Semiosis & Sign Exchange

design for a subjective situationism, including conceptual grounds of business information modeling

Pieter Wisse

chapter 13

Suggestions for theoretical practices

Subjective situationism with its anatomy of meaning can serve as – part of – the axiomatic system of many separate scientific disciplines and professions. I believe it is even more important that this ontology invites tighter interdisciplinary integration. Let me call the pertinent sciences motivational,¹ rather than social. For their decisive characteristic is that (re)actions are motivationally caused. And the sign is the characteristic cause. Motivational sciences and professions therefore emphasize that “information is a difference that makes a difference” (Bateson, 1972).

My own primary concern is with the discipline and profession of business information modeling. It is a design discipline focusing on conceptual models oriented at development and use of information systems for complex business processes. As an epilogue, in this final chapter I apply the conceptual grounds whose design occupies all the previous chapters to the theory and practice of information modeling.

I don’t aim at comprehensiveness, at all. Informally and briefly I touch upon a limited set of what I consider as important aspects of the process and results of modeling. First of all I merely indicate at the practical promise I believe subjective situationism holds. It is both a highly compact ontology and supports large variety through it characteristic organization of variables. Secondly, I suggest some lines for further research on modeling concepts.

13.1 arena of interests

Stating an information system is a tool for sign exchanges is hardly original. Sign users apply it to assist them in engineering and/or observing signs. What really makes a difference from the traditional perspective on information systems, though, is that every sign is a request for compliance. A user who engineers a sign does so to exchange it with a user whose observation and subsequent interpretation is expected, by the original engineer, to lead to compliance with the interest(s) he has invested in the sign. However, the observer has his own particular interests that control his observation, interpretation and, ultimately, (re)action. As the engineer is different from the observer,² the engineer always entertains – again, his subjective measure of – uncertainty about how the effect matches the compliance he produces the sign-as-cause for. The observer’s actual reaction may range from being completely compliant from the point of view of the engineer to completely surprising. As a corollary, the observer is always equally subjectively uncertain about the actual compliance requested from him by the sign’s engineer.

Because signs are both engineered and observed using it for a tool, the information system may be conceived of as an arena of interests. All sign users who act – voluntarily or not, in whatever capacity, directly or at a certain remove, etcetera – as engineer and/or observer of a particular information system are its stakeholders by definition. What counts as a sign may range from the whole tool itself to what it helps to process at the most detailed level possible of information.

A comparison with the built environment shows that sponsors are often highly sensitive to so-called image and can easily be sold by architects on the ‘message’ of their new building as-a-whole. Likewise, some information technology is hotter or, dependent on the language game, cooler than others. I don’t want to compromise the generality of this treatise by naming some technologies that are en vogue at the time of its writing. The reader may, at any time, choose his own examples.

An advantage of the conceptual or abstract approach is that a much larger set of phenomena may be treated as information systems. For a language system in the sense of a so-called natural language is an information system, too. This treatise published as a book is an information system, etcetera, etcetera. Some systems can adapt as far as their informational content is concerned. In flexible cases even their structure can, whereas others are unchangeable.

The inverse relationship also holds between information system and language system, actually. An information system is always a language system, but one not confined to natural languages, of course. It follows because, in its dynamic appearance, a particular information system allows for characteristic expressions of requests for compliance. A (more) stable information system, like this publication, simply is an already completely fixed characteristic expression.

A paradox I point at regards discipline. It seems fixed expressions invite a wider range of interpretations. Isn’t that what makes classical art, precisely, classical? Or is it still a paradox because it has so far been considered too difficult to create dynamic systems that are open to sensible multiple interpretations just as – more – fixed expressions already are?

This last chapter with my final remarks should be taken as ‘only’ applying the conceptual grounds presented in this treatise to modeling information systems that are tools for complex business processes. And ‘only’ such tools are assumed as being both constructed and operationally used through application of digital information and communication technology. This fundamental convergence of digital technologies of information and communication should now be evident from portraying an information system as an arena, or marketplace, for stakeholders. Because an information system is essentially a tool for sign exchanges it is just as essentially a communication system.³ The consequences of conceptualizing communication at the level of dynamics of stakeholders’ interests certainly deserves close attention.

