Introduction

1. The neo-Weberian state cannot be based on Max Weber’s own work since Weber’s understanding of the state and his ideal–typical construction of modern bureaucracy are oriented exclusively to forms of rule by domination (Herrschaft) [section II]. That is to say, Weber was not interested in the “internal” procedures of administration. Furthermore, domination in the sense of the expression of a ruler’s will involves “commands and obedience” and, as such, represents the “central characteristic of state law [at the time]” (Schönberger) which informed Weber’s thinking. For Weber, state and bureaucracy are not identical. Weber constructed an ideal type of modern bureaucracy. Weber’s ideal type of the modern, rational bureaucracy is a conceptual-methodological standard (Maßstab) and not a mere description. The text “Bureaucratism” (Bürokratismus) [MWG I/22–4: 137–234] utilizes the characteristics of the ideal–typical construction, but these characteristics are embedded in a descriptive representation, which creates the possible temptation to consider the ideal–typical construction as a description. However, Pollitt and Burckaert expressly refer to Weber’s ideal type. The above mentioned Weberian text is also comparative: it alludes to the expected development in the USA, and, among other aspects, it also predicts increasing bureaucratization of the system of governmental administration (Scaff, 2011: 188) and emphasizes the process of strict bureaucratization of political parties (MWG I/22–4:181; MWG I/17: 215–218),Footnote 1 which for Weber constitute the actual ruling power. These political parties act within the “urban machines” as a “political business enterprise” demanding a reciprocation of “electoral loyalty” in return (critically Roth, 2021; Offe, 2004: 66). The comparison also refers to pre-bureaucratic forms of exercising power, as part of a demonstration of the technical superiority of a fully developed bureaucracy, through an analogy between a fully developed bureaucratic mechanism and a machine which is superior to mechanical ways of producing goods (MWG I/22–4: 185). This comparative approach is simultaneously the descriptive presentation of a contextual sense of rationality.

This ideal type is neither a description of the prevailing form of bureaucracy then and now (Derlien 1989), nor is it identical to Weber’s understanding of the state. In his conception of the state, Weber distinguishes between the state in its “institutional” form (as “Anstalt”), and the state as an idea of validity (as “Geltungsvorstellung”); in this way, the notion of the state already points to the problem of legitimation (Treiber, 2016a). The first message of this contribution therefore is: There is no Weber in the neo-Weberian State and Weber’s account of the state is outdated. This assessment also holds true when notions of hierarchy and the power of authority (Weisungsrecht) apply. It is worth remembering that these are characteristics of Weber’s ideal-type of modern bureaucracy (and do not relate to his understanding of the state). Moreover, a more differentiated approach that draws on empirical studies shows that those at the highest political level practise an ongoing “dialogue” with the specialists in the basic units (Fachreferate) (Mayntz, 1997b: 193). Müller (2001) provides a vivid description and analysis of the origins of laws, using the example of soil protection (Bodenschutz). She points out explicitly that the “repertoire of professional behaviour for ministerial officials” requires them to observe the following maxim: “try to find a solution as much as possible at your level and in horizontal processes of coordination and agreement and do not harrass your hierarchy as far as the political leadership too much in order to remove conflict” (Müller, 2001: 25). Further, she indicates that “programme management” is one of the tasks of ministerial officials, that is, “the analysis of the conditions for action and the scope for action, and strategies for expanding an existing scope for action and mobilizing support for the subsequent implementation of the programme (draft law)” (Müller, 2001: 21). This is not a task envisaged for Weber’s “neutral bureaucrat”. Programme management does not arise for Weber because he primarily considered the preconditions that enabled the strict enforcement of rule within administration. Yet, neither do Pollitt and Bourckaert examine explicitly these management tasks that German ministerial officials actually undertake. The use of empirical studies would also show that permanent, national and transnational networks have formed in policy making (Schneider, 1986 and 1988).

Weber’s model of bureaucracy is historically specific in two respects: first, in that Weber is concerned with rule in the sense of the expression of the will of a ruler; second, in that “belief, commands, and obedience (…) represent(ed) the actual and central characteristic of state law” at that time.Footnote 2 Since, in accordance with this “central characteristic of state law” (Schönberger, 1997: 53), the model of modern bureaucracy came to include the strict implementation of the will of the ruler, this model represents, at first sight, a certain continuity along the dogmatic “track” followed by state law (Staats- und Verwaltungsrecht) nowadays.Footnote 3 As far as state rule in the present day is concerned, a corresponding ideal type would (have to) look different (Genschel & Zangl, 2008). It is worth highlighting the historical specificity of Weber’s thinking and the terminology he created. Above all, we must take Pollitt and Bouckaert “at their word” when it comes to the term they have formulated, the “neo-Weberian state”, and demonstrate, for example, how a number of the items listed in their collection of “neo-elements” are incompatible with “Weberian elements”.Footnote 4 As we shall see, it is easier to posit these elements than to apply them meaningfully (which is shown by the lex Furgler, among other things [section III, 1]).Footnote 5

2. Because of their particular legalistic orientation, the two authors pay special attention to the German (ministerial) administrative apparatus. First, this effectively restricts them to a set of administrative reforms indebted to the public management movement. Second, they give insufficient consideration to the constraints on the reforms (especially the federalist reforms), constraints that can be traced back, in part, to the constitutional state (Rechtsstaat) (section III). A legalistic orientation refers here both to the validity of the constitutional state (in the sense of strict recognition of rights and laws by the administration including oversight through the administrative courts) – an achievement of the constitution after the Third Reich – and also to the predominance of jurists in the administration, who also use law as a cross-cutting form of knowledge that facilitates co-ordination for them. The overly generalised reference to a “legalistic orientation” appears too broad in scope. A further aim of this contribution is to advocate to Pollitt and Bouckaert a more differentiated approach that draws on empirically informed studies. For reasons of space alone, it is not possible to achieve this fully, but sufficient indications are given to suggest a more differentiated form of analysis (see the following section).

