(P0-12) Knowledge Is Local, Methods Are Local (1)

As we claim that both knowledge and knowledge-acquisition methods are local, we want to show how. We distinguished six scientific positions. Consequently this will become a long blog. Nevertheless we attempt to keep the thing short, and stick to the bare minimum for explaining where we stand. We do not enter into parochial debates on the philosophy of knowledge and take a crude look at the whole. We follow Aernout’s (A.’s) trajectory through the six positions at hand: legal positivism, logical positivism, critical realism, inclusive pragmatism, post-structuralism, story telling.

Legal positivism

It all started for A. in 1966, reading law at Utrecht University. At the time, Hart’s authority as a legal philosopher emerged and legal professionals turned to legal positivism rather eagerly, disavowing in the process any debt to morality and focusing completely on how to interpret valid laws for adjudication of actual situations.

The accomplishments of legal positivism are mainly in interpretation technology. Bobbitt’s work on how the American constitution has been interpreted throughout its history provides a clear enumeration of six interpretation techniques (historical, textual, docttrinal, prudential, structural, ethical — the inventory coincides to a great extent with what Scholten published some 40 years earlier on Dutch legal practice).

The strict locality of use for these legal-positivist techniques is in the interpretion of validly promulgated legal rules (legislation, case law, contract). The strict locality of validly promulgated law is the jurisdiction (often the nation state). The strict locality of resulting knowledge is in legal practices. Such knowledge tends to tally with cultural styles.

One of Hart’s main points is the distinction between types of legal rules (i) primary — addressing the organisation and bahavior of subjects and (ii) secondary — addressing the organization and behavior of legislative powers.

It is concievable  that useful techniques for interpretation of legal texts will spill over to other domains in which interpretation of authoritative formulations is required.

[Of course, as from any intellectual position, destructive oneliners may be derived from legal positivism. “Rules are rules” is a frequent example.]

Logical positivism

During his stay at the university as a student money had to be earned, and to that purpose A. became one of the university’s first computer programmers, in 1969. At the time computers were a new and scarce phenomenon. A few geniuses in applied mathematics provided expertise, permeated with and firmly anchored in logical positivism.

Paraphrasing Wikipedia, in this theory of knowledge only statements that are verifiable, either logically or empirically, are considered meaningful. This approach emerged in early 20th century and contaminated Anglo-American philosophy as its proponents sought refuge from the nazis outside the European continent.

Contrary to its initial ambitions, logical positivism proved to have serious problems in gaining wide applicability status, since empirical verification of abstractions can only work inductively and abstractions (or extrapolations) thus cannot at all be considered open to empirical verification.

And verification through logic requires complete and consistent formal systems, and thus remains restricted to the use of axiomatic formal systems (or artificial games) of a very simple kind.

As a corollary, logical positivism is ill-suited for much of the social sciences and almost all of the humanities. On the other hand, considering its focus on verifiable knowledge, logical positivism (with its preference for formal systems) provides a useful and natural position for logicians, mathematicians, statisticians and computer scientists.

The strict locality of use for logical positivism is dictated by the requirement of verifiable statements, and these are only produced in formal systems. The strict locality of its techniques is in modeling and specifying such systems. And the strict locality of resulting knowledge is in the interpretation their behavior (also discussed as artificial-game behavior in this blog).

Nevertheless the attraction of studying the behavior of formal systems has spilled over to areas where simulating reality with artificial games has become acceptible. This has happened quite early on in the sciences, where mathematical modeling has always been considered quite useful. So artificial gaming has gained considerable appeal in the sciences (especially where universal laws are supposed to exist, lab conditions can be imitated, or models can be used do predict unexpected results that can be falsified (or not) — as in quantum mechanics).

[It is useful to remark (a bit off-topic though) that the applied sciences can (and often do) use legal-positivist-like techniques for the interpretation of authoratative propositions that articulate the requirements for designing a formal system, model or game.]

One of the early critics of logical positivism is Karl Popper who suggested that although meaningful propositions cannot be verified, they can be falsified, and, if not, if falsification is impossible, that such propositions are not meaningful.  In the sciences this adaptation of logical positivism has been very successful and falsification has proved an effective tool there.

