Reply to Critics

Revolution and Evolution

Lafont rightly argues that my proposed dichotomy between the Kantian and the managerial mindset is far too simple to represent the difference of revolution and evolution appropriately, and the focus on legal revolutions therefore might lead to the impression of ‘over-generalizing exemplary cases’ (Habermas, 2014: 2). Indeed, the Kantian mindset is not exclusively revolutionary. On contrary, as well as acting as a driving force of revolutions, the Kantian mindset is also internal to a lot of managerial reformism. Kant himself is the best example for both options. He took the French Revolution as a Geschichtszeichen of moral progress never to be forgotten, and he did this even in spite of the great terror and the execution of the king, which both can and should not be justified or exculpated, as Kant repeatedly noted. However, the impeachment and conviction of the citizen Capet made the republican revolution in Europe factually irreversible, ⁹ and Kant’s writings after 1793 became increasingly radical, republican, cosmopolitan, egalitarian and (ethnically, ‘racially’ and culturally) inclusive (Kleingeld, 2013: 5, 7–8, 92–123). Moral theory has to pay a price once it becomes an existing concept, and this, not lastly, is a strong argument not to take Hegel’s existing concept affirmatively (justifying history’s slaughter bench as God’s progress through history) but to observe it from the negative side of its existence as contradiction to the ‘existing’ (Adorno). ¹⁰ However, Kant’s debate about moral progress does not only refer to ‘revolutionary events’ (Habermas, 2014: 2) because at the same time moral progress maintains a matter of Reform nach Prinzipien [reform according to principles], for example, in Prussia where Kant had close contact with the reformist wing of the Prussian bureaucracy (Langer, 1986). The alternative proposed by Habermas in 1968 – that is, radical reformism instead of post-Marxist revolutionary decisionism – is another example of the Kantian mindset operating from within the managerial mindset. ¹¹ This is true also the other way round. There is no successful revolution without a strong element of instrumental reason and managerially perfected organization. I have mentioned this Leninist element a couple of times, beginning with the Papal Revolution, and ending with the nearby reduction of the revolutionary mindset to instrumentalism by Leninists and Maoists that continued after the revolution, stabilizing party rule and the establishment of new ruling classes (see Brunkhorst, 2014a: 84, 110, 371, 388, 401).

However, as I was absorbed by the reconstruction of the four revolutions (and also the attempt to rehabilitate the concept of revolution), I did not think about it explicitly on a conceptual level. Only when Lafont made it explicit during the conference in Flensburg (see Thornhill, 2014: 1, note 1) did I suddenly realize the conceptual gap. I then immediately integrated her remarks in my critical assessment of Koskenniemi’s far too dualistic opposition of the managerial and the Kantian constitutional mindset, which I have criticized from the beginning (Brunkhorst, 2014a: 48–49). ¹² However, I will now use this opportunity to take a further step towards overcoming this dualism, and I will attempt to increase the internal complexity of my conceptual frame through Table 1, which makes the whole process at the outset less linear than it appears in the book (Habermas, 2014: 3–7). ¹³

OPEN IN VIEWER

Table 1.

The correlation between different mindsets and different processes of historical transformation.

Revolutionary/abrupt change	(I) French Revolution as Geschichtszeichen	(III) Leninism
Gradual change	(II) Reform nach Prinzipien/ radical reformism	(IV) Technocratic incrementalism
Evolutionary change mindset	Kantian	Managerial

Functional Imperatives and Class Struggle

The passage in my book (Brunkhorst, 2014a: 38) on immoral, brutal and gruesome evolutionary experiments quoted by Lafont (p. 9) is indeed open to misinterpretation. As far as actions and actors are concerned, there is no question that actors perform intentional actions. Whatever they think they are, their actions are actions insofar as there are effectively realized intentions motivating their actions, and, therefore, they are responsible for all results of effectively intended actions. All actions which are ‘not unintended, blind or random’ (p. 10) are due to the full accountability of the respective actor and groups of actors. Only in cases where the actor’s behaviour was actually blinded (e.g. drug-driven) and its effects unintended (e.g. trying to help someone, not knowing that a cure makes a complaint worse, or because of force majeure, etc.) are people (usually) ‘exculpated’, ‘mitigated’ or ‘immunized from normative evaluation’ (pp. 10, 11, 13). Nevertheless, all kinds of actions have at least some unintended causal (perlocutionary) effects, which can make effectively intended actions less disastrous than those that are effectively intended (if I steal my neighbour’s purse but the purse is empty), or they can be worse than effectively intended actions (if I steal my neighbour’s purse, and she/he gets a stroke when she/he discovers the larceny).

