painting, for example, can only be understood against the
background of the work of his predecessors and contemporaries. And what’s more, it would
have been impossible without this background. Thus, it is the fact that a tradition already
exists which makes innovation possible.
A fourth function of tradition is, that by complying with it, people “reassure others of
their rationality and predictability, perhaps in the hope of making them act in a reciprocal
way”.21 Thus, by conforming to a tradition we make a claim about ourselves – i.e., about our
own identities: we are the sort of people the behaviour of whom you can predict and on whom
you can rely. Here, a function of tradition mentioned by Hobsbawm is relevant: since
traditions typically are shared by groups, by conforming to a tradition we become members of
a group. This helps to reinforce the identity-establishing effect of following a tradition. The
identity we claim for ourselves is not always identical with the identity other people ascribe to
us. If I claim to be Napoleon, for example, it may well be that most people do not believe me.
But if I claim for myself an identity by conforming to a tradition to which other people
conform as well, I share the identity which I claim for myself with these other people. This
increases the likelihood that other people – both those who participate in the tradition and
3
outsiders – acknowledge my claimed identity.22 Sometimes, by identifying ourselves with a
tradition, we implicitly express the wish that our neighbours conform to this tradition as well.
Socialists, for example, typically want other people to be socialists as well, just as many
Christians would like to convert everyone else to Christianity.
It is clear, then, that participating in a tradition often implies membership of a specific
group; but what exactly is the relationship to the institutions Hobsbawm mentions separately?
For the present purpose, we need not go deeply into this matter. Groups of persons become
institutions when they are characterised by formal patterns of behaviour.23 In practice, this
means that social groups the membership of which involves conforming to traditions, are
institutions.24 Thus, in the present context there is no need to distinguish sharply between
groups and institutions, and we need not pay separate attention to institutions.
Summarising, we come to the following definition of tradition:
A tradition is a set of practices, normally governed by overtly or tacitly accepted rules,
which presuppose a factual context. Traditions may help us (1) to discern order in a
prima facie chaotic world, (2) to know how to act within this world, (3) to participate
in groups and communities, (4) to claim an identity for ourselves, (5) to let other
people know what we expect of them, and (6) to change the contexts within which we
live. These rules seek to inculcate certain values and norms of behaviour by repetition,
which automatically implies continuity with the past. Traditions need not be
invariable, but they do need to exhibit a certain degree of continuity over time.
The other concept we should clarify is that of invention. The verb ‘to invent’ is what is in
philosophy sometimes called an incomplete symbol,25 that is to say that it can never be used by
itself, but always needs an object. One can never invent in general; one always invents
something. The exact meaning of ‘to invent’ is determined by its object. In ordinary English,
‘to invent’ can be combined with two different kinds of objects. On the one hand, one can
invent a machine, a process, a game, etc, namely by being the first to think of it and to create
it. On the other hand, when one invents a story, an excuse, etc., one tries “to make other
people believe that it is true when in fact it is not, usually for a bad or dishonest reason”.26
Thus, when to invent has series of propositions (such as stories and excuses) as its object, it
means to make up a factual untruth. When one invents an excuse, at least one of the
propositions constituting the excuse is false, and one knows that it is false. One has fabricated
it on purpose. On the other hand, when one invents a non-propositional item, such as a
machine, a process or a game, one creates something new and real. It is in this sense that we
speak about ‘inventors’, and we usually do so with much respect. Contrary to series of
propositions, non-propositional items can be invented by accident. This is what we call
‘serendipity’: finding one thing while engaged in the search for something else.