Critical realism

By 1987, A. had joined the Law department of Leiden University and defended his PhD research there. The project attempted to model (and thus explain) the dynamics in 10 years of Dutch legal practice in legal decision making on probation questions.

A. started out in a pure logical-positivist approach attempting to create a single formal game that would simulate real-world behavior of the legal policy makers. He discovered that separate adequate models could be designed for stable periods of decision-making policies, but that any non-linear transformation from policy model at time t into policy model at time t+1 (and the like) was impossible without violating the assumed requirements for designing a consistent and complete system.

The social sciences have to deal with non-linear policies and policy transformations as these assumedly follow (or at least depend on) human decision making. And, looking at political realities, it  is difficult or even impossible to predict such transformations in a complete and consistent formal game. Nevertheless, that is what the critical realist positions aims for.

This seems (and frequently is) a severe drawback for acceptable spill-over of logical-positivist attitudes and techniques to the social sciences and the humanities. The adoption of logical positivist habits nevertheless appears to be a deep-rooted ambition there. And indeed, much effort has been invested into turning the social sciences into “science.”

For a long time, the only social discipline that has successfully adopted into its core the formal modeling approach that is so characteristic of logical positivism is the mathematical branch of economics. There, formal exchage games are modeled on microfoundations (often the assumption that individual agents behave rationally) which are kept as simple as possible. Mathematical economics is good at describing economic trends yet bad at predicting critical transitions in macroeconomic conditions. This results in a quandary: how to react to real-world falsifications of formal econometric games?

This quandry is characteristic for all of the social sciences. The resulting ball park thus cannot be purely logical positivist because the falsification concept has lost its function as unbiased decision procedure for disqualifying a model: in economics, in the social sciences and in the humanities falsification has proven (and still proves) to be not straightforward at all. This observation requires a shift in perspectives (actually a whole cascade of shifts in perspectives), the essence of which is (i) to drop the requirement of knowledge to be absolute and (ii) to come to grips with living in a contingent world.

The perspective that emerged is critical positivism, because the concept of `real world’ reality remains used, yet the concept of model falsification becomes subjective and political. So critical positivism’s strict locality of use is in designing simple formal games and in relating them to stable (in-equilibrium) behavior of social phenomena under investigation. Quite often, critical realism is employed for modelling how civilians, civil servants and institutions can be influenced by non-scientific forces to ignore scientific, `objective knowledge.’ The main accomplishments of critical positivism (there is a huge collection of critical approaches to scientific research, all named and understood as being slightly different – we lump them together here) are to allow that political and social forces come into play, also in scientific research and emphatically so in applocation. And thus that the results of social sientific research may be used to promote the reception or dismissal of scientific work.

[Again, legal-positivist techniques will be used for the interpretation of propositions about (non-)falsification of formal models in critical-realist settings.]

Inclusive pragmatism

After his PhD, A. became a full-time law academic, focusing on ICT situations. These situations generally require multiple disciplinary perspectives to understand. During 2003/7 he looked at the roles of the law  brought to bear towards the emergence of file-sharing. Understanding this situation required at least additional expertise in computer science and in economics. Connecting the separate analyses proved a difficult job (see the fourth book listed at http://l-m-d.net/aernout/selected-publications-online/). Having faced these difficulties and realising that he had encountered similar difficulties while occasionally advising on public procurement proceedings, he jointly (with an experienced legal public-procurement counsel) wrote down his thoughts on how to get useful specifications from a cross disciplinary team for ICT services to be publicly procured (see the fifth book listed at http://l-m-d.net/aernout/selected-publications-online/).

We name the position adopted in these projects inclusive pragmatism, as it supports cross disciplinary cooperation in an iterative process that pragmatically employs the diverse insights on solving a service problem by design. And that is the strict locality of use of the inclusive pragmatic position. It is directed to useful comprehension of cross disciplinary cooperation. In USA legal practice (and theory) it is still finding its feet (see Goldberg 2012), and, to our understanding, it has not (yet?) emerged as a respectable academic position in the knowledge-acquisition arena — even if one cannot suppress the feeling that (rather intuitive, naive) inclusive pragmatism is currently all the thing in the practice of having to address new and complex situations and phenomena.