While this legalistic orientation may have been ascribed to Germany in an overly generalised manner, it nonetheless retains a certain relevance. Keeping this in mind, it is not so much a matter of demonstrating differences between the American and German states, but rather of examining in more detail the differences established by Scharpf (1970) between German and American mechanisms of administrative control. Here we shall consider briefly, as an ideal–typical overstatement, the essential differences between German and American administration (rather than the state!). German administration is “obliged” to meet the “imperative” for the “normative precision” of laws “in advance” (Zwang zur vorwegnehmenden Normierung von Gesetzen) and in as much detail as possibleFootnote 6; this is a requirement of the constitutional state (Rechtsstaat). Apart from some exceptions (e.g. tax law) no such demand is made on the American federal administration. For the most part, this is because the majority of important (administrative) laws are what N. Luhmann terms “purposive programmes” (Zweckprogramme), i.e. problems that need to be solved. In purposive programmes, the “regulation of a problem that has become acute” is delegated “to a particular – often newly and specially created – authority” (or agency); purposive programmes “make use of criteria that require further concretisation” (Scharpf, 1970: 20f.). Here, the appropriate maxim runs as follows: “Here is the problem. Deal with it” (Scharpf, 1970: 21, note 30, after Davis, 1958, vol. 1: 82). This makes it possible for the American administration “to tackle the tasks assigned to it pragmatically at first and then to develop and improve practicable solutions through continuing application of the experience gained” (Scharpf, 1970: 21, 69). It is different in Germany: the “imperative (imposed on the administration) to specify in advance binding regulations for a detailed administrative programme leads to the situation where decisions on certain questions are, to some extent, held in reserve, before they have become practical or political problems. In this sense, the German model of the constitutional state makes greater demands on the norm-setting process than the American one” (Scharpf, 1970: 59). This can be seen to be mirrored in the “specifically American form of group pluralism” insofar as “the relative freedom of decision-making for individual authorities, their wide-ranging organisational autonomy, and the corresponding decentralisation of the parliamentary decision-making process (stand) in a necessary relationship with this group pluralism” (Scharpf, 1970: 63). The unity of administration assumed by the prevailing legal dogma no longer exists in Germany either, but the organisational autonomy of American administrative authorities is considerably greater.

However, invoking Weber (!), we must first consider an aspect that ascribes to present-day (ministerial) administration in Germany a strict hierarchical structure (together with an accompanying power of authority). Already in 1970–72, the empirical study undertaken by R. Mayntz and F.W. Scharpf into the creation of laws in the ministerial administration – in which I was also involved (Mayntz & Scharpf, 1975; Schmid & Treiber, 1975) – demonstrated that, while the power of authority may have applied in theory, it was hardly ever used in practice. Rather, when drawing up proposals for laws, the senior minister makes extensive use of the specialists in the basic units (Fachreferate) “subordinate” to him. As such, the functioning of the ministerial apparatus is not at all distinguished by “hierarchical management” (through the power of authority). Rather, according to Mayntz (1997b: 193), “the relationship between ministerial administration and political leadership” must be characterised “as a kind of ongoing (…) dialogue” (section III). Only the prevailing legal dogma, as it is taught in textbooks and in jurisprudence publications, clings on to the fiction of hierarchy, assuming that only in this way can the will of the democratic lawmakers (parliament) be guaranteed in an unadulterated manner. Empirical studies on the implementation of laws mentioned in the present contribution (Bohne, 1981; Breunung/ Treiber 2000) also show that, in the shadows of formal law, administrative practice has developed informal and altogether wide-ranging mechanisms through which petitioners (for installations requiring official permit) are involved.Footnote 7 In this way, empirically informed publications describe a co-operative or negotiative administration (section III, 3).Footnote 8 This means that critical reflections on the neo-Weberian state must not only take into account legal-historical publications but must also be as empirically informed as possible. It is striking that the literature cited by Pollit and Bouckaert (2017) on Germany repeatedly points to the dominant legalistic culture there but, ultimately, does not make reference to the many empirical studies that reveal structures and processes located in the “shadows” of the written culture of state law (the “strong legalistic tradition”). A notable example here is Hammerschmid and Oprisor’s essay to which the two authors refer: “German Public Administration: Incremental reform and a difficult terrain for management ideas and instruments”.Footnote 9 These studies certainly highlight peculiarities of the German federal system, but they fail to draw on relevant available studies that demonstrate the potential for legal blockades, such as Scharpf’s analysis of the “pitfall of political entanglement” (Politikverflechtungsfalle) (Scharpf et al., 1976).Footnote 10

Of particular relevance for the differing chances of success for new public management reforms in the USA and Germany is Scharpf’s study (1970) in which he compares the mechanisms of constitutional administrative control in Germany with American legal relief (introduction, 2). This comparison is considerably more instructive than a comparison that investigates differences in state structure and internal procedures (in the creation of laws or in mutual coordination) or in relation to the participation of organised and conflict-ready interests. The comparison presented by Scharpf is more insightful because he points out the elements of the constitutional state (Rechtsstaat) that are guaranteed by the constitution and which are obstacles to public management reform, at the same time making clear why the latter is considerably more compatible with the conditions in America. According to Scharpf, “judicial oversight does not focus [as in Germany, HT] on reviewing the content of administrative rulings, but rather on the strict control of the administrative process and on the further development and refinement of its regulations” (Scharpf, 1970: 25). This is complemented by the “weakness of legal obligations and controls” (Scharpf, 1970: 62). An additional factor in America is the “separation of the decision-making and party functions”, whereby the administration is compelled “to articulate its own points of relevance and offer counterarguments at a stage when it has not yet reached a final decision. Citizens discover what has priority in the ‘minds of the administration’, and, thus, the administrative process becomes a legal discussion with an entity whose decision-making premisses are still being developed and which, as a result, can still be persuaded through argumentation” (Scharpf, 1970: 23).Footnote 11 It stands to reason these conditions make new public management reforms considerably easier, since, to a certain extent, they already contain important “building blocks” or instruments for this reform, at least in rudimentary form, which the new public management reformers only have pick up and assemble. In Germany, legislators have to assume that courts will have the “final word” if their laws are not formulated precisely.Footnote 12 This is a significant obstacle to public management reforms, which have an affinity for administrative control as it is practised in America, in which “the American authorities (are) actors in the public political process” (Scharpf, 1970: 62). Insofar as Scharpf pursues the “political costs of the constitutional state”, he brings that state into a direct relationship with these “costs”, which constitute, in part, what is described as a “legalistic culture”. This means that the shift away from the legalistic culture might endanger the constitutional state as an achievement implemented and further developed after the end of the Third Reich. At the same time, this explains why a new public management reform that seeks to moderate or eliminate the “political costs of the constitutional state” has been met with so little affection in the (ministerial) administration.Footnote 13

Pollitt and Bouckaert make no mention of the fact that elements of the “legalistic culture” in Germany can be traced back directly to the constitutional state that was realised there after 1945 (Scharpf, 1970) and which must be seen as an achievement after the Third Reich. General reservations concerning this reform rest on the danger it might pose to these constitutional achievements. This is illustrated by the failed federalist reforms, during which the experts in constitutional and state law (= jurists) in the commission acted in the consultation as custodians of the constitution and the constitutional state, thereby obstructing any moderation or removal of “political entanglement” (Scharpf) and its capacity to acts as blockades (section III).Footnote 14 In addition, the inevitable failure of a long overdue reform, or proposed reform, of administrative law is highlighted, as well as the reasons for that failure (section III.1 and III.2). The proposed idea for the reform was overdue because it sought to “adjust” the assumptions of the dominant dogma to the prevailing administrative situation (i.e. “dogmatic normality”: for example, hierarchy as a guarantee of the strict implementation of laws passed by parliament or the unity and instrumental function of the administration) (section III). In any case, it has to be acknowledged that the concept for the reform of administrative law that was developed by respected academics (and by a judge in the Federal Constitutional Court), and the associated intention to achieve genuine impact (Wirkung) through law, demanded too much of the (administrative) jurists, insofar as they did not hitherto have the necessary methods at their disposal to achieve genuine impact through law.

