The following text has been automatically reproduced by an Optical Character Recognition (OCR) algorithm. It may not have been checked over by human eyes. For matters of precision please consult the original pdf.

Romanticism and technology

Romanticism and
Andrew Bowie

Romanticism and technology are often regarded as
inherently at odds with each other, one supposedly
relying upon a desire to get in touch with a nature in us
and outside us which the modern ‘technologized’ world
risks losing sight of altogether, the other upon the
domination of external nature for human purposes.

However, the term ‘Romanticism’ (leaving aside the
term ‘technology’, to which I shall come later) is now
used in such divergent ways, being attributed to modes
of thought and artistic currents which often entail
contrary assumptions, that productive discussion of
‘Romanticism and Technology’ might seem impossible.

What I want to claim, then, is that the ideas presented
here are simply better than many ideas which others term
Romantic, and, more importantly, are superior to many
philosophical alternatives, ‘Romantic’ or not.

Historically the matter is much simpler: I am concerned
with an admittedly ‘ideal-type’ ‘early Romanticism’, and
not with later manifestations of ‘Romanticism’. Given
the notorious fate of many ideas termed ‘Romantic’ ,1 it
is vital that this restriction be clear from the outset.

Romanticism has often been seen as one of the great
intellectual and political failures of modernity: I want,
however, to argue that what I mean by Romanticism may
not yet have had its historical chance.

The first part of the following passage by Novalis
appeared recently in the London Underground, as part of
the advertising for cultural events running under the
heading of ‘Deutsche Romantik’. The reversal in the
second half of the passage was, characteristically,

By giving the commonplace a higher sense, the
usual a mysterious appearance, the known the
dignity of the unknown, the finite an infinite
appearance, I romanticize it – The operation is the
other way round for the higher, the unknown, the
mystical, the infinite – it is logorhythmized by this
connection – It gains an everyday expression. 2

Novalis, remember, was a mmmg engineer. His
neologism ‘logorhythmized’, which combines the ideas
of rational ordering, verbalization and mathematical
progression with the idea of the music inherent in the use
of everyday language, points to the central issues in early
Romantic philosophy. In this sense the music of
Beethoven, with its unparalleled combination of rigorous
order and free invention, is the perfect metaphor for such
Romantic thinking. The fact is that Novalis’s two-way
movement between the everyday and the imaginative,
the cognitive and the aesthetic, which is characteristic of
Romanticism’s relationship to the natural sciences and
technology, has too often been reduced to a one-way
movement between the aesthetic and the cognitive. Most
of the ideas at issue here initially developed somewhere
between 1794 and 1801, particularly in Jena, in the work
of Novalis, the earlier Friedrich Schlegel, and parts of
the work of Schelling and Schleiermacher, as a response
to Kant’s critical philosophy, and were apparently finally
laid to rest by the middle of the nineteenth century by
materialist critics, following the demise of Hegelianism.

That Romantic philosophy could really be laid to rest in
this way is, given its predominant concerns, very
unlikely: the issue which lies at the heart of Romantic
philosophy is the status given to reason in the light of the
modern demand to establish a ground for rationality
without recourse to a divinity.

A recent example of issues which were central to
Romantic philosophy can indicate what may be at stake
in any contemporary revival of Romantic concerns. In
the debate between sociologists and historians of science,
and a variety of scientific realists, concerning whether
science is a ‘social construct’, one of the realists treated
us to the assertion that ‘Somewhere buried in the
awesome complexity of Nature, lies the truth. It is the
task of science to disclose that truth … ‘ 3 Another
informed us that taking tribal beliefs about the moon as
just as true as our own beliefs was ‘deeply silly … when
you ask practical questions’.4 He did not trouble to ask

Radical Philosophy 72 (July/August 1995)


what ‘true’ might mean, having decided it meant
correspondence to the ‘ready-made’ facts, thereby
ignoring the fact that this was not necessarily even
compatible with his own assertion about truth and
practicality. Seeing the moon as a calabash (his example)
could, presumably, be perfectly practical for some
purposes within a traditional society, and whether you
think going to the moon is practical or not rather depends
upon who you are and what you think is valuable. The
rational response to all this was to say, as some
respondents did, that sciences are, of course, social
constructs, if by that you mean they would not exist if
societies did not construct them, but that this was no
warrant for saying we must validate the practical and
theoretical claims of a science merely in terms of analysis
of the society that constructs it. We don’t entrust the
building of our aircraft to sociologists of science, but, of
course, neither do we regard aircraft builders as the final
arbiters of the scientific, social, economic and political
significance of what they get up to. The fact is that much
of this battle is really about understanding what it is that
needs evaluation. The answer has something important
to do with what ‘Romanticism’ has been understood to
mean, which is why we now need an alternative account

T11e Realists

October 1nth

of Romantic philosophy.

One simple answer to the question of what generates
so much heat and so little light in this context is that some
of the realists – those whom Hilary Putnam terms
‘metaphysical realists’ – want to think science is
‘absolutely true’ of the ‘ready-made world’. The problem
with this position is actually quite easy to show, and
Romantic philosophy was very good at showing it: how
does anyone know that the truth is there hidden in nature
for science to find if they have not either already found
the truth or presupposed that whatever it is that their
science finds out is, or is on the way to being, the final
truth? Would it not be more philosophically tactful, for
example, to understand the sciences as constituting just
one of the dominant ways in which we now respond to
the world?5 Such responses also involve aesthetic,
practical and contemplative approaches, which
themselves feed into scientific praxis, and vice versa.

Inescapable circles loom in every direction if we wish to
make the stronger claim that ‘science represents the
truth’. The Romantic conception of truth is not least
interesting in this respect because it suggests what is
problematic about the very idea of the ready-made world.

Friedrich Schlegel maintains in ‘On Incomprehensibility’

of 1800: ‘In truth you would be distressed if the
whole world, as you demand, were for once
comprehensible’;6 elsewhere he asserts. ‘If
absolute truth were found then the business of
spirit would be completed and it would have to
cease to be, since it only exists in activity.’?