An extreme case is that of the crimes of the Nazis. Hans Mommsen has tried to explain this phenomenon as a functionally effective self-radicalization of the Nazi regime, resulting from the highly advanced de-formalization of law and the increasing fragmentation and erosion of organized state power that occurred throughout the 12 years that the National Socialist German Workers’ Party was in power. This was already analysed in the 1930s and early 1940s in Ernst Fränkel’s Doppelstaat and Franz Neumann’s Behemouth, and their analysis still stands up to most criticism, as far as I can see (see Stolleis, 1999). The fragmentation, decrease and erosion of state power was to a large degree an unintended effect of the way the Nazi regime tried to centralize, increase and stabilize its power with consistently criminal intent. The perverse evolutionary result of this was that the fragmentation of state power enabled and reinforced a previously unthinkable augmentation of atrocities. However, actors with criminal intent, who react in response to chaotic situations (e.g. unclear verbal instructions) with more criminal activities than originally planned, are not exculpated from any single crime which they have committed alone or cooperatively within a criminal organization. On contrary, persons who as leaders with criminal intent tried to increase their own state power clearly also have particular responsibility for the unintended effects of the evolutionary process of fragmentation of state power, and for its disastrous and gruesome consequences, solely on the ground that they approved them. ¹⁴ Therefore, moral and legal accountability seems completely compatible with the thesis that there is no single action or behaviour that is beyond evolution, simply because everything is evolution. As Richard Rorty once said, it is not the theories of Australian neo-physicalists (or Luhmanian functionalists, or the heirs of both, neuroscientists) that threaten our freedom – it is the secret police. It is not law and economics that threatens our freedom – it is the cultural hegemony of law and economics. This evidently does not mean that physicalist, functionalist or neuroscientific concepts of freedom do not deserve criticism, even for political reasons, in particular as far as they determine cultural hegemony.

At least, and here I agree with Lafont, every effectively intended actions, and especially effectively intended negations, are normatively important contributions to the variation pool of evolution. Nevertheless, even if they may have an indirect steering influence (which is what I presume), they cannot finally determine evolutionary outcomes, simply because there are always unintended effects when five billion people act simultaneously in world society. There are, as every family member knows, unintended effects and quickly increasing complexity and over-complexity even when five people act simultaneously in a family, at least as long as not all of their actions are agreed upon and coordinated, and as long as not all of their plans are successful. To this degree, I follow Luhmann. But I agree with Lafont’s view that moral intentions are important as contributions to the variation pool of evolution. To speak in evolutionary terms, tabolition of slavery was only possible because the moral protest against it engendered enough variation that made intentional (ideal and material interests) and unintended legal selection (functional imperatives) unavoidable, and it finally led to the re-stabilization of the prohibition of slavery and the establishment of anti-slavery as an evolutionary universal (see above remarks on Fine, 2014: 4). Therefore, and I am thankful that Lafont has emphasized this point. It is not only the case that unintended functional imperatives are mechanisms of selection (in periods of gradual as well as revolutionary change), but also effectively intended material and ideal class interests function as selective mechanisms. In fact, I distinguish these categories (Lafont, 2014: 10–11; Brunkhorst, 2014a: 295). Good examples are two columns written by Paul Krugman recently in the New York Times (21 March 2014, p. A23 and 24 March 2014, p. A19). In the first column, he describes a functional mechanism, now well established in America and Europe, which he calls the ‘timidity trap’. The timidity trap more or less unintentionally urges left-wing parties (sometimes even against their intentions) to be more and more reluctant when it comes to addressing classical left-wing topics (e.g. unemployment, social security, higher wages, government spending, disaggregation of banks, state regulation, taxes on assessments etc.). As a more or less unintended effect, the parties on the left are drifting further and further to the right, stabilizing functionally the global cultural hegemony of neo-liberal political economics. This is a clear case of unintended functional selection. However, Krugman ends his first column with the remark that, in the next column, we have to talk about effectively intended ‘class interests’. In questions of class interests, the blackmailing power of normatively neutralized functional economic imperatives is transformed into the power used by ruling and ruled social classes intentionally against one another and their respective political representatives, and this power acquires importance in normative terms in relation to general and universal interests (which are the topic of Krugman’s second column). The blackmailing power of functional imperatives then becomes a weapon for the intentional enforcement of class interests – either top-down through (for instance) investment strikes or bottom-up through (for instance) general strikes, both of which operate within a discursive context of normative justification (Brunkhorst, 2014a: 360, 364, 367–368, 379, 450). The two columns nicely show the intertwinement of intended actions based on class interests (the clashes of ideologies mentioned by Lafont) and unintended functional imperatives in the evolutionary process of social selection that finally led to the stabilization of neo-liberal political hegemony by the overwhelming power of the timidity trap and the class interest of global capital. However, even writing against it (as Krugman did) means to presuppose that the outcome of ordinary, non-revolutionary but reformist class struggle must not lead to a stabilization of the neo-liberal status quo but also can change it. It is, by the way, this kind of class struggle to which I refer with the keyword democratic class struggle (Brunkhorst, 2014a: 376–377, 390, 421, 459–461).