3. It can be shown that, without any theoretical foundation whatsoever, a range of different functions are ascribed to the neo-Weberian state as a construct [section IV]: initially it is a question of “evidence/information” taken from reform records; then it is claimed to be a “summary description”, rather than “a theory” or “normative vision”. Next, the discussion moves on to “big models”, “general models”, or “paradigms”. Finally, the neo-Weberian state acquires the status of a “vision, as a modernized, efficient, citizen-friendly state apparatus”. It is noticeable that the absent theoretical foundation is replaced by the “illustrative use” of a familiar and highly appealing image, in the form of menus and dishes on offer. Instead, the specific and intuitively appealing metaphor of “menus” and “dishes” takes the place of that theoretical underpinning, a metaphor through which anyone can imagine anything. Here, reference to Weber would have been more appropriate had precise terms been used in the context of theoretical reflections clarifying unequivocally the status of the NWS as a construct – one thinks here of Weber’s critique of Stammler (MWG I/7: 487ff.). In the absence of a theoretically grounded justification Pollitt and Bouckaert thoughts on Public Management reform acquire instead the status of a reformist philosophy.

In order to demonstrate that the recourse to Weber through the notion of the neo-Weberian state, and, with it, to Weber as a “model” for public management reform, is more than questionable, we shall proceed as follows. Individual sections treat Weber’s understanding of the state and his ideal type of modern bureaucracy (point 1/section II) in order to then explore the fiction of continuity presented by German organisational law (point 2/section III). This fiction views the hierarchical construction of ministerial administration as functioning to fulfil, without deviation, the will of the democratic law-makers. This function is brought into question by evidence of the growing “autonomy of administration” (section III, 1), and this is supported, for example, by empirical studies on the development of particular laws (section III. 2). Indeed, the lex Furgler in Switzerland provides a striking example of the autonomy of cantonal administration, because the cantons gain “their political influence (…) above all through their capacity for implementation [of laws of the federal government], on which the federal government is dependent” (Linder, 2007: 5). Certainly, the cantonal administrations in Switzerland are an extreme, but well-researched example of the autonomy of bureaucrats, an example that is appropriate to call into question the often declared instrumental “ancillary” character of administration (section III, 2). The implementation study on the lex Furgler casts doubt on the guiding force (Steuerungskraft) usually ascribed to a law (section III.1) and emphasises that the reform of administrative law, as it has been conceived by reformist jurists in order to tackle the shortcomings of the predominant “dogmatic normality” in the German Federal Republic, is an extremely difficult undertaking (section III, 2). As long as a reform proposal like this is not adopted by courts and administrations, then it remains a purely academic matter (section IV). The project to “reform” administrative law is a difficult one, intended as it is to use laws to realise the intended effect of the lawmakers in such a way that administrative jurisprudence becomes a management discipline (Steuerungswissenschaft). For one thing, jurisprudence does not have the necessary means at its disposal. For another, it is tied to dogma and its decisions are “subject to” judicial oversight.Footnote 15

Finally, Pollitt and Bouckaert’s notion of the neo-Weberian state is subjected to further critique (point 3/section IV). Among other things, this critique highlights the fact that the shorthand NWS has been ascribed a range of different functions without any theoretical foundation: from simply acting as a “denomination” and “summary description” to becoming a “big model” or “paradigm”, and then further to the “vision” set out in the NWS. In this way, Pollitt and Bouckaert’s public management reform becomes a mere reformist philosophy. Furthermore, there has been no examination of whether, or to what extent, the various Weberian and neo-elements influence one another or are even mutually exclusive.

There is no Weber in the neo-Weberian State (NWS)

A review of the publications since the 1990s dealing with state and administrative reform under the heading of “Public Management Reform” reveals the arbitrary treatment of Weber’s work. To this end, it is necessary to consider in a little more detail the Weberian elements in Pollitt and Bouckaert’s neo-Weberian State (NWS) (2004; 2017: 121f.). It can be assumed that these elements are well-known and need only be listed in brief: (W1) centrality of the state; (W2) reaffirmation of the role of representative democracy; (W3) reaffirmation of the role of administrative law; and (W4) preservation of the idea of public service. The same applies to the “neo-elements”: (N1): “shift from internal bureaucratic rule – following towards to an external orientation” towards citizens; (N2): “supplementation (…) of the role of representative democracy” by consultation and direct citizen involvement; (N3): result orientation – “a shift in the balance from ex ante to ex post controls, but not a complete abandonment of the former” – “use of performance budgeting by central governments;” and (N4): the “‘bureaucrat’ becomes not simply an expert in the law (…), but also a professional manager.”Footnote 16

Weber’s model of modern bureaucracy is oriented towards rule as domination. That is, it guarantees the implementation of the will of the ruler and, thus, the implementation of laws (commands) imposed under the conditions of the Anstalt state, so that no non-compliance or deviation from the will of the ruler takes place in the administration (modern bureaucracy). Two features in particular point in this direction: first, absolute hierarchy and, second, the requirement for unfailingly obedient members of the administration (officials) who are obliged to remain neutral (WuG: 122ff., 124ff.; the features apply to the administrative staff). Hierarchy seems to have a “perpetual value” as a characteristic because a “fixed order of ranks, secured (…) by the right to issue directives, is (seemingly) best suited to ‘commands from the top being executed by the apparatus with the minimum possible frictional loss’”.Footnote 17

The fact that the structural organisation of the German ministerial administration exhibits some of the “iron” traits of every modern bureaucracy, such as hierarchy or specific spheres of competence, leads to the false assumption that Weber’s ideal type of modern bureaucracy represents an accurate description of the ministerial bureaucracy. When measured against the Weberian ideal type,Footnote 18 the German ministerial bureaucracy can still be acknowledged to possess a great similarity, even if the function of exclusive authority is not entirely present. Certainly, Weber’s model of bureaucracy is susceptible to “creative misinterpretations” that draw on Weber (Roth, 1971: 35). Hence, for example, the function of the ideal–typical construction of the model as an analytical tool to enable greater understanding of modern structures of rule has been misconstrued; instead, the formal structure of the ideal type has been “used to investigate the reciprocity between formal and informal relationships and rules. Under the influence of functionalism, Weber’s ideal type has often been equated with ‘manifest’ functions and structures and contrasted with ‘latent’ ones” (Roth & Bendix, 1959: 44). Reference has also been made to the efficiency thesis in organisational theory, which draws on Weber and is said to imply that, “according to all experience, (…) records-based bureaucratic and monocratic administration” is characterised by “precision, consistency, discipline, tightness, and reliability” (WuG: 128). This efficiency thesis has been divorced from the “framework of the sociology of rule” in which it was placed by Weber (Derlien et al., 2011: 206). Furthermore, it has been overlooked that Weber often made assertions of this kind in relation to pre-modern structures of administration, so that the superior efficiency of modern bureaucracy is relativised. If this is not taken into account, then the Weberian ideal type of bureaucracy will only be “received in a partial and ahistorical manner” (Derlien et al., 2011: 206; Derlien, 1989; Meier, Schimank, 2020).

