Or: Like a Rolling Stone – The Ongoing Metamorphosis of Ross in Tû-Tû, On Law and Justice, and His Eventual Abandonment of Logical Empiricism
In Part III of this “Very Scandinavian Saga”, we have seen that Ross, in his Towards a Realistic Jurisprudence of 1946, following Hägerström’s empiricism, conceives the legal phenomenon as the mere result of the inductive interaction between a disinterested psycho-physical reaction of respect and conformity to declarations of authority, synchronously complemented by a system of compulsion necessary to trigger interested conduct actions in the face of the threat of sanction, so as to enforce such declarations of authority on occasions where their spontaneous observance does not occur (Ross 1961: 89-94).
Following this approach, fundamental legal concepts such as ‘right’ or ‘obligation’ would be nothing more than, as Hägerström puts it, “mystical-magical bonds” originating from a superstition transmitted over centuries and crystallized as a socially conditioned psychological function, precisely reinforced by the system of compulsion (1961: 249–250) that confers social stability on legal normativity (1961: 90).
Nonetheless, as will be seen in this part, Ross would gradually distance himself from the radical empiricist scepticism originally proposed by Hägerström. This gradual metamorphosis leading to the scientific acceptance of empirically non-existent legal categories initiated in his Tû-tû of 1951, passing through his paradigmatic work On Law and Justice, up to his effective abandonment of logical empiricism expressed in a letter to Eugenio Bulygin in 1966, amplified in his Directives and Norms of 1968, and effectively concretised in On Self-Reference and a Puzzle in Constitutional Law of 1969.
Let us take a closer look at this metamorphosis.
A Gradual Metamorphosis: Tû-Tû, On Law and Justice, and the Possibility of Attributing Meaning to Expressions Without Semantic Reference
In his succinct yet highly illustrative text Tû-tû (originally published in Danish in 1951), Ross claims that, despite being devoid of semantic reference, certain normative notions – such as “obligation,” “rights,” or “tû-tû”, used by the superstitious Nolt-cif tribe inhabiting the imaginary Noisulli Islands of the South Pacific (1957: 812) – fulfil significant linguistic roles of expressing commands or rules and describing certain “states of affairs” (1957: 813). Thus, although recalling to some extent magical thoughts and powers (1957: 818), they would be useful for simplifying relationships subject to complex layers of normative regulations (1957: 821), thereby facilitating both the understanding of, and influences upon, these very regulations themselves (1957: 817).
Even more drastic and expressive, however, was Ross’ break with the strict empiricism of Uppsala jurisprudence in his paradigmatic On Law and Justice (OLJ), originally published in 1953. In this work, Ross points out that, while the tendency – even in scientific disciplines (2019: 216) – to hypostatize legal concepts – such as ‘rights’ – granting them a kind of substantive or metaphysical existence sometimes associated with a kind of “extrasensory power” (2019: 215) is unwarranted, since in reality legal concepts are nothing more than “tools for presenting normative modalities” essentially reducible to the complex layers of directives aimed at judges (2019: 205). Understanding such categories merely as illogical “mystical formulas” would be, on the other hand, directly incompatible with the very self-perception of those same judges regarding their object of labour and legal-moral conscience as participants in a certain legal culture/tradition (2019: 160).
Even more: Ross states that classifying those concepts as simply as “mystical” would be, in a vocal critique aimed especially at the radicalism of Lundstedt, the result of a “stubborn desire to attribute absurd ideas to others” and a “lack of skill or willingness to understand how others use such expressions to designate certain legal realities” – in a pathological egocentrism that actively limits and prevents the mutual understanding necessary to sustain the scientific community (2019: 225).
In this sense, in OLJ, Ross proposes the identification of Law – qua a valid object of observation for a positivist science – as the ideology present in the minds of a relevant part of the population, especially of the actors in the legal sphere and in particular the judges (2019: 46-47); thus being an interpersonal reality endowed with meaning and motivation, so as to enable the interpretation and prediction of certain social phenomena, since it is socially perceived as a normative “valid” parameter for the guidance and assessment of individuals’ actions (2019: 40-41). Ross indicates its judicial application as the parameter for assessing that very presumed validity within the scope of legal norms, consequently varying its degree of “scientific validity” – not under a requirement of absolute and complete verification, but based on a logical criterion of predictability of its empirical application, naturally dependent on its sources of normative creation (2019: 57).
On this basis, taking upon himself the task of constructing a science of law animated by the methods of observation and verification proper to the modern empirical sciences (2019: 471) and modulated by the rejection of the idea of an understanding of “validity” alienated from the world of facts (2019: 51-52), Ross conditions the validity of scientific propositions about legal norms on their necessary empirical verification (2019: 51-52). This circumstance, given the continuous character of legal operation (2019: 89), ends up being possible only through the combination of a behaviourist perspective – focused on the external and empirically confirmable attitude regarding the pattern of judges’ conduct – and a field of hypothesizing – focused on the factors that govern and motivate the normative ideology adopted by such authorities (2019: 88).