13.2 essence of difference

Traditionally, the concept of an information model refers to the descriptive embodiment of the so-called shared meaning of stakeholders. Through their participation in the modeling process and/or based on the model delivered as the result, all persons involved directly or indirectly with the information system are believed having arrived at identical understanding. That is, they have reached agreement on structure and contents of a particular part of their world. The agreed-upon model is a descriptive synthesis that is next considered a useful, even valid, prescription for the system’s construction. Actually, that is the very purpose of the model-as-specification. In such cases the modeling process is undertaken for achieving of radical convergence, leading to identity, of stakeholders’ meanings that might originally be different. What stakeholders are supposed to learn during the process is to mutually adjust their meanings. Ideally, again, at the end of modeling all stakeholders hold identical meanings, i.e., shared meaning, only. It is of course acknowledged that convergence is not required outside the scope of the information system.

The elimination of differences is experienced as progressively difficult as the number of stakeholders increases. This is easily explained from a Schopenhauerean perspective. For there are principally no two stakeholders, including their intellects, identical. And even when shared meaning would somehow be a theoretical possibility, it must be taken as practically elusive.⁴ Modeling should therefore not at all be oriented at achieving shared meaning. Instead, emphasis must be on the differences between stakeholders. What does not change is that the model is a prescription for construction.

The relevant knowledge of a particular stakeholder may be, very simplistically, depicted by a circle. Suppose there are three stakeholders with completely disjunct knowledge, as shown in Figure 13.2.1.

Figure 13.2.1.
Stakeholders with disjunct knowledge.

Even this case, which is extreme when considering shared meaning, can be described within the boundaries of a single, overall model. The technique of the metapattern with its prominent visualization (see Chapter 4) suits the purpose. It only takes every stakeholder to be separately modeled as a separate highest-level situation. Next, the particular knowledge of every stakeholder is simply symbolized by the corresponding intext. Figure 13.2.2 gives an overview, first for the three stakeholder instances of the current example, next more generally for types.

Figure 13.2.2.
Information model of stakeholders with disjunct knowledge: instance and type version.

Without the overriding requirement of shared meaning, differences need no longer be synonymous with incompatibility. Actually, respected differences often point at opportunities, precisely because they are compatible. For several persons operating as a group usually are compatible precisely because they are different.⁵ It is only the incompatible differences that really need to be sorted out.

Taking appropriate inventory is assisted by broadening the scope from each stakeholder’s knowledge to his overall individuality, i.e., to a unique objectification of the will. What are his relevant interests? How does he conceptualize the situation(s) where he believes such interests are relevant? What does he, pertaining to his interest-driven action, consider his – more detailed – objectified reality to be, that is, his subjective worlds, situation by situation?

Starting from differences, and mainly dependent on the (dis)similarity of their experiences,⁶ the individual models – representing the interpretants – of stakeholders more or less overlap. See Figure 13.2.3. From the perspective of the anatomy of meaning of this treatise I suggest it is useless to speculate whether or not such overlap indicates, after all, a proportion of shared meaning. Each stakeholder will only hold subjective interpretants, anyway. What is apparently similar are only exterior models, that is, the signs, of – some illusive subset of – their interpretants, not the interpretants themselves.

Figure 13.2.3.
Overlapping models by stakeholder.

In this particular example, as can be read from Figure 13.1.3, there are seven different stakeholder groupings with corresponding submodels. The real number of stakeholders of any complex business process is of course prohibitive for such a theoretically encompassing approach.⁷ In professional practice, modeling is mostly done incrementally. Whatever is already modeled at earlier steps is often not ascribed to yet another stakeholder (but, of course, it always can when need be). Usually deserving attention at every incremental step is what is still missing in the overall model as it is drawn from the previous step. Special attention deserve possible incompatibilities between differences. There are many options to resolve them, ranging from settlement by consensus, to settlement by power, to dismissal of one or more stakeholders, etcetera. Or perhaps the very idea of a combined information system needs reconsideration. Options are always contingent upon the situation of the change toward the new information system.

As information modeling is synonymous with classification, “one of [the] central arguments” offered by G.C. Bowker and S.L. Star in their book Sorting Things Out: Classification and Its Consequences (1999) clearly indicates its problematic nature. They write (p 196):

[C]lassification systems are often sites of political and social struggles, but […] these sites are difficult to approach. Politically and socially charged agendas are often first presented as purely technical and they are difficult even to see. As layers of classification system become enfolded into a working infrastructure, the original political intervention becomes more and more firmly entrenched. In many cases, this leads to a naturalization of the political category, through a process of convergence. It becomes taken for granted. (We are using the word naturalization advisedly here, since it is only through our infrastructures that we can describe and manipulate nature.)