Normative Constraints

So far, I agree with Lafont, and I just needed to clarify a bit further what I wanted to say. However, I disagree with Lafont when it comes to the normative constraints that are established, not solely, but in a very significant manner, by the great legal revolutions (Kant’s Geschichtszeichen). I have stated repeatedly that normative constraints are different from the organic constraints of the natural evolution but have a similar function. Social evolution has replaced the functions of organic constraints by those of normative constraints, which ‘shape, give direction to, and act as conditions of possibility for the wider underlying form of society as a whole’ (Thornhill, 2014: 4). Specifically because they are normative, they are often violated, offended, infringed, contravened and impinged. There is no norm without violation. Nevertheless, at the same time normative constraints are not the outcome of clashes of ideology in the same way as ordinary constitutional, legal or moral norms are. This is the case because normative constraints have a ‘quasi-transcendental’ (Habermas) status, which limits evolutionary possibilities, opens and discloses new paths of evolution from within evolution and gives evolution a certain direction at a given moment. Usually, both sides involved in a clash of ideologies must operate (at least by lip service, cheap talk and hypocrisy) within existing normative constraints. A hundred years ago, at the beginning of the Egalitarian Revolution of the 20th century (see Brunkhorst, 2014a: 319–464), political leaders were proud to present themselves in public as leaders of empires, which had a formally acknowledged status in international law. A hundred years later, even those who are in fact leaders of (only) informal empires, present themselves in public as anti-imperialist freedom fighters (e.g. Bush, Putin). The fact that one of them could declare the Crimea or Iraq a formal Russian or US-American colony (or mandate) seems no longer possible. As Matthias Albert rightly has argued in his comment (Albert, 2014: 3–4), there is neither a valid ‘cultural script’ nor a valid legal category available for that to be possible. The same was the case in the argument between Metternich and Gentz over the constitutional status of France after the final defeat of Napoleon (see Brunkhorst, 2014a: 240–242). Metternich clearly saw that it was not possible to restore the old European status quo ante, as Gentz had suggested.

Normative constraints established by revolution always represent a kind of ratchet effect (I discuss this in each part one of the four sections of Chapter 3 on Legal Revolutions, but for a brief and excellent representation see Thornhill, 2014: 4–6). The ratchet effect of revolution (which results from a normative learning process that, as Habermas (2014: 2) rightly observes, is ‘not always presented as revolutionary event’) makes Gentz look as outdated after the Atlantic Revolution, as did the Norman Anonymus after The Papal Revolution, Alanus after the Protestant Revolution and, arguably, Carl Schmitt after the Egalitarian Revolution (see Brunkhorst, 2014a: 96–98, 151–152, 327, 347, 352, 355). Therefore, the great legal revolutions matter for ‘moral progress’ (Kant). Each revolution has ‘instituted universal legal norms behind which the subsequent evolutionary form of society could not easily and enduringly regress’ (Thornhill, 2014: 7).

To conclude, it was not impossible, once and for all time, that Gentz could have prevailed against Metternich in a way that then would have led to a new, and in this case regressive (and actually restorative), set of normative constraints, leading to an alteration in the direction of evolution, through gradual change. This can happen in evolution, and it happens. Normative unlearning is possible. ¹⁵ Even moral occurrences, which ‘are not forgotten’ (Kant), are sometimes forgotten in history. Even Hitler might have won the war, and democracy might have vanished from earth. However, because of the path-disclosing power of normative constraints, it is never possible to rule out such occurrences, but they remain very unlikely, and even more unlikely than the existence and maintenance of egalitarian democracy. Unfortunately, the atrocities caused by authoritarian regimes and failed states are not so unlikely, even if democracy finally prevails. Moreover, it is not unlikely that the democratic institutions after 30 years of effectively intended neo-liberal ‘reform’ will remain unchanged but that democracy (as far as it corresponds to its idea as defined above by Fine) will wither away. This is quite likely because of the ‘grotesque economic inequalities that prevail’ today, and not only in the United States (Scheffer, 2014). Here we see the Dialectic of Enlightenment.

The Concept of Legal Revolution

Echoing Marx’s Eighteenth Brumaire, Luhmann has argued that the great constitutional revolutions were nothing more than soapbox oratories, heart-warming songs and illusions of manageability (feierliche Erklärungen, Gesänge, Machbarkeitsillusionen), which accompanied the sober functional business of the structural coupling of law and politics. My book is nothing more than an attempt to falsify this thesis, and to establish the counter-thesis that normative constraints and normatively progressive ratchet effects are due, not solely, but in particular, to the great legal revolutions. They have ratchet effects that cannot be explained in exclusively functional terms. Therefore, I follow Berman’s thesis that the great legal revolutions were both legal revolutions in law and at the same time legal revolutions of society, changing its ‘collective conscience’ in the direction of the ‘modern tendency of moral conscience’ (Durkheim).