Importantly, Schlegel does not further specify
what this activity is grounded in, because, as we
shall see, this would lay claim to something
which Romantic philosophy was perhaps the first
to renounce.

As opposed to the realists the suspect
fraction of the relativist sociology camp looks, in
a way Schlegel does not, for a big story which
will explain science as really something else – as
driven, for example, by the will to power or the
will to knowledge, or as Heidegger’s
‘enframing’, the manifestation, most visible in
modern technology, of Western metaphysics’

grounding of all truth in the subject as ‘lord and
master of nature’. The unappealing alternative
between hard-line realism and this latter kind of
relativism – which in fact often appeals to a
hidden absolute like the ‘will to power’ – offers
us another instance of what has often been termed

……………… Romanticism. 8


Jacobi and romantic philosophy
Let us now therefore try to establish a few vital historical
and philosophical points of orientation, of the kind that
too rarely get discussed when looking at ‘Romanticism’

in relation to ‘positivism’. The underlying issue raised
by the debate just referred to is the question of what
makes something true. As is well known, in 1781 Kant
tried to overcome his crisis concerning this question,
occasioned by Hume’s arguments on causality, by
suggesting that our thinking was what made things true,
not the world as it was ‘in itself’. Kant maintained that
we had no right to assert that anything was true beyond
the bounds of what we could judge to be true by applying
prior conditions of thought to what was presented to us
in ‘intuition’. He was thereby understood as opening up
a rift between thinking and the world, which his
philosophical contemporaries and successors then tried
to bridge, and which he himself clearly did not think was
the last word on our relationship to nature. Kant toyed
with ways of trying to reunite mind and nature in the
Critique of Judgement, which was vital to Romantic
philosophy, but he kept the links he thought most
significant in the realm of orienting fictions. The rift
Kant’s position can be seen as establishing creates
precisely the space in which a debate like the one just
cited can arise. If Kant’s way of grounding knowledge in
the subject turns out to be a failure there are two main
possibilities. One will be to say, especially given the
evident problem-solving success of modern science and
technology, that knowledge is objective in a pre-Kantian
manner. This is the position of the metaphysical realists,
who regard what science, particularly physics, tells us as
offering an ‘absolute conception’ untainted by the
interference of subjectivity. The other is to maintain both
that knowledge is only intelligible within frameworks of
the kind exemplified in the – probably flawed – concept
of the transcendental subject, and that we have no
necessary access to a pre-existing ‘framework of
frameworks’ which would ground all the various ways
of understanding what knowledge is, though we may,
pragmatically or ethically, wish to try to converge
towards such a framework. In this sense there is no
‘science of knowledge’. This leaves open the door for
sociologies and histories of science, critiques of science
as ideology, hermeneutics and the rest, all of which deal
with some aspect of the conditions of possibility of
whatever is now meant by ‘science’, and all of which
derive directly or indirectly from Romantic philosophy.

‘Science of knowledge’ is, importantly, the way in
which the title of the ‘Wissenschaftslehre, in which
Fichte tried to establish the absolute foundation Kant
failed to provide, has often been translated into English.

The very problem of self-reference implicit in this
translation – does such philosophy not consist in the
attempt to gain ‘knowledge of knowledge’, thereby
leading to the need for a special kind of prior
‘knowledge’? – is one ofthe vital issues that have been at
the centre of much subsequent philosophy. Romantic
philosophy is formed in relation to some of the
alternatives which appeared in key responses to Kant’s
philosophy like that of Fichte: in this sense Romantic
philosophy is the first manifestation of what happens
when transcendental philosophy realizes how
problematic foundationalism can become. 9 Its
relationship to what we now term ‘technology’ can be
understood by looking at some of the responses to
Kantian problems.

The seriousness of the objections to Kant’s dualism
can be gauged from F. H. Jacobi’s response in ‘On
Transcendental Idealism’ of 1787: 10
For even if according to [Kantian philosophy] it
can be admitted that a transcendental something
may correspond as cause to these merely subjective
beings (Wesen [by which he means ‘appearances’]),
which are only determinations of our own being, it
yet remains hidden in the deepest obscurity where
this cause and what the nature of the relation it has
to its effect is.11
In this view there is nothing we can know w~ich could
show how it is that we hold things to be true, given that
Kant had shown how previous explanations in terms of
the world ‘in itself’ were invalid because they claimed
more than they could ever prove. The Romantic answers
to the problems in Kant’s dualism suggested by Jacobi
lead us rapidly in the direction of issues to do with
technology. This is evident in Schelling’s assertion in
the 1797 Ideas Towards a Philosophy of Nature that:

one can push as many transitory materials as one
wants, which become finer and finer, between
mind and matter, but sometime the point must
come where mind and matter are One, or where
the great leap that we so long wished to avoid
becomes inevitable. 12
Schelling thinks of the difference of mind and matter as
only ever a relative difference within a totality which
encompasses both, rather than, as the materialist version
of an apparently similar naturalistic philosophy does,
privileging one – matter – over the other. The vital
problem with this kind of monism was that it could be
construed as leading to the most controversial philosophy
in this period: Spinozism. Schelling himself will spend
most of his philosophical career in an overt and covert
dialogue with Spinozism, in which many of the


subsequent issues concerning technology associated with
Heidegger, the Frankfurt School, and even such thinkers
as Daniel Dennett, are already adumbrated. The tension
between the Spinozist aspects of the thinking of this
period and the attempted alternatives to Spinozism can
provide a model which reveals how the relationship
between Romanticism and technology can be understood.