Ross would therefore adopt a proposal of an “empirical science of law” built upon the idea of factual confirmation of hypotheses concerning the legal ideology that motivates judges’ conduct (2019: 48-49). It logically rests on an attitude of hypothesizing the very notion and amalgam of motivating reasons related to what judges consider as “valid law” (2019: 55), especially if it is understood that, for such recognition, these authorities must adopt a disinterested attitude towards the standards that would bind them (2019: 67-68); presupposing, for that effect, an acceptance of the “minimal moral correctness” of the ideological structure of law (2019: 68-69), co-responsible and co-conditioning, alongside its coercive sphere, for its social maintenance (2019: 69-72).
A Probabilistic Theory of Law? Probably Not: Ross’ Letter to Eugenio Bulygin and His Definitive Abandonment of Logical Empiricism
As Jakob Holtermann highlights (2019: xxxviii), in OLJ Ross seems to adopt a distinctly Carnapian perspective, especially by building his proposal from the idea of “eliminating metaphysical notions, addressing the problem of the ‘nature of law’ through the detailed analysis of the concept of ‘scientifically valid law’ as an intrinsic element of scientific propositions about legal norms” (2019: 19), and taking “scientifically valid law” as the “normative ideology operational in the minds of judges, understood as socially binding, once effectively observed” (2019: 46).
But since the factual verification of what is judicially taken as “valid law” (2019: 48-49) would be precisely conditioned by its effective judicial application; a fundamental question remains: would this be a probabilistic theory of law?
Curiously, according to a letter sent by Ross to Eugenio Bulygin in 1966, apparently not:
It was with great pleasure, now on my summer holiday, that I read your article. I find the construction and writing of it superb, and I am inclined to grant most of the points you make. Please do note that I am in complete agreement with your analysis of the propositions on efficacy (‘if certain conditions…’). Perhaps mistakenly, however, I assumed that such sentences are predictions. My mistake, then, is more a matter of logic than of legal analysis. Also, in that exposition, I was still under the influence of logical empiricism, according to which the meaning of a sentence is determined by the conditions for its verification.
(Bulygin 2015: 288-289)
In such letter, Ross seemly ends up recognizing as erroneous the conditioning of the validity of scientific propositions about legal norms on their necessary factual verification through judicial application, leaving apparently only their suitability to be used for grounding a possible judicial decision as the sole epistemological parameter for that matter (Bulygin 2015: 288-289).
Ross’ abandonment of strict logical empiricism seems to become even clearer in his Directives and Norms (DN) of 1968; in which he precisely rejects the idea of a “view of the legal man” – a concept proposed by Kelsen and very much alive in Ross’ TRJ, Tû-tû, and OLJ –, claiming that such a perspective would not be a “reliable guide for the logical analysis of Law, since it is saturated with ideological concepts reflecting emotional experiences, devoid of any function of describing reality” (1971: 146).
Eventually – ironically, considering what he had originally developed in TLS from 1926, and in an apparent total departure from his previous works –, when addressing the issue of constitutional self-reference and the normative foundation for the amendment of articles regulating constitutional amendments in his On Self-Reference and a Puzzle in Constitutional Law (OSRPC) of 1969, Ross adopts a strict perspective of the “systemic validity of law”. This perspective is based on the understanding that the “validity” of norms necessarily derives from the “validity of the higher norms that regulate their creation” (1969: 1-2), a chain that inevitably must end in a basic, non-derived norm (1969: 2-3), from which – as an original systemic founding fact itself not demonstrable – the validity of the system’s norms is presumed (1969: 6-7).
Coming Up Next: Olivecrona’s Split and the Vienna Circle as a Movement with Concerns and Divisions Parallel to the Uppsala School of Legal Realism (and a Farewell)
The gradual metamorphosis of Ross (especially from Tû-tû onwards) towards what would be the scientifically appropriate use of certain legal concepts in a realistic and “non-metaphysical” manner (Ross 2019: 205), seems to have had some recognized appeal in Olivecrona. Indeed, in the fully revised 2nd Edition of his Law as Fact of 1971, Olivecrona also shows a certain hesitation in fully committing to the radical scepticism inherent in Uppsala Legal Realism, notably regarding the outright rejection of such notions by Hägerström and Lundstedt (Olivecrona 1971: 177-178).
In the fifth and final part of this “Very Scandinavian Saga”, we will turn to the parallels between the Vienna Circle and the Uppsala Legal Realist School regarding their concern with the construction of an empirically supported conception of science, as well as their developments and internal splits concerning the scientific acceptability of empirically non-existent legal categories, especially as developed in the aforementioned late work of Olivecrona.
This work was realized under a PhD Scholarship granted by the Lisbon Public Law Research Centre, funded by Portuguese national funds through the FCT – Fundação para a Ciência e a Tecnologia de Portugal.
My work on this blogpost results from efforts conducted within the Horizon Twinning project “Advancing Cooperation on The Foundations of Law – ALF” (project no. 101079177). This project is financed by the European Union.
SUGGESTED CITATION: Bueno Assalin, Bruno: “A Very Scandinavian Saga (Part IV)”, FOLBlog, 2026/01/12, https://fol.ius.bg.ac.rs/2026/01/12/a-very-scandinavian-saga-part-iv/