Thornhill remains sceptical. There are, he argues in agreement with me, legal revolutions in law that might have some ratchet effects for all society. But these revolutions are not at the same time legal revolutions of society. There might have been ‘rather diffuse bundles of related revolution’ (p. 13) in law and in other functional spheres of society, and ‘law had obvious utility in each of these’ (p. 14) – but there is no such thing as one revolution of society expressed in slogans such as Liberté, Egalité, Fraternité. The emancipatory unity of revolution only exists in books. It makes heart-warming songs and conventionalized normative stories of hard-won revolutionary constitutional advances. But these stories are just telling a myth. They are ‘a projection of normative analysis’ (Thornhill, 2011: 10). Therefore, it remains unclear why in Critical Theory ‘law is a privileged bearer of human freedoms, and as such, it is uniquely able to transcribe diverse human freedoms into an emancipatory grammar, which can be applied across, and made constitutive for, all parts of society’ (p. 13). My claim, applied here to the Papal Revolution, that functional differentiation has been triggered by unintended side effects of that revolution (growth of legal complexity) seems to contradict my other thesis that ‘law implicitly conducts learning processes, and resolves antinomies within these processes, for all society’ (p. 13). Constitutional law, Thornhill argues, simply becomes overburdened once it is taken as expression of a unifying idea of freedom for all society. To explain my overburdening of law, Thornhill presumes (close to the Marxist critique of ideology) that I ‘quietly’ ‘rest’ my theory on the ‘quasi-anthropological’, old European ‘notion that law has primacy in the normative apparatus of society because it is the societal medium that is best equipped to generalize human essence and human species-life’ (Thornhill, 2014: 13).

However, as Thornhill himself acknowledges that the concepts of legitimization crisis and normative learning present ‘problems of the normative self-construction of society’ that ‘are categorically not expressions of mono-original or essential-anthropological demands for ideal self-realization’ (p. 11), I do not rest my argument on a philosophical anthropology of law as generalized human essence. On the contrary, I explain the highly unlikely combination of technically advanced law (that in the classical Roman Empire was just a law for coordinating the economic interests of the ruling classes) with a religious world view that contains a comprehensive and utopian idea of egalitarian emancipation as normative self-construction of a specific society that is modern. Only after the societal self-construction of law as emancipatory law, could the abstract idea of emancipation as human essence (that originates historically from the metaphysical and religious world views of Axial Age and was more or less the same in all world views that emerged between Old Athena and Xian) become an existing contradiction that is internal to modern society and no longer in need of any reference to human essence and human species life.

The really crucial difference between myself and Thornhill is rather different. Thornhill agrees that each of the four legal revolutions has ‘instituted universal legal norms behind which the subsequent evolutionary form of society could not easily and enduringly regress’ (Thornhill, 2014, forthcoming: p. 7). However, he restricts this agreement on normative constraints and ratchet effects exclusively to ‘functional norms that underpin modern power, and that permit societies recursively to apply and reproduce their power’ (Thornhill, 2011: 374). I agree that the stabilization of normative constraints can and must be explained functionally in this way. But only stabilization can be explained functionally. For example, the pious monks and church leaders who reconstructed the Corpus Juris Romani, established the Corpus Juris Canonici, initiated a legal discourse, erected a system of courts, founded law schools and universities wanted to establish just law in the service of salvation and reform, but nobody wanted (or even could expect and foresee) the functional differentiation of a self-referentially closed legal system that was simply an unintended side effect of intended legal organization, discourse and professionalization.

However, this does not mean that the legitimating and socially integrative meaning of the intended and discursively grounded establishment of emancipatory law is exhausted by its unintended functional effects. On contrary, the legitimating and socially integrative meaning of law in service of salvation and reform must be realized in a way that is independent of the functions that law fulfils for the stabilization of society. Therefore, the existence of mere functional norms alone cannot explain how the first formation of law that is emancipatory could emerge in Europe between the 10th and the 13th centuries. For example, the finding of a copy of Justinian’s Corpus Juris Romani in a library in Pisa in 1050 was not an accident as systems theory is pleased to presume because it fits in so nicely with blind and contingent evolutionary processes. If anything it was the result of intentional, intense and systematic search, embedded in growing legal scholarship, ideas for radical reform, unresolvable social conflicts and powerful class struggles. Moreover, functionalism cannot explain why the emancipatory potential of law was not swallowed by a functionally differentiated legal system that was designed (by the imperatives of evolutionary adaptation) to minimize ‘unpredictable resistance’ (Thornhill, 2011: 374). ¹⁶ Hence, functionalism cannot explain why emancipatory constitutional law strikes back, repeatedly and sometimes revolutionarily.