The totality which encompasses mind and matter was
referred to in this period as the ‘Absolute’ or the
‘unconditioned’, terms which are generally
misunderstood. Discussion of ‘the Absolute’ in the
English-speaking world has tended to move in one of
two directions, neither of which owes much to the
Romantic conception. In the first because it is beyond
anything we can know discursively, and can therefore
only be ‘known’ ‘intuitively’, the Absolute is regarded
as a mystical issue for theo]ogians: this is often, wrongly,
assumed to be the Romantic position. In the second, the
Absolute has been thought of as the account of the world
which is uncontaminated by the relative ways the human
subject ‘sees’ it within particular local interpretations:

truth here is the final correspondence of ‘objective’

scientific knowledge to the ready-made world. Bernard
Williams, who in this respect (though not in others) is
thoroughly in line with a Spinozist conception, has
approvingly termed this version of truth the ‘absolute
conception’ .

The issue for Romantics that leads to questions about
the Absolute is how to come to terms with the relativity
of particular claims to knowledge without becoming
trapped by the paradoxes of relativism and the regresses
we are about to consider. The continuing significance for
contemporary theory of what is involved in the
Romantics’ reflections derives from their attempt to
sustain an orientation towards truth, even as they
renounce the idea that specific truths can be
epistemologically grounded. In an argument whose
echoes are evident both in Nietzsche’s view, in The Birth
of Tragedy and in other texts, of modern science as
leading to the abyss of ‘nihilism’, and in Heidegger’s
work on the principle of sufficient reason and on
technology, Jacobi first showed, thereby actually
bringing the world ‘nihilism’ into wider currency, why
the ‘unconditioned’ was to remain such an issue for
modern thought. 13 J acobi’ s contributions to the
‘Pantheism controversy’ that began in 1783 – the
controversy over whether G. E. Lessing, the central
figure of the German Enlightenment, was a Spinozist and
thus to all intents and purposes an atheist – revealed a
regress which, in various forms, both bedevils attempts
at systematic philosophy in the modern period, and
prevents metaphysical “realist positions being


philosophically defensible. Jacobi convincingly shows
why foundational philosophy leads to problems
philosophy cannot solve: it is this insight that is so
significant for the Romantics and for a perception of
technology which can be developed via Romantic

The key fact about Spinozism in this context was its
establishing of a system on the basis of ‘all determination
is negation’. This aspect of Spinozism can be linked to a
model of technology which is still very current in
contemporary debate. 14 In Spinozism every aspect of
being gains its identity via its ‘negative’ relations to other
aspects, not via anything intrinsic to itself. Spinoza’s
thought can, as such, be construed as already mapping
out the basis of digital technologies, in which the most
divergent information is rendered equivalent by an
encoding that gives each element its identity or function
via its relations to the other elements, and not via
anything inherent to the element itself. The totality of the
dependent aspects of being is Spinoza’s God, who, as
Henry Allison says, ‘functions … as the logical ground
of things. The latter follow from his nature in precisely
the same way as the conclusion of a valid argument
follows from its premises’ .15 In line with the Kantian
suspicion of dogmatism, J acobi suggests that if all our
knowledge of what things are depends on the principle
of sufficient reason, which both he and Kant understand
in terms of the subject seeking the ‘conditions’ of that
which it would explain, then all our explanations depend
merely upon whatever we began with, leaving a problem
of circularity which must be overcome if the explanation
is to be grounded in any final manner. If everything is
itself both conditioned and the condition of something
else, what about the totality of conditions – which Jacobi
comes to call ‘being’ – that cannot be of the same order
as what we construct in the differential operations of our
thinking? Is being also to be explained, as it is in
Spinozism, by the principle of sufficient reason?16
J acobi’ s main worry concerned the effects of Spinozism
on theology, because it turned God, and thus the
existence of the world, into something which excluded
anything beyond what he termed the ‘mechanical’.

Though this side of Jacobi’s argument may no longer be
of any concern to those not bothered about theology,
J acobi’ s insight does have serious philosophical
consequences, which relate to real consequences for the
world in which the effects of the natural sciences become
so dominant.

The problem J acobi reveals is that we have no
ultimate way of philosophically explaining either what
we think truth is, or of explaining our very awareness of
a world which gives rise to questions of truth. If each

particular belief we hold true is dependent upon other
beliefs, none of these ‘conditioned’ beliefs can ever show
why anything is held true at all, nor can it explain how it
is we are aware of the world of conditions that constitutes
what we claim to know. For this reason Jacobi insists on
the need for a conception of ‘belief’ , which he defines as
‘holding as true’ (Furwahrhalten), and which is prior
even to axiomatic truths, because the possibility of truth
cannot itself be demonstrated. Something like this view
has been recently echoed, of course, by Donald
Davidson, who talks of our ‘general and pre-analytic
notion of truth’, and of an ‘intuitive grasp we have of the
concept’ .17 For Davidson, as for Jacobi, we cannot finally
give a theoretical description oftruth, because we always
already rely upon it in order to describe or understand
anything at all. The need for the assumption of what is
‘unconditioned’ seems therefore inescapable if there is
to be a philosophically groundable account of truth, but
the unconditioned itself seems philosophically

In the 1787 Preface to the first Critique, Kant
maintains that, while reason most postulate the
‘unconditioned … in all things in themselves for
everything conditioned, so that the series of conditions
should thus become complete’, 18 by restricting
knowledge to appearances rather than ‘things in
themselves’, the contradiction of seeking ‘conditions of
the unconditioned’ can be avoided. This did not solve
the real problem, however, as Fichte realized when he

maintained that the subject, the condition of possibility
of appearances, must therefore itself have an
unconditioned status: ‘It is … the ground of explanation
of all facts of empirical consciousness that before all
positing in the I the I itself must previously be posited.’ 19
The I cannot be posited in the same terms as that which it
is to explain, because this would lead to another vicious
circle of self-reference, or to a regress: how does the I
explicate itself as the ground of knowledge, if it already
has to presuppose its own existence as spontaneous ‘nonobject’ before it can explain anything objective? As such
the I became a self-caused ‘deed-action’ (‘Tathandlung’)
for Fichte, not another conditioned ‘fact’ in the knowable
world. In one sense Fichte’s I thereby echoed Spinoza’s
God, which is also ’cause of itself’. Fichte’s position
therefore leads Jacobi in 1799 to claim in his published
letter Jacohi to Fichte that
A pure, that is a thoroughly immanent philosophy;
a philosophy made only of One piece; true system
of reason is only possible in the Fichtean manner.