For Weber, the twin characteristics of hierarchy and subjugation to instruction point to the Anstalt state of the nineteenth century, especially as the “recognition of the state as Anstalt” reflects the “identification of the state with bureaucracy, something which stood to reason in Wilhelmine Germany,” this bureaucracy being seen as “an entirely independent apparatus (of rule) from the citizenry”.Footnote 19 State and bureaucracy are not identical for Weber, and this is not respected by a formulation such as the “neo-Weberian state”. This does not take into account the “ancillary role” of the administration. It is also more appropriate to speak of Weber’s understanding of the state, since he never developed a “theory of the state” in the manner which the title of Anter’s book, “Max Webers Theorie des modernen Staates” (2014), might suggest. Weber’s understanding of the state has two core dimensions: the state as Anstalt and the state as a notion of validity (Geltung) (Treiber, 2016a: 67ff., 74ff.). The state as Anstalt refers to the debate undertaken in the second half of the nineteenth century concerning the juristic concept of Anstalt. An Anstalt is an enterprise (Unternehmen) calculated to endureFootnote 20 and is characterised by rationally constituted (legal) frameworks (Ordnungen) which are imposed on those subordinate to the leadership of the Anstalt (the coerced members) within a particular sphere of activity (territory). Only a marginal role falls to parliament in this Anstalt, the latter proving to be an organisation of rule (Herrschaftsverband) by virtue of its imposed authority: “The idiosyncracy in the German path of development lies in the way that this bureaucratic Anstalt state does not respond to the will of its citizens. Rather, as the personification of the Anstalt state, the officials of the monarchic administration tolerate certain rights of participation by parliament. In a particular way, the monarchic-bureaucratic apparatus acquires a separate status as the ‘state’. This assumption of autonomy is also actually the central element in the doctrine of state law advanced by Gerber, Laband, and Jellinek” (Schönberger, 1997: 27, 30; Lukas, 1908). Since Weber’s sociological concept of Anstalt respects the methodological principle by which very precise juristic terms are invoked when creating sociological concepts, only to then be given an altered meaning (MWG I/12: 405), he also ties himself to the constitutional law of the time. As Hermes (2016, 2006: 212) has demonstrated, “the way that the formulation of concepts in state law is positioned within organisations and forms of rule” determines “to a considerable degree the meaning and structure of the parallel sociological terminology”. In other words, Weber’s sociological concept of Anstalt is shaped to a significant extent by the equivalent juristic term, as well as reflecting the contemporaneous conditions in monarchic Wilhelmine Germany.

As far as the other central meaning is concerned, the state as an idea of validity, there are two references in Weber which result from his rejection of “collective concepts” and from which his intended meaning follows (WuG: 7). The following passage can be found in Economy and Society:

“The interpretation of action must recognise a fact of fundamental importance: that collective constructs belonging to everyday thought or to legal thought (or that of another specialism) are ideas of something in the minds of real people (or not only judges and officials but also the ‘public’) which, in part, exists and, in part, should be valid, and to which their action is oriented; and also that, as such, they have a very powerful and often almost dominant causal importance for the manner in which the action of real people proceeds. Above all, as ideas of something that should be valid (or also that should not be valid). (A modern ‘state’ consists, first, to a not inconsiderable degree, in this manner: – as a complex of a specific cooperative action of people, – because particular people orient their action to the idea that it exists or should exist: thus, that frameworks of this legally oriented manner have validity.)”Footnote 21

Weber expresses himself more concisely and more incisively in a letter to Liefmann:

“In a sociological sense, the state is nothing but the chance that particular types of specific action take place, the action of particular individuals. Otherwise, nothing (...). What is ‘subjective’ about this is that the action is oriented towards particular ideas. What is ‘objective’ is that we – the observers – make judgements: the chance exists that this action, oriented to these ideas, will take place. If the chance no longer exists, then the “state” no longer exists.”Footnote 22

Since those undertaking action have in their minds this aspect of what “ought to be” as an idea of validity, the concept of validity is co-determined by a “particular element within action” (Lübbe), which the observer must deduce according to the standard of what is objectively judged to be possible.Footnote 23

The applicability of Weber’s model of bureaucracy to the philosophies and proposals for the reform of modern administration,Footnote 24 which are aimed at the internal structure of modern administration, also fail. For, Weber never fulfilled what he was demanding for the structure of bureaucracy when, rejecting “the action of collective personalities”, he spoke of sociology meaning “thereby rather a particular type of procedure involving the actual social action of individuals or that conceived as possible” (WuG: 6f.). Weber conceives the “state” as a “complex of specific cooperative action by people” and thereby invokes the sociological concept of validity that has proved itself an important element in Protestant ethics: “binding” religious beliefs in the minds of the puritans, whereby there really existed the chance that the latter would orient themselves to those beliefs and, in this way, even have historical influence (Lepsius, 2003: esp. 35ff.). In this way, Weber both satisfied his “causal requirement” (MWG I/9: 490) and reduced the collective notion of the “state” to “comprehensible action, and that means, without exception, to the action of the individuals involved” (MWG I/12: 405). However, his use of the concept of validity in his basic sociological concepts situates it in close proximity to the concept of legitimacy, through the “superadditum” of prestige invoked in this latter term. For Weber, a “complex of specific cooperative action by people” means those people subject to rule, as the letter addressed to Robert Liefmann on 9 March 1920 shows. Even if a “state” ceases to exist sociologically, the administrative staff can continue to exist in the form of modern bureaucracy as a result of the officials’ “cultivated attitude of submissiveness” and/or the “objective indispensability of the apparatus that once existed” and its “own impersonality”.Footnote 25

Those jurists who were engaged with the state in the 1871 Empire in the context of so-called “legal-constitutional positivism” (Dreier, 1991: 94), and to whom made Weber reference, conceived this as an “association of will” (Willensverband): “It is the highest form of will, the ‘will of wills’, the centre of will. The power of will in the state is state power. And the juristic construction of the state is the cardinal law of this new doctrine. ‘Like all juristic construction, it is to do with detecting and delimiting the relations of will.’ Hence, the state consists only of relations of will, in particular those between ruler and ruled, that is, relations of rule, and they characterise the relations of the subjects.”Footnote 26 In this context, it is important to refer directly to Weber’s definition of rule, which he conceptualises in such a way “that a manifest will (‘command’) of the ruler(s) seeks to, and actually does, influence the action of others (of the ‘ruled’)” (WuG: 544; MWG I/22–4: 135). The concept of rule is also defined by the “implementation of will” and thereby indicates how greatly Weber was influenced by the prevailing view of the time. As such, Weber is indebted to a large extent to the predominant contemporaneous theory of state law, in which, in the second half of the nineteenth century, “beliefs, commands, and obedience represented the real and central characteristics of (German) state law” (Schönberger, 1997: 53). Jellinek expresses a similar view in his Allgemeine Staatslehre (General Theory of the State), in which we read: “More specifically, the state consists in the relations of will of a majority of people. People who give commands and those who grant obedience to these commands, constitute the substratum of the state”.Footnote 27 Further, Gerhardt suggests that “this practically and immediately comprehensible construct that lies behind the commands which a person or others give, (…) that (is) exactly what Kant calls ‘will’”.Footnote 28

In this way, the administrative staff or modern bureaucracy remains a “black box”. In other words, we learn nothing from Weber about the “specific cooperative action” of the members of the administration (officials) who act within it. The division of labour brought about by specialisationFootnote 29 might have made it possible to “look” into the “inner workings” of the apparatus by means of the coordination processes, it makes necessary and to consider the level of action (in the sense of the “cooperative action of people”). However, Weber never undertakes this further analysis.