Maintenance of the emancipatory potential of law and its ability to strike back cannot be explained alone ‘as produced through society’s specific responses to its own inner functional exigencies’ (Thornhill, 2014: 15). First, law that is modern – and Thornhill does not dispute this – cannot be reduced to negative freedom, purposive rationality and the instrumental use of law (see Larmore, 2014: 150–151). Second, in spite of the heterarchical decentration of the political–legal–sacral complex of hierarchically stratified societies through the functional differentiation of law, politics and religion, at least the political and the legal system cannot cut their links with the general public sphere (Brunkhorst, 2014a: 49–51, 79–81, 433–434). It is this – and here I disagree with Thornhill and concur with Habermas – that distinguishes law and politics from ‘subsystems such as economy, family, education, health’ and ‘religion,’ which are no longer indispensable (if not superfluous) for political legitimization and social integration, but not for fulfilling special societal functions (Habermas, 2014: 4–5, 16).

Therefore, the Kantian mindset ‘cannot be eliminated’ from public politics and public law (that covers its internal differentiation in private and public law reflexively) – except (and this is just an empirical hypothesis) at the price of a legitimization crisis (Somek, 2013: 8). In the same way that the political system needs the ‘living power of the people’ (Arendt), the legal system needs the regenerative feedback from the general and diffuse public sphere. At the least, constitutional law is not just a ‘lawyer’s contract’ but a ‘layman’s document’ (Roosevelt). Legal norms that are a self-construction of modern society, I want to repeat here, cannot relinquish their internal connection with the colloquial language and the moral self-understanding of their addressees; the normative closure of the legal system is not only in need of cognitive and systemic adaptation to preserve itself (as in Luhmann’s theory), but also the continuation of normative learning. All law that is public is opened not only cognitively to its environment but also normatively to the general and diffuse public sphere (Brunkhorst, 2014a: 50).

Politics and law are functional systems, but as public endeavours they remain internally linked to the general and diffuse public sphere, and to one another. Therefore, ‘the legal system cannot differentiate itself from politics in the same way’ as the economy, the family and religion ‘were able to do’ (Habermas, 2014: 5, my emphasis). It is only because of its still indispensable public character that the decentred political system can use positive law to ‘influence, even shape and steer’ all social subsystems ‘in an indirect fashion, without adversely affecting the internal logic of those self-regulating systems’, which themselves all must be ‘legally constituted’ (Habermas, 2014: 5). However, specifically because law and politics are at the same time, in a ‘dialectical mode’ (Habermas, 2014: 5), functional systems in the sense of Albert, Luhmann and Thornhill, they can have normatively disastrous and colonizing effects on the social life world of the people, and they can cause crises of motivation and legitimization. They can even trigger revolutions as in the case of the Protestant Revolution, which strongly articulated its total alienation from canon law and opposed violently the oppressive concretization of divine and natural law in the existing system of private and public law (Brunkhorst, 2014a: 178–188). The same was true with the negative experiences of the people and European intellectuals on the eve of the Atlantic Revolution. At this time, the customs and conscription officers of the emergent modern state treated ‘free human beings like mechanical works’. And, as it is stated in the Oldest Systematic Programme of German Idealism, ‘it should not do that; therefore it should cease.’

However, I take this statement from the end of the 18th century (in a similar way to the Twelve Articles of 1525) not as an expression of human essence vis-à-vis the absolutist monster but as an expression of the existing contradiction that is internal to modern constitutional law. Therefore, I take the concurrency of systems formation and structural coupling of law and politics with the growing dependency of law and politics on an untameable and anarchic public sphere as a ‘further justification’ that up to the present modern law ‘promoted freedoms that were […] generally meaningful in all spheres of social practice’ (Thornhill, 2014: 14). If viewed solely with regard to their functional outcomes, the Papal Revolution, the Protestant Revolution, the Atlantic Revolution and the Egalitarian Revolutions might easily each be split into five, six or seven revolutions – in law, education, science, religion, media and so on. However, from the actor’s perspective of egalitarian freedom, the functionally differentiated results of the great revolutions are embodiments of the same idea of egalitarian freedom, which is modern – at least once they use it in a practice that strikes back. This is so not only in political practice but also retrospectively. Both future-oriented practice and retrospective memory are as important and indispensable for the normative self-construction of society as the mere cognitive ‘learning processes that occur within the law and the law alone’ (Thornhill, 2014: 14). If we as social actors refer, for example, to the abolition of slavery in the 19th century, to the right for women in the 12th century to say no in case of marriage, to the legalization of divorce since the 16th century, to the legalization of homosexual relations and the implementation of gender-invariant rights to marriage at the end of the 20th century and in the early 21st century, to the enforcement of equal voting rights for all adult men and women in the course of the 19th and 20th centuries and to educational affirmative action in the second half of the 20th century, do we not refer, and must we not refer to different sides of the same idea of ‘socially universal freedom’ (p. 14), at least as long as we are involved in public discourse? (Larmore, 2014: 151). I think that this is not – in ‘critical idealist fashion’ – a comprehension of freedoms ‘as species-freedoms’ (p. 14) but still a ‘self-construction of society’ (p. 4), and it is a critical idealism not in the idealistic manner of Plato but in the pragmatic manner of John Dewey.