Obviously everything must only be given in and
through reason, in the I as I, in egoity, and already
be contained in it, if pure reason alone should be
able to deduce everything from out of itself alone. 20
Because he thought this position entailed an indefensible
omnipotence on the part of the I, which consequently
remains trapped in a kind of narcissistic self-enclosure what now usually gets termed, for example by Derrida,


‘self-presence’ – this meant that philosophy could not be
finally grounded. J acobi therefore referred to his own
position as ‘Unphilosophie’, thereby already initiating
the question as to what might come ‘after philosophy’, in
a manner not so far from the later Heidegger’s attempts
to overcome metaphysics. Even more importantly, in our
present context, J acobi maintains that both Spinozism
and Fichtean philosophy have essentially the same

This last contention is the key to the Romantic move
against foundationalism. Jacobi claims that Fichte’s
transcendental idealism is in fact an ‘inverted
Spinozism’ .21 Each aspect of the Spinozist system gains
its identity from its relationships to the other aspects
within the whole, just as in structuralism the signifier
gains its identity via its relations to other signifiers, or as
information is digitally encoded. The vital question lies
in how these systems are grounded. Whereas Fichte’s
transcendental idealist position establishes its system via
the absolute ‘positing’ of the I, the Spinozist system
establishes it on the basis of the ‘substance’, which,
Jacobi asserts, is, like Fichte’s I, ‘nothing but the
unintuitable [in Kant’s sense of ‘intuition’, meaning that
it is not empirically accessible] absolute identity of
subject and object … upon which the system of the new
philosophy, of the independent philosophy of intelligence
[thus of the’ l’ in Fichte’ s sense] is grounded’ .22 The point
is that both theories circumscribe the truth in an internally
grounded system of relationships which is founded in
what is both subject and object – the ‘I’ or Spinoza’s
substance – and is thus, in the last analysis, supposed to
be fully transparent to itself. Whether this is conceived
of in idealist or materialist terms makes no difference, as
the net result is a totality which is ultimately transparent
in itself, like the world of metaphysical realist science.

Now this sort of argument ought to sound familiar to
anyone who has considered accounts of the relationship
of ‘Western metaphysics’ to the growth of the
domination of technology in the modern world, and who
is familiar with Derrida’s deconstructions of Hegel and
Husserl. Given Spinoza’s relationship to Cartesianism,
and Fichte’ s subjectivism, we have a version of the link
between philosophies of the subject and modern
technology’s subjectification of being by its overcoming
the Other in the name of the Same, which is also the
argument Heidegger uses against Nietzsche’s ‘will to
power’ as the culmination of the subjectivist tradition of
Western metaphysics. Is this, however, an adequate
account of the philosophical options on offer in this
constellation? As we shall see, the Romantics offered
important alternatives here, which are still relevant to the
contemporary versions of these debates.


Metaphysics and technology
At this point it is worth becoming shamelessly
metaphorical, by very briefly taking aspects of these
positions as models for some of the contemporary debate
over the new technologies, artificial intelligence, and
artificial life. Claus Emmeche has recently contended,
for example, in relation to the issue of ‘artificial life’,
that ‘If natural processes are essentially computational
ones, we can also study them by producing the same or
similar processes in a computer’ . Reviewing Emmeche’ s
book, Anson Rabinbach concludes that:

If our understanding of technology pre-emptively

dissolves the difference between original and copy,
life and syntax, nature and its mimesis, the
products of the age of digital reproduction become
mere components of a self-referential universe …

the danger is that the computer can all too easily
become a substitute reality, a ‘magical black box’

endowed with infinite wisdom and knowledge of
nature. Like Laplace’s demon, such a computer
would have to be identical with the universe
This seems to me to map pretty exactly onto what is at
issue in the constellation that I have just outlined. The
computer identical with the universe is the Spinozist
God, as understood by Jacobi, in which the totality of
conditions was to be made potentially accessible .to
philosophy by constructing a necessitated system of
internal relations. The real problem is how such a totality
can ground itself in a manner that we find intelligible,
which is why Jacobi’s objections are so important.

‘Re-producing’ life computation ally involves the
reproduction of life as known, as otherwise one could
not claim to have ‘re-produced’ life: what makes the
second product identical with that which is to be
reproduced, unless this identity is somehow presupposed
from the outset? In this sense one requires an account of
that which knows, which leads to another regress: what
knows must presumably already be alive and must know
what being alive is if it is to re-produce life. Much the
same argument can be applied to the problem of selfconsciousness in relation to artificial intelligence: how
does one differentially reproduce the condition intelligence – which enables differences to be intelligible
as differences in the first place? Computers can clearly
register and record differences and identities in, say,
linguistic elements, but this does not mean that syntax
can computationally become semantics: rules of
combination, which rely upon differentiation, have to be
understood if they are to give rise to meanings. In short,
how does one artificially reproduce intelligence without

already presupposing what intelligence is, and without
thereby merely presupposing a ground which, as Jacobi’ s
argument suggested, would lead to a merely arbitrary
circularity if one wanted to claim absolute status for that
ground? The kind of knowledge required to ground
intelligence absolutely cannot be of the same order as the
functions of which artificial intelligence consists:

otherwise we end up with another regress of functions
for functions, which could never account for the
consciousness of what a function means. It is this latter
problem that gives legitimacy to those aspects of Fichte’ s
position which insisted on the impossibility of
objectifying subjectivity, which were developed in
Romanticism, as well as to Jacobi’s contentions about
the necessarily intuitive aspect of truth. The question is
again the problem of the Absolute, which must, as both
Fichte and the Romantics realized, include an account of

The Fichtean and Romantic reason why the Absolute
must also include consideration of self-consciousness,
rather than be thought of in merely ‘Spinozist’ terms,
has, significantly, reappeared in Hilary Putnam’s – in this
sense at least – ‘Fichtean’ arguments against Bernard
Williams’s ‘Spinozist’ absolute conception: ‘It cannot
be the case that scientific knowledge (future fundamental
physics) is absolute and nothing else is; for fundamental
physics cannot explain the possibility of referring to or
stating anything, including fundamental physics itself. ’24
Putnam’s argument raises problems of self-reference
which are analogous to those which occurred in the
Romantic attempt to understand the Absolute. The
problem of self-reference is, crudely, the problem of how
to gain the external perspective – which would allow one
to assert that the universe is ‘really “x'” (e.g. ‘law-bound
matter’, ‘computational functions’) – without leaving the
place from which that assertion is made outside what is
presumed to be the grounding reality.