A more differentiated consideration of legalistic culture, taking into account, from an internal perspective, ongoing dialogue and, from an external perspective, administrative cooperation – a legalistic culture that is closely connected to the achievements of the “constitutional state”

If we compare the necessity for hierarchy present in the Empire of 1871 and its rationale with the customary justification for the principle of hierarchy in the democratic state, then we discover a high degree of congruence: “Because of the aim to realise the will of the democratic lawmakers with the least possible deviation, the hierarchical model based on the office system, the strict right to issue directives, and a professional civil service also proves to be suitable and sufficient for democracy, together with its plausible sounding advantages of greater objectivity, greater impartiality, and clearer lawfulness” (Dreier, 1991: 126, 305). However, here it is necessary to heed Scharpf’s warning. When considering the legal review and regulation under constitutional law of laws or provisions presented by the ministerial bureaucracy, he explains that “in this area, (legal) relief under constitutional law appears to function as an instrument of an administrative and disciplinary hierarchy in relation to the executive authorities, especially the federal states and municipalities. However, the criterion of correctness (Richtigkeit) changes its functional meaning as soon as the prescriptions of the administration are formulated less precisely and require further concretisation. Here, the intensive scrutiny of what is correct must lead to a transfer of decision-making in the matter to the administrative courts and of the competence to concretise the programme of the administration”.Footnote 30

Examining the “long path from the absolutist machine model to the complex role of administration in democratic parliamentary states” (Dreier, 1991: 307), Dreier first acknowledges, with a series of additional assumptions, that, “against the background of far-reaching alterations to the system of state forms, the sustaining principles of administrative organisation since the genesis of the absolutist machine model prove at first sight to be surprisingly invariable: that is particularly the case for the fundamental assumption of their ancillary function and, with it, their instrumental character. In the course of development – so it would seem – the (specific) princely will [that of the ruler] was replaced as the decisive impetus simply by the (general) law” (Dreier, 1991: 305). However, Dreier criticises this “fundamental assumption”, referring to traits that, according to Schmidt-Aßmann, characterise the so-called “dogmatic normality”. This is distinguished by three central assumptions of conventional dogma: first, by the assumption that the “administration must be conceived as a whole, which implements laws through subsumption and is subject to complete judicial oversight”Footnote 31; second, by the “influential general principle of organisational law”, which has in mind the “monocratic and hierarchical structure of administration”; and by the furnishing of the hierarchy with the right to issue directives as the guarantee that laws adopted by parliament will be strictly implemented (Schmidt-Aßmann, 1997: 25f.). Nonetheless, the positive evaluation of hierarchy in organisational law overlooks the fact that the administrative courts have the final say in cases where the ministerial bureaucracy has developed imprecise laws. Furthermore, empirical studies show that, despite the existing hierarchy and the validity of the power of authority (“commands”), in practice an “ongoing dialogue” was established between the political leadership and the specialist units at the base (Fachreferate) (Mayntz, 1997b: 93).

It is necessary here to briefly explore the peculiarities of German federalism, because its characteristic “political entanglement” (Politikverflechtung) (Scharpf) encourages an incremental form of politics and also causes possible blockades in the system. Scharpf characterises political entanglement through two “contrary styles of interaction” that act upon one another – “negotiation between member state governments, which is directed towards protecting one’s own interests through the principle of compromise, and confrontation between party-political positions, which is staged for victory or defeat” (Scharpf, 1994: 68f.). According to Scharpf, “competition between parties, plus federal political entanglement,” leads to a “form of antagonistic cooperation which in reality tends towards political immobilism” (Scharpf, 1994: 69). Together with blockades, we find strategies to “minimise the need for consensus” (e.g. by reducing the number of participants); if these strategies are not sufficient to achieve a consensus of action, “rules for decision-making that minimise conflict” are used (such as the prevailing of existing structures, equality of treatment, protection of vested rights, conflict deferral, or the renunciation of intervention). In turn, implementing “rules for decision-making that minimise conflict” has consequences for applying very specific governance programmes with very specific governance instruments, which, for their part, are ill-suited for resolving “problems of distribution and interaction”.Footnote 32 This applies to the Federal Republic of Germany, but not to Switzerland, where “the canton governments do not have at their disposal a veto on federal politics” and where the cantons gain “their political influence (…) above all through their capacity for implementation, on which the federal government is dependent”.Footnote 33

In his study on the “reform of federalism”, Scharpf also analyses numerous restrictions on this reform project, not all of them inherent in the federalist structure. In this way, he provides a striking example of how difficult it is to carry through reforms and of which restrictions “stand in (their) way”.Footnote 34 It would also have been easy to establish a connection to Max Weber here. In his reflections on the politics of the Weimar constitution, Weber discussed the advantages and disadvantages of a solution based on either a Staatenhaus (a representative system) or a Bundesrat (a delegate system). However, he was convinced that the federal assembly (Bundesrat) would remain the solution since “the governments of the individual states (would never) allow themselves to be forced out also of their shared decision-making position in the administration” (MWG II/10–1: 374f.). As such, Scharpf sees in the “combination” that prevailed at the time – “of a comprehensive law-making competence in the central state and an equally comprehensive power of veto on the part of the member-state governments in the second chamber [Bundesrat]” – the core of the “present political entanglement” (Scharpf, 2009: 17).