Overcoming the Schism

Thornhill and Albert reconstruct the sociological argument of my book in a very illuminating way. On the one hand, they show that most of my argument is aligned to functionalist sociology, or systems theory. On the other hand, Thornhill in particular makes the points very clear where I try to implement the explosive charges of Critical Theory.

It is, as Thornhill rightly says, my interest to ‘to rectify the damage and loss of explanatory capital caused by’ the ‘deep conceptual schism’ between Frankfurt and Bielefeld in particular since the 1990s (Thornhill, 2014: 1). For this purpose, I take (I) from Adorno the dialectical method of determinate negation; (II) from Neumann, Mauss and Habermas the basic idea of an emancipatory potential that is, ‘however inchoately’, embodied in ‘sociologically formative legal realities’ (pp. 1–2). Therefore, I conceive, in Thornhill’s fitting formulation ‘constitutional norms as socially generated but normatively essential principles’ (p. 7). This clearly goes beyond Kant and Adorno, and so does (III) the follow-up empirical thesis, adopting some arguments of Habermas from the 1970s, that law is the bearer of ‘communicative norm-rationality’ that is internal to social evolution alone. Methodologically, Critical Theory must ‘give authority to its normative claims by examining the legal norms of freedom in historical elaborated contexts’ and ‘offer a sociological account of the processes through which’ constitutional norms as ‘discursively mediated agreements’ ‘come into existence’ (pp. 3–4). Therefore, (IV) my general explanatory goal is ‘to offer a strictly inner-sociological and causally differentiated – that is […] sociological – explanation for the production of law’ (p. 11).

For Thornhill, the very idea behind my work is that law ‘in the sense of Kant, Rawls, Habermas, or Dworkin’ has a ‘philosophical’ (or ‘legal-philosophical’) side that conceives it as ‘a privileged bearer of overarching norms’ that are relevant for the ‘legitimization’ of ‘society as a whole’ (Thornhill, 2014: 2). However, the philosophical dimension that makes law a ‘medium of innerworldly transcendence’, is so completely internal to the ‘factual sociological production of law’ that – in ‘legal-sociological reflection’ – ‘the normative origins of the law can never be strictly separated from the conditions of its emergence and application’ (pp. 3–4).

Here I would say: yes and no.

Yes, because factually – from the sociological point of view (p. 3) – laws legitimizing and transcending, and, hence, emancipatory normativity, can never be separated from the specific historical conditions of its emergence.

No, because the evolutionary emergence of laws legitimizing and transcending normativity is, at an evolutionary level, co-original with the emergence of discourses that are – from a philosophical point of view – concerned with the validity and justification of this normativity. These discourses are also part of social reality, but they are a part that distinguishes itself from the emergence and application of law, and the subject of this distinction is not the system but the people who participate in the discourse as a social group, which in addition is in need of functional stabilization (and only in this respect can we say that ‘discourses are systems’, as Luhmann argued in his first debate with Habermas in the early 1970s). As factual discourses, and there are no others, the discourses on the validity and justification of law are – as everybody knows from every scientific or philosophical debate – always already intertwined with power games and many other overlapping language games that are not just motivated by the forceless force of the better argument alone. They are often deeply entrenched in ideological formations, hegemonic and counter-hegemonic cultures, microphysics of power and so on. Nevertheless, sometimes they can become the very medium of ideological critique, of the unmasking of hegemony, and of emancipation from oppressive power, and there is no other medium to do that. Moreover, the unity of the dialogically contradicting extremes of discursive exercises (ideology production vs. critique of ideology) is that both parties (from within social evolution) must presuppose in the performance of their discourse that ‘reason cannot be reduced to power, nor power to reason’, and that ‘good reasons do not make bad arguments’ (Lafont, 2014: 2). They all know that they have to give reasons and not to reach for their guns. There is (nearly) nobody who cannot distinguish the gun shooting game from the give and take reasons game. Moreover, the rationality of the forceless force of the better argument, pooled in discourses, can be transformed into a ‘rational motivation’ (Habermas) to strike back, usually with words, but sometimes also with paving stones. This is the way that the common man of 1525 used the then hegemonic legal discourse for emancipatory purposes (Brunkhorst, 2014a: 180–186).