An obvious consequence of the way this problem
manifests itself has been shown by Herbert
Schnadelbach, in a discussion of answers to the question
‘What is humankind?’ – which inherently requires an
answer to ‘What is nature?’ Canonical recent answers to
the question of humankind are, Schnadelbach suggests:

A highly complex organism; a whim of evolution;
an intrusive special case in the biosystem of the
earth; a historical configuration; a nexus of the
social ensemble – and it is always noticeable that
such explanations are simply applications of what
the particular explanatory sciences (biology,
evolutionary theory, ecology, history, social
science) know more exactly in their particular core
area. 25

All these approaches exemplify Jacobi’ s key insight into
the circularity of epistemological grounding and thereby
always already raise questions about any attempt
scientifically to ground an absolute position. 26
The awareness of the dangers of a scientifically
conceived absolute is crucial to early Romantic
philosophy and forms the real substance of what has
often falsely been assumed to be a generalized antipathy
to natural science on the part of Romanticism as a whole.

The vital fact about the Romantic view is that it does not
conceive of truth as ground or origin in any
philosophically definable sense. Novalis puts the issue
concerning any knowable absolute ground as follows:

‘There is no absolute beginning – it belongs in the
category of imaginary thoughts. ’27 Any beginning must
always be relative to what succeeds it for it to be known
to be a beginning at all, rather than something completely
indeterminate. Because of this dependence any
beginning will always, if one tries to make it absolute,
land one in the kind of circle or the kind of regress we
have been considering. This is the source of Romantic
anti-realism and of the concern with those modes of
articulation, like art, which subvert final grounding or
final interpretations. Such ideas clearly also feed into the
development of a modern hermeneutic approach to truth.

The essential Romantic move is, then, a rejection of a
‘ready-made world’ – the notion of which is already
implied by Novalis’s ‘absolute beginning’ ….. a rejection
which also informs much recent anti-realist philosophy.

Schlegel claims in the Notebooks on Philosophy from
1805 that
there is only one inherited fundamental mistake the fundamentally wrong concept of the thing which takes merely relativefinitude [the particular
transient object]’ as absolute and abstracts the
shadow concept of BEING from life – Being is
merely apparent, finitive only relative. Being =life,
without life, being = appearance. 28
The Romantic world is not a world of things but a world
of relationships whose determinacy depends upon our
interaction with them, an interaction which therefore has
no final foundation either in the subject or the object.

The early Schelling sees scientific experiments as

Every experiment is a question addressed to nature
that nature is forced to answer. But every question
contains a hidden a priori judgement; every
experiment which is an experiment is prophesy;
experimentation is itself a production of the
phenomena. 29

In this way, judgements about the kind of experiment to


be undertaken become the central philosophical issue:

there will always be some kind of result, whatever one
does to nature, including to ourselves, and the modern
period has enough examples of this that do not bear
thinking about. Any attempt on our part to ground the
activity of experimentation can only lie in an ethical and
hermeneutic approach to what we disclose in nature, not
in the idea that we are contributing to the ‘final truth’ . In
the Romantic sense it is incoherent to think that a
complete account of the world in terms of scientific laws
is absolute, unless self-consciousness, the fact of
intelligibility, and our ethical sense itself could be
explained in a completely law-bound manner. It is this
aspect of Romantic philosophy which is most relevant to
the question of technology, not a vague aversion to the
damage done by science to an innocent nature.

The problem that forms the core of Romantic
philosophy is that any explanation, including of
subjectivity, must be of the kind used to explain a
phenomenon of nature like any other, but the whole point
of transcendental philosophy, which Fichte first saw
more clearly than anyone, is that the condition of
possibility of explaining natural phenomena via laws
cannot itself be of the same status as those phenomena,
on pain of another regress. To this extent we cannot
finally draw a line between nature and consciousness and
are compelled to confront the consequences of
consciousness’s inability finally to ground itself. We
must at the same time, though, take account of self-


consciousness’s undeniable familiarity with itself. If
consciousness is to explain itself qua object it must
already be unquestionably familiar in a non-objectified
manner with what is to be explained: otherwise it would
have no criterion for knowing that it had explained itself
The prior condition of my seeing myself as myself, rather
than as a random object, in a mirror is that I must be
already familiar with myself in some way which does
not require a mirror. This awareness is of a different order
from the awareness I have of the ‘not-I’, the appearing
deterministic object world of natural science. As N ovalis
puts it in the Fichte Studies: ‘Can I look for a schema [as
that which enables one to identify objects] for myself, if
I am that which schematizes?’30 Consciousness’s
inherent familiarity with itself is, furthermore, even
different from the cognitive objectified account it may
give of itself once it has questioned its own status in
transcendental reflection upon the conditions of
possibility of knowledge. Novalis maintains in this
respect that ‘What reflection finds, seems already to be
there’ .31 The already existing I must in fact be prior to the
ways the I of reflection attempts to describe itself in
transcendental philosophy, and can never be
scientifically determined by something which actually
depends upon it as its – thus never ultimately knowable
– ground.