Consequences of the “dogmatic normality” (dogmatische Normalsituation): The limited guiding power of laws

Dreier certainly questions the “guiding power (Steuerungskraft) of laws” insofar as he demonstrates a variety of restrictions on that power.Footnote 35 However, he does not pose the fundamental question about the circumstances which make it difficult or impossible for an administration that is obliged to obey the laws to exercise the guiding function which is ascribed to them (Rechtsstaatserfordernis). Those reform jurists who recently conceived administrative law as an academic branch of governance (Steuerungswissenschaft) have sought to answer this question, thereby freeing it from its position of a contingent, functional resource for the application of lawFootnote 36 and seeking to transform it into an academic discipline. The “crisis of so-called regulatory law” addressed in this way is contrasted with the approach in management science, which decisively criticises the reductionism of the situation of “dogmatic normality” with the suggestion that the latter could no longer appropriately capture “the variety of forms in which administration manifests itself under the conditions of Europeanisation and internationalisation”.Footnote 37 However, what separates Dreier from the reform jurists is their perspective, gained through metaphorical transferFootnote 38 and borrowing from Mayntz’s governance model, which appeared well-suited to replace an understanding oriented towards legal acts (Rechtsakte) with one oriented towards effect. Hoffmann-Riem has formulated the intention associated with this as follows: “Administrative law is a medium of governance, first, in that the administration itself is ‘governed’ and, second, in that it acts to govern the behaviour of others: the aim is to bring about effects.”Footnote 39 It is apparent that an administrative jurisprudence that sought to identify interdependencies and achieve effects would, for one thing, be overwhelmed and, for another, be able to accomplish little against the autonomy of the (cantonal) administrations.Footnote 40 Administrations provide evidence of their autonomy by applying one and the same law in a manner different from the intentions of the lawmaker and with completely different results. It can be assumed that this finding, together with phenomena such as a cooperative bureaucracy that does not rule but rather negotiates, would appear to Weber as a “turn to the irrational”.

The limits of a reform of administrative law

A more sober perspective on this ambitious project of the reformist jurists is provided by the rigorous and extensive long-term case study on “Grundstückserwerb durch Ausländer in der Schweiz” (“Land Acquisition by Foreign Nationals in Switzerland”, Delley et al., 1984). This case study demonstrates how complex interdependencies can be in reality and how difficult it can be to determine effect, in particular when there is a “lack of clarity in the legislative objectives” or when a law expressly fails to name any goals. This was the case for the lex von Moos, which preceded the lex Furgler, and sought to limit the acquisition of property by foreign nationals in Switzerland (without explicitly naming this goal). The lex von Moos required that foreign applicants purchasing land (property) prove a “justified interest” (“berechtigtes Interesse”). Without entering into further detail (Treiber, 2017b: 433ff.), what is important here is not the obvious explanation for the implementation deficit at the cantonal level in the absence of stated legal objectives, but rather the autonomy of the cantonal administrations involved in implementation, administrations that the lawmakers sought to guide but that guided themselves. If we follow Linder on the lex Furgler, “the cantons (had) used their competence in the law for their own ends, and in a completely different direction, and had shaped its implementation in even more divergent ways. While Lucerne linked the limit on land acquisition to the moderate development of tourism and followed the aims of the federal government more or less exactly, other cantons instrumentalised the federal law for completely different purposes: Geneva undertook the construction of social housing and Valais pursued tourist development. Each used a similar incentive system: the authorisation sought by applicants with foreign residency was tied to conditions that indirectly contributed to financing the construction of social housing or tourist projects” (Linder, 2012: 198). According to Linder, this example remains “relevant today” (that is, in 2012). At the same time, Ellwein and Wollscheid (1986: 320) see in the difference between that which the administration was intended to achieve by the lawmakers and that which they eventually undertook not the “oft cited implementation deficit, but rather one possible realisation of what was wanted [in reality on the part of the administration]”.

We can consider the intended effectiveness of a law as follows: “Measured against the proclaimed intentions (aims) of the lawmakers (governing subjects), we can speak of ‘selective ineffectiveness’, whereas the autonomy of the cantonal administrations helps to transform this into the ‘selective effectiveness’ of their own objectives, in the manner of self-governance” (Treiber, 2017b: 437, lightly altered). Linder has attempted to generalise the evidence of the lex Furgler on the basis of a reasonable number of case studies and with the help of a two-by-two table. According to this table, the “extent of consensus at the federal and cantonal levels” is decisive for the “extent of implementation of federal laws” in Switzerland (Linder, 2012: 198ff.). Here, he is able to draw on Scharpf who demonstrated, with reference to Weber, how judgements of probability that rest on a “generalised consideration of the individual case” (von Kries) are theoretically justified through nomological and ontological knowledge. Nonetheless, Scharpf overlooked the fact that Weber himself drew on the theory of objective possibility conceived by Johannes von Kries (Buldt, 2019). According to Scharpf, the method adopted by Linder is appropriate if “politically active (corporative) actors in differing underlying institutional conditions react to differing situational challenges with differing cognitive and normative orientations for their action, and (…), thus, the constellations of relevant explanatory factors do not often arise in identical form”. This concerns their ontological conditions (von Kries).Footnote 41

The role of the ministerial bureaucracy in the development of laws: results of an empirical investigation: Dialogue instead of hierarchy

If we take into consideration not only implementation studiesFootnote 42 but also empirical investigations into the ministerial organisation (of the FRG) in the 1970s that are guided by the question of the participation of the bureaucracy in the development of laws (programme development/ Programmentwicklung), then the construction of administration as a “technical apparatus determined and led by a separate will” (Dreier, 1992: 137) diverges from the evidence that also grants the ministerial bureaucracy a certain “autonomy”. Above all, the study of programme development prompts a shift in perspective,Footnote 43 by which the structural and procedural organisation of the ministerial bureaucracy becomes apparent, that is, the”inner workings” of the “black box”. Weber touches to some extent on organisational structure when he introduces the traits of a “fixed hierarchy of offices” and “fixed competences of offices” into his ideal-type construction of modern bureaucracy (WuG: 126). Here, Weber might have asked how the specialists are co-ordinated with one another or what particular problems arise in their “cooperative action” (WuG: 7). It is left to Scharpf to analyse the problem of “negative coordination” created by the division of labour or, more precisely, by specialisation and “selective perception”.Footnote 44

If we consider the kind of empirical investigations that use a policy science approach to study German ministerial bureaucracyFootnote 45 – or more exactly its “contribution” to programme development (the development of laws) – then the expression “administrative autonomy” acquires particular importance. This is simply because laws are created by the ministerial bureaucracy and many programme initiatives proceed from specialised units at the base (Fachreferate) and often also “from outside”, prompted by organised and conflict-ready interests. Viewed in this way, government ministers find themselves in an inherent position of dependency on the base, but one which can be “moderated” by virtue of the fact that it is introduced at a relatively early stage in the process of programme development. In this sense, the “ancillary role” ascribed to the bureaucracy must be placed into question.