Critical Idealism

Thornhill accepts the functional side of my normative universalism (I) that normative constraints (as functional norms) are ‘expressions of universal patterns of evolution’, which ‘impact on all aspects of society, across the dividing lines between national states’ (p. 10). This directly leads to the co-evolution thesis to which Albert, Thornhill and I all subscribe (see the section on ‘the co-evolution thesis on national and cosmopolitan statehood’). However, from a purely functionalist point of view, Thornhill, to be consistent, is obliged to reject the basic idea of my book that (II) normative constraints are ‘constructed as articulations of an encompassing evolutionary orientation towards general freedom’ (p. 10) as ‘critical idealism’ and ‘implicit essentialism’ (p. 12). Therefore, the project of overcoming the schism between Bielefeld and Frankfurt must finally end with the unresolved antinomy between a ‘theoretical construction of society as shaped – lastly – by evolution’ and a ‘theoretical construction of society as shaped – lastly – by normative teleology’ (p. 12).

However, I would not say teleology, and I would not say that the orientation towards general freedom is encompassing. In contrast, I try to avoid any Hegelian teleology of freedom. Therefore, I would prefer to use Dieter Groh’s formulation of a trace of reason in a history of many traces. Thornhill himself offers this as an alternative (p. 9). A trace of reason first appears accidentally, like a trace in the sand of evolution, it proceeds in a certain direction, it is forced back by a bundle of functional imperatives, by environmental problems, by powerful and urgent class interests, it finds its way back to its original trace, discovers a new one and finally disappears in the sand of history – or not. Hegel already came close to that insight in talking about the modern state of reason as a vanishing point in the ocean of history. ¹⁷ Such a trace, if it exists, exists sociologically, and if it exists, then adaptive cognitive learning processes of learning machines can never give sufficient explanations of its turns and sequences. For sufficient explanation of the small traces of reason in the constitutional evolution, a recourse to normative learning processes of social groups seems unavoidable. These groups are nearly absent in Thornhill’s great book on Sociology of Constitutions. But they exist in history, and their class struggles are not only about material but also about ideal interests, ‘clashes of ideologies’ (Lafont), and principles of justice which often are established against functional imperatives and selective mechanisms.

Normative learning is learning that social actors themselves interpret as a kind of moral progress. Their own interpretation of evolution is part and parcel of social evolution. They matter for understanding social reality just because ordinary people in their everyday life are not only layman constitutional lawyers (Roosevelt) but also layman sociologists (Alvin Gouldner) (see Beck, 1972). Therefore, the actors’ understanding of themselves and their society resides on the same epistemic level as the theories of the professional sociological observer. Normative learning does not presuppose, as Thornhill assumes, ‘a unified source of reflexive norm-rational agency, which, under certain circumstances, demands freedom’, and ‘stable norm-generative resources as the most essential substructure of society’, and its ‘innermost residual substance’ (pp. 11–12). This is the wrong, old European model. There is not, on the first level of human nature (state of nature), a competent subject in a state of latency whose competencies then, on the second level of societal conditions of manifestation, are performed if they meet these conditions (state of society). On contrary, conversely, old Europe must be turned upside down. Then, we will see that it is societal communication, mediated through negative speech acts, that engenders a socialization process of the internalization of authority that at once stabilizes authority and enables autonomy and the social formation of a sense of injustice. Depending on the contingent social conditions of socialisation, individual human beings learn to strike back, from within the frame of normative self-construction of the respective societal formation [Gesellschaftsformation] (see Habermas, 1981: 52ff.; 1988: 187ff.).

Framing Social Evolution

Habermas’s main objection resembles strongly that of Thornhill, but with the opposite evaluation. Both argue that I finally do not resolve the antinomy between systems theory of evolution and history of revolutions. Whereas Thornhill rejects my theoretical construction because it is shaped – lastly – by ‘normative teleology’ (p. 12), Habermas argues that my theoretical construction is shaped – lastly – by the ‘abstractions’ of systemic evolution (pp. 18–19).

I agree with Habermas that societies are ‘systemically stabilized networks of action and communication among members of socially integrated groups’, which are conflict and cooperate (p. 18), and I use a similar idea from the Theory of Communicative Action throughout my book. Even if I do not make much use of it, I also agree with the (Durkheimian) three-component model of the social life world (culture, person and society) as a starting point of the evolutionary differentiation between the societal component (which then differentiates in functional systems and their environment) and the life world (which then differentiates internally, following the differentiation of discourses and validity claims) (Habermas, 2014: 19–20; 1981: vol. 2).

If I see it correctly, there are two main differences between systems theory (Luhmann) and Critical Theory (Habermas), which both make sense within the complex process of evolutionary differentiation. The problem of theory construction is how to coordinate them appropriately.