N ovalis’ s at first sight abstract metaphysical point
connects much more directly to technology than it might
seem. The reference to ‘schematism’ concerns that aspect

of Kant which Heidegger will later question in analogous
ways with respect to the Romantics. Kant understood
‘schematism’ to be the aspect of judgement which
overcame the regress that would ensue if we tried to
ground the rules we use for subsuming images under
concepts, which would lead to the need for rules for the
application of rules, and so on. The ‘talent’ required to
prevent a regress of rules for rules in judgement depends
on a ‘hidden art in the depths of the human soul’ which
will, Kant maintains, probably remain inaccessible to us:

this ‘art’ is ‘schematism’. In line with dominant
assumptions abut modern technology Novalis says ofthe
schema, in 1795-96, that it ‘renders the treatment of a
single case more easy, because it teaches me to apply the
universal laws of the class (Gattung) of these cases and
thus spares me the effort of again looking for the laws of
this case’ .32 Schematism thus plays a central role in the
constitution of a world of determinate truth and of objects
which can be instrumentally manipulated. As such,
though, the ground of truth becomes a ‘technique’, but
not in the sense of something governed by rules, but
precisely in terms of a ‘techne’, an ‘art’ which requires
judgement on the part of the subject. This argument
obviously goes very deep, and it links the Romantic
concern with aesthetics to vital questions of
epistemology and truth. Much of the debate concerning
technology is in these terms a debate over what ‘techne’

actually signifies. The Romantics evidently do not regard
it as something algorithmic, and it therefore involves
aspects of ourselves which do not fit a ‘Spinozist’ model.

Romantic nature
The way the technologies develop this point again leads
in directions which are now central to contemporary
philosophy. Schelling and Schleiermacher claim that
language itself also depends upon the ‘art’ of
schematism. Even though, for Schleiermacher, there is a
‘grammatical’, rule-bound side to interpreting language,
there is also a ‘technical’ non-rule-bound side, without
which language could never be understood in real-life
situations. In the System of Transcendental Idealism of
1800, written while he was in close contact with Schlegel
and Novalis in Jena, Schelling says of Kant’s schema:

‘The schema … is not an idea [Vorstel!ung] that is
determined on all sides, but an intuition of the rule
according to which a particular object can be produced. ’33
As intuition of a rule, thus as immediate access to the
rule, the schema cannot itself be determined: for that, as
Kant has shown, intuitions need logically prior concepts.

There can, then, be no further grounding of this intuition
beyond the fact of its own functioning, which is what
renders the world intelligible. Schelling further maintains

that ‘From this necessity of schematism we can infer that
the whole mechanism of language must rest upon
schematism’ :34 for words to have iterable significances indeed, for words to mean anything at all – they must
depend upon some ground of identity which both makes
manifest and, so to speak, ‘holds steady’ what is meant,
while the actual data of empirical reality continually
change. This ground is, though, not a stable quantity. If
one is to learn the rules for using language appropriately,
one must, in order to be able to understand the same
utterance in differing situations, understand in a manner
which cannot be further grounded by the notional rules
of language: otherwise a regress of rules for rules ensues
once again.35 The proximity to Davidson should again be
apparent here: the Romantic idea is exactly what leads
Davidson to say that language has no ‘rules in any strict
sense’ .36
The Romantic understanding of schematism makes it
clear that this ‘hidden art’ plays an important role in
understanding our relationship to the natural world. The
vital question for a contemporary approach to
Romanticism is to understand this role appropriately.

Schematism essentially plays the role of the ‘asstructure’ in Heidegger, the structure which makes a
world intelligible. The danger often seen in schematism
is that it reduces the inherent diversity of intuitions to
repeatable forms of identity, so that, as Nietzsche put it,
‘overlooking the individual and the real gi~es us the
concept’ .37 Within the story common in various forms to
Nietzsche, Horkheimer and Adorno, and Heidegger, the
danger lies in the way modern science is seen as the
product of the subject’s attempt to dominate its Other.

What, though, is the Other, if, as we have seen, the divide
between nature and consciousness cannot finally be
determined? This does not mean that the issue of
schematism should be dismissed, but rather than we need
to rethink the position from which criticism of
technology can be advanced.

One Romantic way of considering the dangers of
schematism’s reduction of difference was to suggest that
if the schema is the key function of the productive
imagination, which Fichte makes the centre of a
philosophy grounded in practical reason, then nature
becomes reduced to what concepts generated via
schematism and the goals demanded by the subject’s
attempt to bring nature in line with practical reason
demand. In 1806 Schelling therefore contends against

in the last analysis what is the essence of his whole
opinion of nature? It is this: that nature should be
used … and that it is there for nothing more than to
be used; his principle, according to which he looks
at nature, is the economic teleological principle. 38


The underlying structure of this argument reappears in
the reflections upon alienation and commodity structure
in the early Marx (who knew Schelling’s work), which
are central for the early Lukacs and the Frankfurt School.

Implicit in the idea of schematism and its relationship to
the principle of equivalence is the idea ofthe domination
of the Other in the name of the Same, which has become
a central issue in the analysis of the pathologies of
capitalist modernity. A philosophically effective
approach to this issue must adequately confront the
complexities it involves.