Organisational and labour differentiation create an awareness of the relevant sphere of competence and the associated environment in which organised and conflict-ready interests act. This corresponds to the tendency towards “selective perception”: “Each specialist unit tends to limit its awareness to its own sphere of competence, to perceive less clearly problems that lie beyond its boundaries, and to consider those problems less important” (Scharpf, 1973a: 81). The discrepancy between a “structure of problems that are interdependent and structures for dealing with those problems and for decision-making that are segmented” is responsible for the pattern of adjustment known as “negative co-ordination” (negative Koordination) (Scharpf). The units involved in internal coordination, which also understand themselves as “advocates” for the interests acting in their surroundings, tend to “scrutinise any proposed decision that is under discussion for potentially negative consequences for their environment and to block all alternatives in which these negative consequences cannot be ruled out” (Scharpf, 1973a: 88). The result of all of this is a “politics of small steps” in the sense of an incremental form of politics. In other words, the prevailing structural and procedural organisation does not exactly encourage “innovative policies”, but these are possible under particular preconditions. As the empirical study of the policy programme for transport (Verkehrspolitisches Programm) has shown, a certain level of innovation could be achieved by not involving heads of basic units (Fachreferate) in this programme with their numerous external contacts, but rather their Hilfsreferenten (“section assistants”)Footnote 46 who were not embedded to the same extent in the structures outlined above. These Hilfsreferenten, who worked in departments responsible for particular modes of transport (Verkehrsträger), were employed in these departments in the mornings in order not to be disconnected from the flow of information but were then brought into an operating unit where they worked on the innovative programme in “secrecy” and freed from the differentiated structures of the organisation, hence transcending specific areas of responsibility (Kussau & Oertel, 1974). Although these studies are already somewhat dated, they do allow us to identify empirically grounded trends. The study cited also revealed that the normative image of a top-down decision-making process with a strict functional division between politics and administration, as assumed by Weber,Footnote 47 needs to be set against a cooperative form of dialogue and mutual influence between the government minister and the apparatus.Footnote 48

The role of Weber’s “neutral bureaucrat” has also changed, although the diverse studies present contradictory results (Derlien, 2003b: 420; Treiber, 1977: 218ff.). Weber’s ideal–typical opposition between politicians and bureaucrats is strongly influenced by his historical observations and his remarks on officialdom under and after Bismarck (Mommsen, 1974; Röhl, 2000; Schluchter, 1980). This critique includes the charge that Bismarck’s Caesarist regime (Hofmann, 1986) hindered the selection and development of politicians, so that, after his removal, German politics was increasingly shaped by individuals with a “civil servant’s mentality” (Beamtenmentalität) who were anything but suited to pursue the “political profession”, which is determined by the “will to power” (Weber 1992: 189f., MWG I/17). Empirical studies, such as those presented by B. Steinkämper and Putnam,Footnote 49 indicate that, particularly among younger officials and those working in the field of administrative planning, a shift in mentality had taken place (Blankenburg & Treiber, 1972), such that they were unable to identify with the role of the classical bureaucrat.Footnote 50 The study of policy programme development from the perspective of policy science has also noted the change in mentality, in that the officials who were heavily involved in programme development counted consensus building and conflict management as part of their duties and, hence, viewed the implementation of the recommendations (jointly) developed by them as their primary task. (Müller, 2001: 21ff., 25).

The arbitrary use of the construct of the neo-Weberian State (NWS) renders the public management reform conceived by Pollitt and Bouckaert a reformist philosophy

In light of the discussion presented above, it is now pertinent to undertake a more detailed consideration of Pollitt and Bouckaert’s neo-Weberian state (NWS) and, more specifically, its four characteristic Weberian elements and neo-elements.Footnote 51 First of all, we must clarify the status afforded to the NWS. To this end, the different characterisations of the NWS in Pollitt and Bouckaert can be “ordered” in a meaningful way. The starting point is the suggestion that the NWS involves “some rather general common denominations” which the two authors have identified across reform reports relating to six European countries (Pollitt & Bouckaert, 2017: 122).Footnote 52 From this perspective, the questionable neo- and Weberian elements constitute findings drawn from “reform records” or transactions, and the suggestion here is that the NWS is not to be seen “as ‘Weber’ plus NPM”. They add that the NWS is a “summary description, not a theory, not our normative vision” (Pollitt & Bouckaert, 2017: 122). This can be related to the statement that the NWS (as well as NPM and NPG) represents “the best way of describing and classifying what has been going on internationally in public management reform in terms of big, general models” (Pollitt & Bouckaert, 2017: 26). Hence, the NWS is assigned the designation and, with it, the status of a “model”, accompanied by the observation that the NWS represents an “heuristic attempt” “to give a rough shape to a very complex and messy reality” (Pollitt & Bouckaert, 2017: 120). This raises an important function of a model, namely to simplify or reduce complexity, since, although the term model is repeatedly used together with the term paradigm, it can be assumed that the two terms are instead applied as commonly used “labels”. As Wagner (2013) has shown, this kind of usage is at odds with a careful reading of Kuhn’s concept of the paradigm. Since any more detailed (theoretical) justification for the use of “model” and “paradigm” as concepts is missing, it is reasonable to assume that this justification has been replaced by the extremely vivid metaphor of the menu, which is treated as having “equal status” to big models and paradigms (Pollitt and Bouckert 2017: 26, 28, Fig. 1.3).Footnote 53 Naturally, there is no indication of the extent to which this changes the evidence identified in the reform records, for example, whether they become constitutive features of the NWS model or reform aims: the authors suggest that “some commentators have used it [NWS] as a normative vision”, before confessing that “in this 4th edition we too have allowed it something of that quality” (Pollitt & Bouckaert, 2017: 122). In this context, they continue that the NWS “serves as an omega”, that it is “a vision of a modernized, efficient, citizen-friendly state apparatus” (Pollitt and Bouckert 2017: 122). Finally, we discover that “the general idea of an NWS has been constructed as part of a political strategy responding to globalization”, more precisely “as a defensive strategy by previously corporatist regimes (Germany, France, the Netherlands, Sweden) to try to protect the ‘European social model’ and the ‘European administrative space’ from the depredations of globalized neo-liberalism” (Pollitt & Bouckaert, 2017: 122). The extent to which there are connections between the neo-pillars and the Weberian pillars is not discussed, although it can be gathered from the book that the realisation of W3 (“reaffirmation of the role of administrative law”) impinges on the implementation of N3 (“limited use of performance budgeting by central governments” (relating to Italy, Belgium, and Germany) (Pollitt & Bouckaert, 2017: 80, Fig. 4.2; 121f.).

With a certain duality of meaning, the most striking “commonality” between the NWS and Weber may arise from the fact that the NWS follows Weber’s sociology of rule from a chronological perspective, thereby justifying the recourse to the Weberian conceptual framework for legitimising purposes.Footnote 54 This becomes evident when we examine individual Weberian elements in greater detail. In W1, for example, we read: “Confirmation of the role of the state as the most important institution that facilitates solutions for new (…) problems.” There is no requirement to use Weber in order to ascribe the state an active role in dealing with new kinds of problems.Footnote 55 Furthermore, Weber rejects the collective concept of the state, as the corollary of the Weberian state as an Anstalt state which, entirely in keeping with the exercise of rule, imposes laws onto its subjects and attributes a peripheral role to parliament. Moreover, the concept of the Anstalt state is the theoretical reflection of officialdom within a monarchic state of that historical period. The conceptual parameters of the Anstalt state are not tailored to deal with the kinds of problems that need to be addressed and solved today.