First, we can construct an ideal point of departure on a journey that leads both Luhmann and Habermas, albeit in different directions, out of old Europe and its metaphysical thinking. An ideal point of departure is Husserl’s Fifth Cartesian Meditation (1928). Owing to Husserl’s supposed failure in this work, Luhmann decided to abandon truth and intersubjectivity, but to retain the abstract structure of the self-reflexive subject, and apply it to any kind of self-organized Turing machine, be it organisms, computers or social systems. To avoid any association with humanism, Luhmann replaced the self-reflexive subject with the self-referentially closed system. Habermas drew the opposite conclusion. He abandoned the subject and retains intersubjectivity and truth and applies them to the internal differentiation of the social life world. ¹⁸ He replaced the subject (in a formulation from Faktizität und Geltung [translated as Between Facts and Norms] that seems to come close to Luhmann but does not) with subjectless circulations of communication. From this starting point, both Habermas and Luhmann argue that society reproduces itself through communication. In this thesis, both deny the one-dimensional information model of communication, and they follow instead the three-dimensional model proposed by Karl Bühler (and others, such as Austin and Searle). However, whereas Luhmann conceives communicating actors (in a Hobbesian manner) as reflexive systems of conscience (psychic/personal systems) which are only significant for one another as bearers of high and risk-filled environmental complexity, Habermas conceives communicative actors as human individuals who are motivated by truth- and validity claims (that are internal to their speech acts) to reach an understanding. ¹⁹

From these conceptual premises follows the second step. From the point of view of the systemic observer, society appears as an endless flow of communication (Habermas, 2014: 19). As Albert rightly emphasizes, this communication is not in need of any form of social or systemic integration; this is simply because, for Luhmann, such integration is not possible (Habermas, 2014: 8–9, 20). Therefore, from the systemic perspective, we can only observe if and how society solves the quasi-biological problem of keeping communication running, and we can only explain this through reference to causal and functional mechanisms. Habermas admits that systemic self-organization through causal and functional mechanisms alone is possible. But it is only possible for functional systems, which, in evolutionary terms, are relatively young. Functional self-organization is possible in particular (and in accordance with Marx) for the capitalist (or any other) system of modern economy. However, Habermas argues – from the internal point of view of the social actor – that the life world of social groups and classes, and the public sphere of politics and law, are results of social integration via intersubjective understanding that is bound to truth claims. Therefore, he concludes, quite rightly, that a post-truth democracy would no longer be democratic.

Third, there is sufficient empirical evidence to support both perspectives, and both are still constitutive for ‘progressive research programmes’ (Lakatos). Writing on my personal computer (that clearly belongs to the social world), there is no doubt that ‘systems exist’ (Luhmann) without any subject or intersubjectivity, and not because they are socially integrated but simply because they are effectively adapted, and serve to keep communication running. Writing and replying to other voices on my personal computer, there is no doubt that social groups exist that must be integrated through understanding and truth claims. Whereas the ‘capitalist system’ (Marx) is a social system that is paradigmatic for the ‘object domain’ (Habermas, 2014: 18) of the functional perspective, the modern public sphere is paradigmatic for the object domain of the participant’s perspective.

Taking this for granted, I do not deny the crucial and indispensable role of cultural knowledge and personally mediated ‘learning’; that is, the ‘acquisition of insights through the negation of errors’ that ‘only occurs through the participant’s perspective’ (p. 19). However, I focus on the two sides of society, not in the ‘more narrow sense’ of system (p. 20), but in the broader sense that covers social integration and functional stabilization. When we turn our attention to the take-off of social evolution, we do indeed find a striking argument in Luhmann, which could easily, but ought not to, be misunderstood as a concession to Critical Theory (although it might be influenced by Luhmann’s long debate with Habermas). This argument is as follows:

Variation is triggered […] by communication that refutes or rejects communicative propositions. […] The refutation contradicts the expectation of acceptance. It contradicts the tacit consent that everything continues ‘as always’. All variation therefore is contradiction as disagreement, that is, not in the logical sense of contradiction, but in the original dialogical sense. (Luhmann, 1997: 461)

However, for Luhmann this original dialogue is nothing other than reciprocally observed double contingency that immediately leads to the differentiation of functionally stabilized social systems, which, decoupled from personal systems, are not able to integrate. In consequence, only expectations of acceptance are denied and not truth claims (a differentiation that makes no sense for systems theory), and there is only tacit consent and no idea of reaching an understanding that is (completely) ‘embodied in social facts’ (Habermas, 2014: 19, note 13). Once a process of systems formation is triggered, systems begin to learn cognitively in the same way as computers and other Turing machines learn through adaptation to their environment and to augment adaptive capacity. Despite this, at the same time, the same the original dialogue must be ‘conducted in determinate cultural and social contexts’ (Habermas, 2014: 19, note 15), and so it triggers another (a) ‘cognitive’ and (b) a ‘socio-cognitive and moral learning process’ (p. 19), which are as unavoidable as adaptive learning but intersubjectively directed by problems and truth claims. ²⁰ In this way, social groups and classes begin to

form, change and accumulate normative insight;

In my book, I have tried to combine Critical Theory and systems theory from this starting point, with the intention of letting critique do its subversive job in this way from within systems theory (Brunkhorst, 2014a: 11–20). From there I went straight to the revolutionary emergence of modern law, abstracting broadly from most of the evolutionary steps in between. ²¹