As we have seen, traditions are practices. Practices do not consist of sets of
propositions, some of which can be made up. Thus, one would expect, when tradition is the
object of invention, invention means creation. Is this indeed what Hobsbawm means? There is
some support in Hobsbawm for this interpretation. Thus, Hobsbawm speaks of “creating …
ritual and symbolic complexes” and “constructing” a tradition.27 Does Hobsbawm mean
conscious creation? No, not necessarily. He distinguishes between traditions that are “actually
invented”, i.e. “deliberately invented and constructed by a single initiator” or “officially
instituted and planned” (probably by a group) and traditions that “are partly invented, partly
evolved”.28 Sometimes, he seems to suggest that the distinguishing characteristic of an
invented tradition is not that conscious planning plays a role, but that they emerge “within a
brief and dateable period”.29
4
On the evidence I have just given, Hobsbawm’s interpretation of invention is quite
clear. However, this is only one half of the evidence. The other half suggests, that at least
sometimes Hobsbawm uses the phrase ‘invention of tradition’ in the other sense which it can
also have in ordinary language: the making up of an untruth. He asserts that invented
traditions “normally attempt to establish continuity with a suitable historic past”.30 Moreover,
he suggests that “insofar as there is such reference to a historic past, the peculiarity of
“invented” traditions is that the continuity with it is largely fictitious”.31 Furthermore, he
asserts that “(i)t is clear that plenty of political institutions, ideological movements and groups
– not least in nationalism – were so unprecedented that even historic continuity had to be
invented, for example by creating an ancient past”.32 In statements like these, Hobsbawm
distinguishes between a real and an invented past. Though here also he uses the phrase
‘invention of tradition’, he concentrates on the invention of continuity with the past where that
in fact is absent, as part of and legitimation of the construction of a new tradition. This form
of ‘invention of tradition’ also figures prominently in other articles in Hobsbawm & Ranger’s
volume, such as Hugh Trevor-Roper’s essay on the Highland tradition of Scotland. There,
Roper shows the Scottish kilt to have been invented by an Englishman in the early 18th
century; the wearing of the kilt, however, was promoted by Scottish nationalists who falsely
claimed it to be the dress of “our Celtic ancestors”.33 After the publication of The Invention of
Tradition, Hobsbawm himself continued to use ‘invention’ in this way. In an address of more
recent date entitled “History is what priest and educators make of it”, Hobsbawm talks about
the “invention of history”.34 Paul Post asserts that he employs the term “very explicitly in the
sense of myth-forming and manipulation, in short, misuse. In this context of the rose of
intolerant ideologies and their misuse of history, … Hobsbawm calls for a separation between
fact and fiction, and makes a passionate plea for intellectuals to resist against the construction
of national, ethnic and other myths”.35
Thus, just like ‘invention’ in ordinary language is used in two ways, Hobsbawm uses
‘invention of tradition’ in two ways. Sometimes he focuses on the construction of the tradition
itself, and at other times he concentrates rather on the fabrication of a fictitious past that
functions as a factual assumption in a new tradition. This fundamental ambiguity in
Hobsbawm’s own presentation of the term is mirrored in the ways in which others use it, and
has lead to much confusion. On the one hand, there are those who concentrate on the
fabrication of a fictitious past. At their hand, “‘Invention of tradition’ … receives a rather
pejorative context, and … is used to call up both the idea of artificially developed myths and
the falsification of history, as well as that of manipulation”. Thus, ‘invention of tradition’ is
used in “unmasking or demythologizing” contexts.36 On the other hand there are those who,
like Paul Post, invoke the other strand in Hobsbawm’s writing on the subject and reject the
demythologising use of ‘invention of tradition’: “Hobsbawm is investigating the cultural
construction of history and is not occupied with the unmasking of invented pasts or with
laying bare manipulations, but with research into the construction, creation or innovation of
traditions”.37 As I have argued above, in Hobsbawm we can find support for both uses of
‘invention of tradition.’ Ordinary language is ambiguous, and Hobsbawm does not remove
this ambiguity. Those who do make attempts to remove it, however, are often led by
pragmatic considerations. When, for example, Paul Post wants to concentrate on the
construction of ritual, that is exactly what one would expect from a liturgist; for a liturgist, the
manipulation of history is a less fruitful subject than the construction of ritual. However, there
are some forceful objections to this way of narrowing down the concept. Must we always
remove all ambiguities from our concepts? In some cases, is it not sufficient for us to be
aware of these ambiguities? Is it not the fact that the phrase seems to carry an implicit
suggestion of a fictitious past – a mere suggestion, and not an implicit, but undeniable
5
implication – which has made the idea that tradition can be invented so fruitful? And should
not the context within which Eric Hobsbawm introduces this idea – a collection of articles,
several of which discuss the role of untrue propositions about the past – also be taken into
serious consideration? In the case of ‘invention of tradition’, is it not possible that precisely
these ambiguities have made the phrase so popular?
I suggest, therefore, that we should retain this fruitful ambiguity, and continue to use
‘invention of tradition’ in the double meaning of both the creation of tradition and the
fabrication of a fictitious past.