Neo-element 1 declares that there is a “change from internal orientation to bureaucratic rules to external orientation to the needs and wishes of the citizens”.Footnote 56 Since Weber also subsumes laws under “bureaucratic rules”, this contradicts to some extent Weberian element 3, which discusses the confirmation of an appropriate, modernised administrative law. Dreier’s analysis outlined above demonstrates how difficult and demanding a reform of administrative law on Germany is. On the one hand, he indicates the autonomous action of the administration; on the other, he only selectively criticises the assumptions of “dogmatic normality” (without any impact). The necessary reform of administrative law propagated by reformist jurists, which should render the state more capable of acting under “the conditions of Europeanisation and internationalisation”,Footnote 57 led to a three-volume handbook which has had little impact on practice. The reform project to ensure laws achieve their intended outcomes could not be realised in reality.Footnote 58 Lüder’s (2004) informative contribution in the Festschrift for König sets out the difficulties involved in the reform of resource management envisaged under N3 even in Germany, since this is dependent on a number of contextual conditions that are very different in different countries. The restrictions of the “political entanglement” present in the federal structure of the FRG – which can be traced back to the retention of the Bundesrat in the Weimar constitution and which, at best, “allow” incremental reform – are disregarded in the NWS.Footnote 59 The reasons for including Weberian element 3 – “confirmation of the role of representative democracy” – among the Weberian elements have to be questioned.Footnote 60 Weber’s Anstalt state attributes a very peripheral role to parliament, and Weber’s constitutional suggestions for rendering the state parliamentary were only published in 1917 and 1918 in the Frankfurter Zeitung (Weber, 1971a and b), after which he participated in December 1919, under Hugo Preuß, in the deliberations on the Weimar constitution. Weber’s suggestions on this matter are clearly also a reaction to the Wilhelmine Empire and to the “Caesarist regime” of Bismarck, which helped to establish the civil service state after his removal (Weber, 1971a: 347; Hofmann, 1986; Röhl, 2000).Footnote 61 Neo- und Weberian elements no. 4 are connected with one another. Since the nineteenth century, jurists have been preferred in recruitment in the Federal Republic (Bleek, 1972).Footnote 62 As long as jurists are preferred as recruits to the German ministerial administration and a legalistic culture exists,Footnote 63 which adheres to contemporary administrative law (or remains bound to it), a professional (and, where possible, non-specialist) manager will remain the exception, even if ministry officials actually today practise management tasks (Müller, 2001: 21ff.). Furthermore, it is worth recalling here Scharpf’s (1970) study on “Die politischen Kosten des Rechtsstaates” (“The Political Costs of the Constitutional State”). By tracing these “political costs” back to the constitutional state (Rechtsstaat) and proving that these are fundamental components of the “legalistic culture” in Germany, Scharpf makes it clear that any new public management reform applied to them provokes the fear that it will at the same time endanger important achievements of the constitutional state.

Conclusion

It is difficult to connect the neo-Weberian state (NWS) with Max Weber, especially since the corresponding theoretical foundation is absent. The selection of four Weberian elements, their possible links with one another, and the possible implications of the four neo-elements all lack any embeddedness in an underlying theoretical framework through which at least some, if not all, of the different functions attributed to them could have been justified. Here, use could have been made of Weber, for whom ideal types are “theoretical constructions making illustrative use of the empirical” (MWG I/7: 222). Pollitt and Bouckaert have clearly considered the illustrative use of the empirical when they indicate that their book is “not primarily a ‘theory and methods’ book: it is a book about what has happened and why” (Pollitt & Bouckaert, 2017: 29). Yet even the “why” requires a minimum of theoretical justification that explains the rationale for combining the three menus discussed (NPM, NWS, NPG) – professed to derive from empirical evidence – with very particular dishes. Thus, the metaphorical use of combinations of menus and dishes in Fig. 1.3 remains at the level of illustration (Pollitt & Bouckaert, 2017: 28), even if it does appear plausible because of its highly suggestive nature (in the sense of its apparent “conclusiveness” or Beweiskraft). However, here again Weber would have been of value, when he cites Goethe as follows: “The peculiarity of meaning in artistic representation is inherent to any purely suggestive [representation]: ‘Everyone sees what they carry in their heart’ (…)” (MWG I/7: 227, slightly altered).

Moreover, the neo-Weberian state rests on a distortion: even under the influence of a “legalistic culture” particular structures and forms of action develop that cannot be captured with the label “law-based state” and which, in this way, reproduce “law in the books” rather than “law in action”.Footnote 64 The label “legalistic culture” alone merits a more differentiated analysis, which, based on Scharpf’s study of 1970, both raises awareness of the constraints feared to be imposed on the achievements of the constitution and shows, drawing on empirical studies, that cooperation or negotiation takes place in the administration both internally and externally, cooperation/negotiation that has developed in the shadows of the law. In this sense, we can speak, from an internal perspective, of dialogue between ministers and heads of basic units (Fachreferate) and, from an external perspective, of a “negotiative or cooperative administration”. In this contribution, discussion of the selected empirical studies has sought to draw attention to this situation. There would have been much to be gained from an account that limited itself exclusively to the contemporary “modern state”. That alone is an extremely ambitious task.

Scharpf (1991) pursued this task with the help of theoretically founded reflections, which were followed by empirical studies at the Max-Planck-Institut für Gesellschaftsforschung (Cologne) into what are known as “sectors close to the state” (such as the health system). These reflections provided a much more favourable assessment of the “negative coordination” that had previously been viewed critically, analysing it in combination with “positive coordination” (Scharpf, 1991: 17ff.) – precisely in the context of “selective networks” within long-standing relationships (Mayntz, 1991). As a first step, Scharpf (1991: 10ff.) compares, among other things, “hierarchy and negotiative systems” against the theoretical standard of welfare economics, using the Kaldor criterion and the Coase theorem (Scharpf, 1991: 10f. and 15f.). This involves comparing the respective preconditions and optimal performance of the classical state model (of hierarchical coordination) with the “optimal functioning of negotiative systems” (i.e. horizontal coordination), with particular reference to the “negotiative dilemma” (the dilemma between, on the one hand, problem solving and, on the other, the struggle over distribution, in the game theory sense of “mixed-motive games”). Following this comparison and taking into account the possible combinations available, Scharpf urges addressing the “interrelationship between hierarchical and non-hierarchical political forms” in interdependent systems from his own theoretically grounded perspective: “At the end of the twentieth century, the state acts in an ever denser web of negotiated relationships, both internally and transnationally, that severely constrains its capacity for unilateral hierarchical management. However, (…) the [observable, HT] negotiating systems have at their disposal their own autonomous welfare potentialities.Footnote 65 As a result, they do not require extensive management, but instead only corrective interventions and complementary participation by instances of the state. To this end, even when they are no longer able to issue commands, they possess unique orientations and potential for their actions, which are also effective in complex negotiative networks.Footnote 66 Bearing this in mind, the question of the state’s capacity to act at the end of the twentieth century might have an altogether more positive answer” (Scharpf, 1991: 29).