We cannot simply step outside of the forms of
equi valence entailed in schematism, and the attempt to
suggest we could has done great damage to many
progressive approaches to these problems. As such we
must be sure what is really at issue here. A crucial short
circuit in many of the arguments concerning technology
and subjectivity has been to claim the two are necessarily
linked, in the manner Schelling fairly justifiably suggests
is the case in Fichte. In this perspective the sort of identity
suggested by J acobi between Spinozism and Fichte,
which is based upon an internally grounded system of
relationships, is supposed to take up all the philosophical
space for reflection upon technology. This is one way of
understanding the impetus behind the later Heidegger’s
attempt to overcome metaphysics
as the
‘subjectification’ of being. The attempt to criticize the
Heideggerian position is these days too often regarded as
requiring the adoption of an invalidly external
‘metaphysical’ position in defence of subjectivity or
humanism, which repeats the domination on the part of
the subject in another form. Schnadelbach has shown
how mistaken this assumption is, in an essentially
Romantic argument against the kind of scientism which
I cited at the beginning of this paper, which is implicit in
many of the current arguments about artificial
intelligence and artificial life. Schnadelbach’s contention
is, however, also valid against that side of Heideggerian
thought which in fact sees the subject’s relationship to
itself as constituted solely in terms of its relation to the

Philosophy is the articulation in thought of our
theoretical and practical relation to ourselves,
which is admittedly located in a relationship to the
world, but does not disclose itself – as scientism
maintains – from the relation to the world: Instead
we can, on the contrary, rather illuminate and
understand our relations to the world via our
relation to ourselves. 39
This relation is developed not just via what we think we
determinately know of ourselves, but also in aesthetic


modes of exploring our self-understanding. The key
Romantic insight lies in the refusal to reduce our relation
to ourselves to what we can determinately know about
our relation to the world, and it is this position which a
revival of Romantic philosophy needs to develop.

Such a position is exemplified in the following long
passage from Novalis’s Remarks on Physics and
Medicine of 1799-1800. The text is in line with
Schnade1bach’s contentions, thereby suggesting that we
may only now be arriving at the truth of the Romantic
position. In the text Novalis addresses some of the
structures I have tried to elucidate in this paper, as well
as giving the lie to the idea of Romanticism as a
movement concerned with mystical access to the
Absolute. He does so with the kind of literary talent
which all the efforts to reduce our relationship to
ourselves to objectifiable relations to the world will never
come even near to understanding. He begins with the
Spinozist world, but then takes us in very different

253. No force, no phenomenon can be explained
by itself in nature – e.g. gravity. All forces are what
they are – by division into chains. One is what the
other is – only modified by its location, by its

254. Universal assertions are not valid in natural
science. They must be presented practically,
technically, really – developing step by step constructing like the description of a technical

Contemporary physics is very meagre – Our
physics speaks only of the universal elements of
nature – of the universal effective concepts – or the
forces of nature – it is a real [largely in the sense of
‘materialist’] metaphysics or logic. Plants,
animals, stars, human beings are already synthetic
products of nature – higher natures. Nature is a
church of infinite natures. Everything is limited,
even human science is to be determined according
to time and place. Humankind cannot reach any
higher than to see what knowledge is appropriate
for its particular stage – for the duration and
constitution of its life – and to see that it does not
pathologically favour the drive for knowledge that it leaves it in harmony with its other powers
and dispositions.

Humankind is not destined for science alone a human being must be human – destined to be a
human being – a tendency to universality is
indispensable for the real scholar. But human
beings must never, like a fantasist, seek something

indeterminate – a child of fantasy – an ideal- They
should only proceed from determinate task to
determinate task. An unknown beloved admittedly
has a magical charm. Striving for the Unknownthe indeterminate is extremely dangerous and
disadvantageous. One cannot produce revelations
by force.

The truly idealist path of the physicist is not to
explain the composite, the compound, by the
simple, the disintegrated, but the other way round.

A state [Staat] will never emerge out of a state of
nature – but a state of nature can result from a
state[.] Nature arose by degeneration [Ausartung,
with the sense of moving out of a genus, an ‘Art’,
thereby into new genera, not in the now usual sense
of degeneration]. Gravity is explained by
sensibility – sensibility is not explained by gravity
– electricity etc. The emergence of gravity is
explained by thoughts. The first chapter in physics
belongs to the realm of mind. Nature cannot stand
still, it can only progressively – be declared to be

Eventually there is not supposed to be any
nature any more – It is gradually to make the
transition into a world of the mind. If the
unchangeable laws of nature are not to be an
illusion – not to be extremely unnatural.

Everything works according to laws, nothing
works according to laws.

A law is a simple relationship which it is easy
to overlook.

We look for laws for the sake of convenience.

Has nature a determinate will- or no will at all? I
believe both – It is all things to all men. 40
I shall make only a few comments on a highly complex
text: to do otherwise here would be to lose some of the
point of any attempt to initiate a new exploration of
Romantic modes of thought. I do not mean to suggest by
this that one should take the passage’s meaning to be
merely indeterminate, but rather that it contains
possibilities which can only emerge by putting it into
context with other Romantic texts, as well as with more
recent attempts to confront these questions.

Novalis’s text is a piece of communicative action,
which both opens up issues and undermines the attempt
to reduce what is said to a series of propositions: in this
sense it exemplifies that the Romantics meant by
‘Poesie’, ‘literature’, in the sense of art which has truth
potential. The vital factor lies in the way the text puts the
meaningfulness of the world and ourselves first, without
this leading to advocacy of scientific domination over
the Other of nature. However much the methods of

modern science may rely on breaking the world down
into determinable elements, the point of science in the
Romantic view is to help make more sense of the world
as something already intelligible, hence natural science’s
role as only one of the forms of our self-understanding.

Clearly one can reduce the world to the elements of its
intelligibility, or to computational functions, and this
plays a necessary role in the modern scientifically
oriented relation to the world – including in dangerous
and destructive manners. But this gives no way of
approaching the world’s very intelligibility itself, an
intelligibility which is in the Romantic view made most
apparent in aesthetic products and aesthetic activity, but
which is also revealed in the need for the holistic
approach to science suggested by Novalis.

Schlegel illuminates the essential Romantic approach
in the following wonderful fragment: ‘If the chemist
thinks a thing is not a whole because he can dissect it,
that is just the same as what bad critics do to literature Didn’t the world emerge from slime?,41 It is in aesthetic
products which can be associated with the ideas of early
Romanticism – such, most notably, as the music of
Beethoven and Schubert – that we can perhaps still
glimpse what can be produced by a combination of the
technical ‘Spinozist’ demands of modern rationality with
a Romantic sense of the other meanings of ‘techne’.