3 Counterfactuals and Religious Tradition
In order to remove the depreciative connotation that may resonate in terms like ‘untrue
proposition’ and ‘falsehood’, let’s from now on talk about counterfactuals. I will use
‘counterfactual’ as shorthand for counterfactual propositions and statements.38 By using the
term ‘counterfactual’, I will also do justice to the fact that one can assert such propositions
and statements without knowing that they are false. With a proposition we assert that a certain
state of affairs exists (existed, will exist) in reality; propositions have an indicative or
constative function.39 They may be true or false. When they are false, they go against the facts.
Counterfactual propositions about the past are propositions about the past that do not refer to
any state of affairs that has in fact existed in the past.
In two respects, the idea that counterfactuals about the past may play an important role
in the construction of new traditions, seems an eye-opener to me. From a religious point of
view, it must be said that the historical reports about the origins of most religions are rather
unreliable. To begin with a few examples from my own religious tradition, in the Roman
Catholic Church it is claimed that Jesus Christ founded the Church, established its
hierarchical structure, instituted the priesthood and willingly excluded women from it, etc. I
don’t think that I am too much of a rebel when I claim that there are some historians who have
some doubts about these alleged facts. Despite these doubts, which we nowadays encounter
both within and without the Roman Catholic Church, the practices and traditions that are
legitimated by means of them, still continue to exist. Other world religions have similar
problems: their origins also lie in the remote past, and even where there are no positive
reasons to doubt certain reports about their origins, there often is very little evidence to
substantiate them. Viewed in this light, we have some prima facie reason to think that
Hobsbawm’s thesis that counterfactuals may play an important role in the establishment (and,
later on, the continuation) of new traditions, may be fruitfully applied to religious traditions.
The parallels with similar mechanisms in other contexts may throw some light on the way
they work in religion.
There is a second respect in which Hobsbawm’s thesis about the role of
counterfactuals may be an eye-opener. The analytical tradition in philosophy, in which I have
been trained, is often very rigid in its rejection of classes of utterances that suffer from what
we might call cognitive handicaps. To give an example, it is commonly claimed that when we
violate the law of non-contradiction, we do not succeed in asserting anything.40 Likewise, it is
claimed that we cannot live by incoherent views of life.41 Furthermore, it is also sometimes
claimed that the success condition of a proposition is whether it is true or false; false
propositions fail as a result of their counterfactuality.42 What Hobsbawm suggests, however, is
that propositions may be successful in other ways than by being true. Even false propositions
(counterfactuals) may be successful in playing an important role in the establishment of new
traditions. In a sense this is something we all know. If people were not aware of the fact that
falsehoods may sometimes be successful in other respects than being true, they would never
lie. In the case of successful lies, however, it is quite clear what determines their success: the
6
fact that they are believed. A lie succeeds if and only if those to which it is addressed believe
it. In the case of counterfactuals playing a role in the establishment of religious traditions, it is
much less clear what determines their success. It is not sufficient for a proposition to be
believed to make it fit to play a role in the establishment of religious traditions. What, then,
are its success conditions? This seems to be an especially important question in connection
with religious traditions, because there – as we have seen – many propositions the truth of
which cannot be ascertained play an important role.
One success condition has been mentioned already: counterfactuals can have a
tradition-generating function only when they are believed to be true (and not counterfactual).
Thus, though the truth of the counterfactual cannot be ascertained, its untruth should not be
established either. As soon as a statement has been shown to be false, it can no longer be
invoked to legitimate or justify (elements in) a tradition. In exceptional cases, the untruth may
be both firmly established and believed by a majority of those who are familiar with the
counterfactual in question. Even then, however, those who are part of the religious tradition in
which that counterfactual plays a legitimating role, should believe in its truth. This will often
be so in the case of religious sects. To give only one example, it has been a well-established
fact for a long time that there are many analogies between the Book of Mormon (1830) and
Ethan Smith’s View of the Hebrews (1st ed. 1823, 2nd ed. 1825), a book Joseph Smith
probably had read before publishing his ‘translation’ of the Book of Mormon. For those who
are not members of the ‘Church of Jesus Christ of Latter Day Saints’, the detailed similarities
between both books throw serious doubt on its alleged divine origin. The Mormons
themselves, however, cannot accept that the Book of Mormon was really composed by Joseph
Smith himself – on the price of giving up their faith. This seems a clear example of a
counterfactual that has been unmasked as being untrue but can still continue its traditiongenerating function, because the unmasking is disbelieved – and hardly known – within the
religious tradition in question.43 By way of parallel secular example, we may again mention
Trevor-Roper’s findings on the Scottish kilt; it is only because it was – contrary to the
available evidence – widely believed to be of medieval Celtic origin, that they could become a
symbol of the Scottish national identity.