Post-structuralism (with a complex adaptive networked system twist)

These misgivings (intuitive, naive) urged A. since 2010 (having earned emeritus stature) to persue a search for a (to his mind) less naive and intuitive, yet promising position — which presented itself in the form of what e.g., Mitchel calls complexity theory and what Cilliers (2011) calls post-structuralism.

It is remarkable, that complexity theory (which emerged from modernist-positivist scientific approaches) and post-structuralism (which emerged from post-modernist deconstructivist scientific approaches) meet so emphatically around 2010. We cite Cilliers 2011 to support this submission:

Poststructuralism builds on the essential systemic ‘method’ of structuralism […]. The very structures which make meaning possible introduce a distortion in the system of relationships. These structures, sometimes called hierarchies, can therefore not be final, but are in constant transformation, both through external intervention and by their own dynamics […].

and claim that his characterization of the objects of poststructuralism are functionally identical to what we consider complex adaptive networked systems.

We think that the post-structuralist position clearly shows what has been missing in the four positivist positions mentioned earlier: it has eyes for what the sensors of these positions could not see or feel. Taking non-linear dynamics and non-equilibrium seeking, evolving ecologic multi-level systems behaviors seriously implies the acknowledgement that such behaviors are too difficult to model in a single overarching formal game that is complete and consistent.

So in order to gain any progress we will have to drop the requirements of overall completeness and consistency. One way to do this is to let collections of complete-and-consistent models (agents) interact and operate in and on their environments. Designing and looking at the behaviors of collections of models in an environment might be the only way to provide for further useful information.

To our minds, agent based modeling and simulations define the strict locality of use of the complexity (or poststructuralist) position. But, as we will not be able to claim completeness or consistency of such behavior, we will need an additional position for explaining what we see when we look at such simulated behaviors in aggregate. This will be the story-telling or narrative position.

Yet first we have to acknowledge that, in the evolution of notions in the pilosophy of knowledge, the different positons here have predominantly been discussed as rival. We do not immediately see a reason for such rivalry, as we cannot imagine that the philosophy of knowledge could be represented in a single formal system. We guess that we will need all different positions in parallel, even in cooperation. Our suggestions concerning localities of use attempt to accomodate this view. Elsewhere we will address the subject in more depth.

Story telling (narrative knowledge)

We can know what has happened. We cannot know now what will happen. And we often do not in fact know what has happened. And we often are capable to predict correctly what is going to happen. Apparently, we can create coherence in our thinking and in our comprehension by designing appropriate stories — and by telling them.

In unbrideled story telling, anything goes. It is only very recent, say around 2012, that A. brought himself to accept that story telling is essential in academics. Important requirements for scientific story telling are brought to  the fore by Aumann 1985. But that issue deserves more focused attention. Again, we will address the subject elsewhere in more depth.


Aumann, R.J. 1985. What is game theory trying to accomplish? Pages 28–76 of: Arrow, K., & Honkapohja, S. (eds), Frontiers of economics. Blackwell.

Bobbitt, Philip. 1982. Constitutional Fate: Theory of the constitution. Oxford University Press.

Cilliers, Paul. 2011. Complexity, poststructuralism and organization. The sage handbook of complexity and management, 142–154.

Geertz, Clifford. 1983. Local knowledge: Further essays in interpretive anthropology. Basic Books.

Goldberg, John C.P. 2012. Introduction: Pragmatism and private law. Harvard law review, 125, 1640.

Hart, H.L.A. 1961. The concept of law. Oxford University Press (Claren- don), London.

Popper, Karl R. 1963. Conjectures and Refutations: The growth of scientific knowledge. Routledge.

Scholten, Paul. 1931. Algemeen deel (derde druk 1974). Tjeenk Willink.

Print Friendly, PDF & Email

4 thoughts on “(P0-12) Knowledge Is Local, Methods Are Local (1)

  1. This is a test comment for to find out how registration through commenting works