Romanticism in this sense can now be understood as a
vital philosophical reminder from the beginning of
modernity of how the separation of moral and aesthetic
achievement from technical advance is one of the
greatest threats to emerge from the new possibilities
offered by modernity.

The original version of this paper was given at the conference
on ‘Technology and Subjectivity’ at the University of
Middlesex, 29 October 1994.

1. This fate is perhaps most graphically and eclectically
explored in Thomas Mann’s Doktor Faustus.

2. Novalis, Novalis Band 2 Das philosophisch-theoretische
Werk, edited by Hans-Joachim Mahl, Munich and Vienna,
Hanser, 1978, p. 334. All translations from this and other
German texts are my own.

3. Peter Atkins, Times Higher, 30 September 1994, p. 19.

4. Richard Dawkins, ibid.

5. Even leaving apart the question as to whether all forms of
natural science have equivalent cognitive status.

6. Friedrich Schlegel, Kritische Schriften und Fragmente (6
vols), Vol. 2, Paderborn, Ferdinand Schoningh, 1988,

7. Friedrich Schlegel, Transcendentalphilosophie, edited by
Michael Elsasser, Hamburg, Meiner, 1991, p. 93.

8. See, e.g., my review of Ernst Tugendhat’s Philosophische
AuJsatze in the European Journal oJ Philosophy, vol. 2,
no. 3, December 1994. Tugendhat uses the division as a
way of characterizing the difference between analytical






and hermeneutic approaches to philosophy, which
usefully suggests how the philosophical and the scientific
issues can be seen as interrelated.

Walter Benjamin’s 1919 Ph.D. thesis, The Concept of Art
Critique in German Romanticism, was until very recently
one of the few texts to begin to appreciate what was really
at issue in Romantic philosophy in this respect.

Jacobi himself could not be termed a Romantic, but
without the responses to his arguments there would be no
Romantic philosophy in the sense intended here.

Friedrich Heinrich Jacobi, David Hume iiber den Glauben
oder Idealismus und Realismus ein Gesprach, Breslau,
Gottl. L6we, 1787, p. 224.

Schelling references are to Friedrich Wilhelm Joseph
Schelling’s Sammtliche Werke, edited by K. F. A.

Schelling, I Abtheilung Vols 1-10, 11 Abtheilung Bde. 14, Stuttgart, 1856-61: here 112, p. 53.

13. I have dealt with this issue in detail in ‘Rethinking the
History of the Subject: J acobi, Schelling and Heidegger’ ,
in Peter Dews and Simon Critchley, eds, Deconstructive
Subjectivities, New York, SUNY Press, 1995.

14. Whether these arguments are actually fair to Spinoza
cannot concern us here: the main point here is what he
was understood to mean, not a philosophically adequate
assessment of his arguments.

15. Henry E. Allison, Benedict de Spinoza: An Introduction,
New Haven and London, Yale University Press, 1987,

16. The proximity of aspects of these arguments to those of
the later Heidegger in particular is striking: I suspect the
link is via a certain reading of Kant and via the later
Schelling, who owed more to Jacobi than he usually
admitted. Heidegger, of course, borrowed much from
Schelling (see my Schelling and Modern European
Philosophy, London, Routledge, 1993; and my translation
of Schelling’s On the History of Modern Philosophy,
Cambridge, Cambridge University Press, 1994).

17. Donald Davidson, Inquiries into Truth and Interpretation,
Oxford, Oxford University Press, 1984, pp. 223, 267. The
parallels of Davidson’s holist approach to semantics with
key aspects of Romantic philosophy are substantial, and I
consider them in more detail in my forthcoming From
Romanticism to Critical Theory: A Philosophical
Introduction to German Hermeneutics and Literary
Theory, London, Routledge, 1996.

18. Immanuel Kant’s Critique of Pure Reason, trans. Norman
Kemp Smith, London and Basingstoke, Macmillan, 1933,
Bxx, p. 24; translation amended.

19. G. J. Fichte, Werke I, Berlin, de Gruyter, 1971, p. 95.

20. Friedrich Heinrich Jacobi, Jacobi an Fichte, Hamburg,
Friedrich Perthes, 1799, p. 14.

21. Ibid., p. 4.

22. Ibid., p. 3.

23. Times Literary Supplement, 9 September 1994, p. 4.

24. Hilary Putnam, Realism with a Human Face, Cambridge,
Mass. and London, Harvard University Press, 1990,

25. Herbert Schnadelbach, Zur Rehabilitierung des ‘animale
rationale’, Frankfurt am Main, Suhrkamp, 1992, p. 281.

26. The crassest version of such a grounding is, of course,
socio-biology, the nature of which will, I suspect, only
become finally transparent in the wake of the demise of
the politics which underlies it.

27. Novalis, p. 699.

28. Sch1egel, Kritische Schriften und Fragmente, Vol. 5,

29. Schelling, I, Vol. 3, p. 276.

30. Novalis, p. 162.

31. Ibid., p. 17.

32. Ibid., p. 160.

33. Schelling, I, Vol. 3, p. 508.

34. Ibid., p. 509.

35. The probably insoluble problems artificial intelligence has
with ‘background knowledge’ confirm the importance of
this approach.

36. Donald Davidson, Truth and Interpretation, edited by E.

LePore, Oxford, Blackwell, 1986, p. 446.

37. Friedrich Nietzsche, Samtliche Werke, Vol. 1, edited by
Colli and Montinari, Munich, Berlin and New York, dtv,
1980, p. 880.

38. Schelling, I, Vol. 7, p. 17.

39. Schniidelbach, p. 320.

40. Novalis, pp. 793-4.

41. Schlegel, Kritische Schriften und Fragmente, Vol. 5,

The Woburn Book Shop
10 Woburn Walk
London WC1 H OJL
0171 388 7278
Secondhand and antiquarian books
bought and sold
… cultural studies, social history, philosophy,
anthropology, Jewish studies, cinema …

Opening hours: Monday to Friday 11.00 – 7.00
Saturday 11.00 – 5.00


Download the PDFBuy the latest issue