What more success conditions can we mention for tradition-generating
counterfactuals, besides being credible for those participating in the tradition? A reasonable
guess would be that counterfactuals are tradition-generating by virtue of the fact that they
contribute to the function of that tradition. As we have seen, it is one of the functions of
traditions that by participating in traditions, we can become members of specific groups and
communities. If a specific counterfactual shows us a way to become such a group member,
that would certainly add to the tradition-generating potential of that counterfactual. Here
again, the specific dress of the Scottish Highland can function as an example: Trevor-Roper
argues that “(b)efore the later years of the seventeenth century, the Highlanders of Scotland
did not form a distinct people. They were simply the overflow of Ireland”.44 One of the ways
in which the myth of a distinct people was created, was by means of the invention of the
tradition of a distinct dress. Thus, the counterfactuals about the ancient origin of the kilt
contributed to the creation of a group identity, and gave people one specific way of belonging
to this group, namely by wearing the kilt. This probably is one of the reasons why the myth of
kilt was so successful.
Can something similar be shown to be true in the case of religious traditions? In other
words, is it typically the case that the counterfactuals that have proved to be of value in the
creation of religious traditions, can be shown to have made substantial contributions to the
functions of these? One would expect that where people invent a past, and this invented past
contributes to the establishment of a tradition, this invented past should have an obvious
7
connection to the functions of that tradition. Otherwise, why would people care to make up a
fictitious past? And how could the invented past have come to contribute to the creation of the
tradition? And why would the members of that tradition continue to believe in it, ignoring the
evidence to the contrary, which at least in some cases is overwhelming?
I will briefly discuss four examples. The first is an example that will be discussed in
more detail in another contribution to this volume: the use of the Exodus story to create a
national and religious identity for the Ephraimite population. Historians like H. Jagersma
argue that it is most unlikely that all the tribes of Israel left Egypt in a massive Exodus.
Probably, small groups of people related to only some of the later tribes of Israel left Egypt on
several occasions. There is evidence that these groups took different routes through the desert,
and arrived in Israel at different times and places.45 If that is the case, the original stories about
the ways in which these small groups left Egypt, must have been transformed into the ‘Exodus
myth’ later on, e.g. after the creation of an autonomous Israelite state under Jerobeam I (ca.
925). This partly fictitious history served to create one homogeneous national and religious
identity for the rather heterogeneous population of Israel. If we may assume that this picture of
how the Exodus story came into existence is correct – and I am not able to judge whether that
really is the case46 – the function of the Exodus counterfactuals seems quite clear. They define
the population of Israel as one nation with one religion, and thus constitute a group identity.
They simplify a complex history and thus create order in a complex world. They testify to
YHWH as a powerful and liberating God, and thus provide reason to worship YHWH, that is
to say, they give guidelines for (religious) action. Finally, they nicely illustrate Popper’s thesis
about the innovative power of tradition: the Exodus-story was not created by canvas-cleaning,
i.e. by denying the existing (his)stories and making up a new one, but by revising existing
historical material in such a way as to make it fit with the new situation (the State of Israel).
The second example that I would like to discuss is that of the Pseudo-Isidoric
Decretals.47 This collection of council documents and letters of several early popes was used
by bishop Hincmar of Laon in his conflict with his uncle archbishop Hincmar of Reims (9th
century). The Decretals, which are a mix of genuine and fake documents (cf. again Popper’s
arguments against canvas-cleaning), supported the authority of local bishops over against
archbishops and secular authorities by invoking the primacy of the Pope. At the time the Pope,
being far away in Rome, must have appeared less threatening to bishops like Hincmar of Laon
than his nearby uncle must. From the 11th century onwards, the Decretals gained authority,
and several Popes used them in support of their own authority. Though their genuineness was
doubted from the very beginning, they were unmasked as counterfeits only in 1628 by the
protestant theologian David Blondel.48 Here again, a mix of traditional and recent
(counterfactual and counterfeited) material is used to a specific end; its genuineness is
doubted mainly by those who would profit by their unmasking, and the fraud is finally proved
by a protestant theologian, who cannot count as disinterested either. The counterfactual
material was introduced primarily because of the ‘norms of behaviour’ that were implied: they
allowed local bishops a relatively large autonomy. And later on, they were invoked primarily
because of related norms of behaviour: they emphasised the primacy of the Pope, and
consequently, the submission of other secular and ecclesiastical authorities. The
counterfactuals served other functions of traditions as well. The interpolations in and
additions to genuine historical documents suggested an unambiguous interpretation of
ambiguous documents. As in the case of the Exodus, complex historical evidence was
simplified so as to create order out of chaos. The Pseudo-Isidoric Decretals are a clear
example of how partly counterfactual material can be used in the establishment of religious
traditions. They are also a clear example of the unmasking of such counterfactuals, and of
8
tradition-generating counterfactuals that are nowadays generally regarded in a pejorative
way.49
The third example of counterfactuals employed in the establishment of a new tradition
is the view of (Reformed) scholasticism adopted by many protestant theologians. In
contemporary protestant theology, scholasticism meets with little sympathy. It is seen as one
of the ways in which Roman Catholicism went wrong, and is commonly associated with
speculative hair-splitting. Thus, it is seen as one of the defining characteristics of protestant
theology that it is anti-scholastic. In this way, much is made of the discontinuity with RomanCatholic theology. A major problem for this view is the existence of the phenomenon called
‘Reformed scholasticism’: Calvinist and Lutheran theologians mainly from the 17th-century
who considered themselves as Scholastic theologians and employed Scholastic methods. On
the view we have just sketched, these theologians constitute a lapse from the true
Protestantism. The Reformers wanted the Church and theology to go back to the roots (sola
Scriptura); for these theologians, this no longer sufficed and they allot an equal place to sinful
human rationality. Recent research, however, suggests that the anti-Scholasticism of
Reformation theology is a later invention.50 Calvin and Luther themselves were in many
respects children of their time, and there is much more continuity with the Middle Ages than
discontinuity. Moreover, Calvin and Luther continued to use Scholastic distinctions, and did
not attack the Scholastic method as such. There is more continuity between late medieval
theology, early Reformation theology, and Reformed Scholasticism, than the traditional view
admits. Here again, the past is reconstructed in the course of the establishment of a new
tradition. In this way, a complex past is simplified, guidelines for doing theology are given, a
new identity for Protestant theology is defined, some kind of ‘group code’ is instituted, and
the existing tradition is changed.
A fourth example of counterfactuals employed in the establishment of a new tradition
has already been introduced above: the alleged special origin of the Book of Mormon. It is not
difficult to understand what the functions are of the counterfactual that Joseph Smith
discovered the Book of Mormon written in hieroglyphs on golden plates, and that he is not its
author but its translator. It clothes the book with a divine authority, and – given the contents of
the book – which claims that the American Indians are Israelites to whom God’s promises
were made, and who should be evangelised – provides a clearly defined group identity to the
American Mormons, sets them a goal, etc., etc.
In short, the four examples that I have briefly presented show that the counterfactuals
which play a role in the establishment of religious traditions typically are closely connected
with the functions of these traditions. This is what we expected, and can hardly count as a
spectacular result. It is impossible, of course, to conclude on the basis of a selection of
examples that this must always be the case.
4 Conclusion and outlook
Though Hobsbawm and Ranger’s volume, and the introduction by Hobsbawm himself, have
made the concept of ‘invention of tradition’ very popular, Hobsbawm and Ranger themselves
did not define their terms very clearly. Hobsbawm gives some indications of what he
understands by a tradition, but these are not very precise, and in so far as they are, they are not
always correct. His view that traditions are invariable, for example, is mistaken (and
incompatible with some of the examples Hobsbawm himself provides). By comparing
Hobsbawm’s views with those of Popper, I have tried to arrive at a more tenable view of
tradition. The other central concept, that of invention, I have shown to be fundamentally
ambiguous; its meaning depends on whether it is a propositional or non-propositional item
that is invented. Hobsbawm does not remove this ambiguity, and uses ‘invention’ in both
9
ways. The same goes for Hobsbawm’s followers. I have argued that we need not remove this
ambiguity, as long as we are aware of it. Moreover, I have argued that it would be a pity to
remove the aspect of the invention of counterfactuals (e.g., the making up of a fictitious
history) as one of the aspects of the establishment of a tradition. As I understand Hobsbawm
and Ranger, there is nothing much new about their use of ‘invention of tradition’ except
precisely this, that they do in a suggestive way draw attention to the fact that invented
counterfactuals may play an important role in the establishment of new traditions. It is not
surprising, then, that much is made of this aspect by many of those who have taken up the idea
of ‘invention of tradition’. Since this seems to me the most interesting aspect of Hobsbawm’s
article on the invention of tradition, I have devoted the third part of my paper to it. I have tried
to show that in order to play a successful part in the establishment of a tradition, a
counterfactual (1) must be accepted as true by those inside this tradition, and (2) must be
closely connected with the functions of tradition. The first seems to me a conceptual necessity
that can be proven, the second an empirical hypothesis which I have tried to make plausible on
the basis of four examples. The evidence, however, is far from conclusive. As I understand
Hobsbawm, he does not give us many more clues regarding the role of counterfactuals in the
invention of tradition.
This brings me to my final point: What is the relevance of Hobsbawm and Ranger’s
volume The Invention of Tradition for theology and religious studies? In my view, this
relevance is limited to the historical approach to religion. The volume confronts us with the
idea that many traditions that have been presented to us as ancient may well be constructions
of more recent date. Moreover, Hobsbawm draws attention to the way in which the illusion
that a tradition is very old may be created, i.e. by the use of counterfactual claims about the
origin or past of the tradition in question. Thus, Hobsbawm & Ranger’s volume can function
as an eye-opener within historical studies of religious traditions. Because the concept of
‘invention of tradition’ is a rather under-developed notion in Hobsbawm himself, however,
the concept cannot be more than an eye-opener; it cannot function as a tool, or as a method or
theoretical framework. Where it does function as an eye-opener, one would expect a
demythologising, deconstructing, critical and unmasking approach to religious traditions.
For the systematic disciplines within theology, such as dogmatics, ethics and
philosophy of religion, the concept seems less important. They do not primarily deal with the
genesis of religious traditions – though facts about that genesis discovered by historians may
be very relevant. Thus, a Mormon dogmatician who becomes convinced that Joseph Smith
wrote the Book of Mormon, and did not translate it, will probably change his dogmatic views.
Probably, but not necessarily! He might argue, for example, that though Joseph Smith had to
lie about the genesis of the Book of Mormon in order to make it acceptable to some of his
contemporaries, this does not take away our experiential evidence of the truth of the things
contained in the book. This is what happened in the case of the pseudo-Isidoric decretals; it
has been unmasked as a counterfeit, but the authority of the Pope has increased rather than
decreased.51 However this may be, although those involved in the systematic disciplines may
need to be aware of recent developments in historical scholarship; it is in no way their task to
repeat historical research. For them, the idea that traditions may be invented is indirectly
relevant rather than directly. This does not mean, of course, that the creative application of
this idea by historians may not lead to interesting and even fascinating results, such as can be
found in some of the other contributions to this volume.
1
Eric Hobsbawm, “Introduction: Inventing Traditions”, in: Eric Hobsbawm, Terence Ranger (eds.), The
Invention of Tradition, Cambridge 1983, 1-14; I will quote from a ‘Canto Reprint’ from 1996. Paul Post, “Rituals
and the Function of the Past: Rereading Eric Hobsbawm”, Journal of Ritual Studies 10 (1996), 100 n.7 gives a
10
useful survey of the use of ‘invention of’ terminology, but the titles he mentions are either very recent, or do not
mention tradition.
2
Karl R. Popper, “Towards a Rational Theory of Tradition”, The Rationalist Annual 66 (1949), 36-55 (42-43).
The lecture was reprinted in his Conjectures and Refutations: The Growth of Scientific Knowledge, London
3
1972, 120-135. Here, I will refer to the original publication in the Rationalist Annual. The phrase ‘invention of
tradition’ occurs at p.44: “The Greek philosophers invented a new tradition”. Popper also addresses the rise of
traditions in another lecture included in his Conjectures and Refutations: “Back to the Presocratics”, (the
presidential address delivered to the Aristotelian Society in October 1958), 136-165 (150 ff). There, he also uses
the phrase ‘invention of tradition’: “the critical or rationalist tradition was invented only once” (151).
3
Paul Post claims that the concept of invention of tradition is popular in cultural studies, in the social sciences
and in historical studies; he himself provides a number of examples of the use of the concept in ritual studies and
liturgical studies. See Post, “Rituals and the Function of the Past”, 85-107.
4
An exception is Post, “Rituals and the Function of the Past”.
5
Popper, “Rational Theory”, 38.
6
Popper, ‘‘Rational Theory”, 38.
7
Popper, “Rational Theory”, 39-40.
8
Hobsbawm, “Introduction”, in: Hobsbawm, Ranger (eds.), Invention of Tradition, 1. In fact, this quotation is
part of Hobsbawm’s definition of invented tradition; it seems clear, however, that the part which I quoted is his
definition of tradition-in-general, and that the lines which follow – and which we will discuss below – are
intended to distinguish invented traditions from other traditions.
9
Hobsbawm, “Introduction”, 2.
10
Determinism cannot be falsified empirically. Implicit in the main text is my definition of a metaphysical
statement: a metaphysical statement is a factual statement which in principle does not admit of empirical
verification or falsification. The clause ‘in principle’ is inserted to distinguish metaphysical statements from
factual statements like ‘some dinosaurs had grey eyes’ which may in fact not admit of verification or falsification,
but could in principle be verified by a well-placed observer (an observer co-existing with dinosaurs). Cf. Vincent
Brümmer, Theology and Philosophical Inquiry: An Introduction, London 1981, 268-282; H.J. Adriaanse, Het
idee van een religieus apriori, Amsterdam 1998, 12, 23-25.
11
As the famous dictum says: Ought implies can. Rules and guidelines presuppose a context in which the rulefollower is capable both of doing and of refusing to do what the speaker requires of her. Cf. Brümmer, Theology
and Philosophical Inquiry, 25.
12
This is the question of the identity of a tradition. Cf. Vincent Brümmer, “The Identity of the Christian
Tradition”, in: Marcel Sarot, Gijsbert van den Brink (eds.), Identity and Change in the Christian Tradition,
Frankfurt a.M. 1999, 23-41.
13
Here, ‘function’ should not be taken in the sense of ‘goal’, but in that of ‘proper activity’, ‘appropriate
behaviour’ or ‘role’. Cf. my “Religion, Meaning and Imitation: The Christian Ideal of Imitatio Christi as a Way
of Making Sense of Life’, in: Sarot, van den Brink (eds.), Identity and Change, 73-90, esp. 73-74.
14
Hobsbawm, “Introduction”, 9.
15
Popper, “Rational Theory”, 43.
16
Popper, “Rational Theory”, 49.
17
Popper, “Rational Theory”, 49.
18
Popper, “Rational Theory”, 49.
19
Popper, “Rational Theory”, 50
20
Popper, “Rational Theory”, 51; cf. Conjectures and Refutations, 343-344.
21
Popper, “Rational Theory”, 51.
22
As Jan Platvoet and Karel van der Toorn remark, “The fact that one’s identity is determined by the group to
which one belongs means in practice that an individual has a multiple identity”. Jan Platvoet, Karel van der
Toorn, “Pluralism and Identity: An Epilogue”, in: Platvoet, Van der Toorn (eds.), Pluralism and Identity: Studies
in Ritual Behaviour, Leyden 1995, 352. One and the same person can be Christian, scientist, socialist, etc.
23
Cf. Arthur S. Reber, The Penguin Dictionary of Psychology, London 1985, 362.
24
Cf. Popper, “Rational Theory”, 52-53 who argues that we speak of an institution where a group fulfils “certain
prima facie social functions … which serve certain prima facie social needs …, while we speak of traditions
mainly when we wish to describe a uniformity of people’s attitudes, or of ways of behaviour, or of their aims or
values, or of their tastes. Thus traditions are perhaps more closely bound up with persons and their likes and
dislikes, their hopes and fears, than are institutions. They take, as it were, an intermediate place … between
persons and institutions”.
25
Here I use the term in a broad sense, and not exactly in the same way as Bertrand Russell, who introduced the
term, used it.
11
26
John Sinclair (ed.), Collins Cobuild English Dictionary, London 1987, s.v.
Hobsbawm, “Introduction”, 4 et passim. Popper employs similar terms.
28
Hobsbawm, “Introduction”, 1, 4.
29
Hobsbawm, “Introduction”, 1. That Hobsbawm does not take the term ‘invention’ so broadly that all emerging
traditions can be called invented becomes clear from one of the opening sentences of his “Introduction’:
“‘traditions’ which appear or claim to be old are often quite recent in origin and sometimes invented” (italics
mine – MS).
30
Hobsbawm, “Introduction”, 1.
31
Hobsbawm, “Introduction”, 2.
32
Hobsbawm, “Introduction”, 7.
33
Hugh Trevor-Roper, “The Invention of Tradition: The Highland Tradition of England”, in: Hobsbawm, Ranger
(eds.), Invention of Tradition, 15-41.
34
Here I depend on Post, “Rituals and the Function of the Past”, 96, who summarises a Dutch translation of the
speech that appeared in the Dutch newspaper De Volkskrant, December 4 1993.
35
Post, “Rituals and the Function of the Past”, 96.
36
Post, “Rituals and the Function of the Past”, 85.
37
Post, “Rituals and the Function of the Past”, 92.
38
Thus, I do not use counterfactual here in the sense of counterfactual conditional. An example of a
counterfactual statement is ‘Pope John Paul I has been pope from 1978 to the present day’. An example of a
counterfactual conditional is: ‘If John Paul I had lived much longer, he would have removed the restrictions on
contraceptives’.
39
Cf. Brümmer, Theology and Philosophical Inquiry, 17-18.
40
See, e.g., Brümmer, Theology and Philosophical Inquiry, 26-27; William A. Christian, Meaning and Truth in
Religion, Chicago 1964, 149; C.S. Lewis, The Problem of Pain, London 171964, 14-16.
41
Brümmer, Theology and Philosophical Inquiry, 139; Ian Barbour, Myths, Models and Paradigms: The Nature
of Scientific and Religious Language, London 1974, 143. Cf. Christian, Meaning and Truth, 150-155.
42
Brümmer, Theology and Philosophical Inquiry, 17-18; the term ‘success conditions’ appears on p.21.
43
For a survey of the evidence, see Wesley P. Walters, “The Human Origins of the Book of Mormon”, originally
published in the Journal of Pastoral Practice 3/3 (1979), 123-152, now available on http://www.irr.org/mit/hobom1.htm and http://www.irr.org/mit/HO-BOM2.HTML. See also H. Michael Marquardt, Wesley P. Walters,
Inventing Mormonism, Salt Lake City, Utah 1994, and cf. the review by Joel B. Groat, available on
http://www.irr.org/mit/inventbk.htm, which shows by means of examples that LDS Church leaders attach great
value to the historical accuracy of some tradition-generating counterfactuals.
44
Trevor-Roper, “Invention of Tradition”, 15.
45
H. Jagersma, Geschiedenis van Israël in het oudtestamentische tijdvak, Kampen 1979, ch.3.
46
My main worry is that the absence of any independent evidence in favour of the Exodus story seems sometimes
to be taken as positive evidence against its historicity. This way of dealing with the absence of evidence is a form
of the argumentum e silentio or ad ignorantiam fallacies.
47
See, e.g., H. Fuhrmann, “Pseudo-Isidor”, LThK3 VIII, 864-866, C.C. de Bruin, Handboek der kerkgeschiedenis
II, Leeuwarden 61980, 34-36.
48
David Blondel, Pseudo-Isidorus et Turrianus Vapulantes, Genève 1628.
49
Is this merely because nowadays we have less sympathy for the aims served by the Decretals (establishing
papal primacy) than for the aims served by the Exodus story (establishing the Jewish identity)? Or is it also
because in the case of the Exodus story, the fabrication of history has not been definitively proven?
50
See Willem J. van Asselt, “Studie van de gereformeerde scholastiek: Verleden en toekomst”, Nederlands
Theologisch Tijdschrift 50 (1996), 290-312; W.J. van Asselt, P.M. Wisse e.a., Inleiding in de Gereformeerde
Scholastiek, Zoetermeer 1998, esp. Ch.2.
51
There are more ways in which traditions once they have been established could cope with the falsification of
factual claims that are presupposed within these traditions. Instead of admitting the claim to be untrue, one might
also try to reinterpret it. In this way, a claim that is not literally true (God created the earth in six days, and took a
rest on the seventh day) could be reinterpreted either as a factual claim that cannot (yet) be falsified, e.g., ‘for
their (continuing) existence, heaven and earth depend on God’, or as a non-factual claim, e.g., the evaluative
claim ‘heaven and earth are good, not evil’. It seems implausible to me, however, that counterfactuals that play an
important role in the creation of a new tradition, can be interpreted symbolically while they play this role. The
claim that the earth was created by God has, to my best knowledge, never in this way played an important role in
the creation of a new tradition. The claim that Jesus founded the Church, and instituted the sacraments, has
played such a role; at the time it played this role, however, it was taken literally, and not symbolically. Only those
within the Church for whom not much depends on the question whether Jesus really had the intention to found
27
12
the Church, and for whom, as a consequence, the claim is no longer tradition-generating (is no longer needed in
justification of the tradition), will be satisfied with a symbolical reinterpretation of this claim.
13