I cannot stress it enough that overlap naturally results from orientation at differences. It is simply the application of aggregation to partial models that, after all, are similar enough for the purpose of the information system. The reverse approach, however, doesn’t have this guarantee of necessary and sufficient completeness. When shared meaning is presupposed as the result, differences are simply not allowed to enter the model. Whatever starts as an aggregate can never be properly disaggregated, i.e., differentiated into necessarily partial models. The false assumption of shared meaning explains many failures of information systems. When using ‘his’ tool a user often discovers that his particular interests and corresponding requirements for sign engineering and/or observation are not supported. Without the paradigm shift toward the sign as individual request for compliance, those problems remain endemic to information systems.

Except for information systems with special requirements for accountability, it is likely that in the overall model mention of specific stakeholders, and their particular interests, is avoided.⁸ However, what always must remain is the essential recognition of differences. And it is perfectly compatible with the interest-driven nature of every stakeholder that he will concede more overlap when he feels his very own interests acknowledged. Thus a traditional paradox quite naturally disappears. For the result of respecting more autonomy by the, say, coordinator of the participants often is greater appreciation of his coordination by participants.

13.3 enneadic dynamics

It might be raised as an objection against fundamental recognition of, and follow-up on, differences that their number is practically overwhelming. I argue that in such cases denial of real differences undoubtedly frustrates corresponding interests, and thereby stakeholders. When an information system supporting relevant requisite variety is thought impossible to develop and/or to maintain, it must be judged an inappropriate tool in the first place.

As I have already suggested in the previous paragraph, an incremental approach to modeling secures overview throughout. For every additional difference is structurally integrated into the overall model before undertaking the next modeling step. In practice, at the start of each step it is mainly a matter of convenience to consider the originating stakeholder as occupying a completely separate situation of interests, etcetera. The particular sequence of steps of course influences the modeling process and the resulting model. It follows from the irreplicability of semiosis (see also my argument in § 1.12).

Taking a difference really seriously may every time result in a partial, or even complete, reorganization of the model.⁹ Suppose the behavior of an additional – type of – object is largely similar to that of an object, or object type, identified earlier. The overall model is then kept compact by adding a level of classification, making for example bicycles and airplanes each a kind of vehicles. The metapattern allows, on the other hand, to apply such reorganization locally, i.e., as pertaining to a particular situation. So, in one situation, for example bicycles and airplanes are taken as similar while in other situations they may be kept different as before.

In general, metapattern-based models support any focus imaginable. Then some particularly narrow foci might be combined. Or what has earlier been modeled as a high-level focus might be decomposed to make room for more differences at lower levels. What counts as situation, object, and behavior, is all relative. A model should support such flexible interpretation onto focus, background interpretant, and foreground interpretant because it allows for relative positions of signature, context, and intext. Especially the visualization with the metapattern technique supports those enneadic dynamics.

Focusing on a particular signature within the overall model, its intext may be discussed relative to its context. In other words, the ground of an intext consists of its unequivocal context, to which it is ‘connected’ through a unique signature. The model-as-sign thus refers to a particular situation as the ground of an particular object’s particular behavior.

Any metapattern-based information model may be seen as a networked collection of signatures. Every signature is the starting point for enneadic dynamics based upon the configuration of nodes. Dependent on the process instance of sign use it figures in, a particular node can serve as signature, or as – part of a – context, or as – part of an – intext. Another advantage of a model as a compact collection of, say, multipurpose nodes is that different process instances of sign use (also read: interpretation) most likely don’t run in parallel. They ‘connect.’ Such intellectual interference is beneficial for acquiring overview by any individual stakeholder.

The question is whether complete overview is necessary or even possible. Is it not sufficient when every stakeholder recognizes his own interests in the overall model? Is the rest of the model not essentially irrelevant to him?

Indeed, a stakeholder with little empathy may concentrate on the part, only, of the model which reflects his narrowly defined interests. But a stakeholder with wider interests (also read: a wider horizon for space and time) undoubtedly feels invited to learn, through the overall model, about other stakeholders. Any comments he might raise on contributions other than his own should preferably, at least initially, not be taken to convince another contributor to relinquish differences but rather to become even more aware of them. The grounds of argument are, ultimately, individual interests. For they always drive requests of compliance. Instead of the vain pursuit of joint grounds in arguments, participants should supportively challenge each other about their individually different grounds. It is only when such differences are clear enough that they can be judged compatible, or incompatible. Of course, those judgments are once more subjective. Power is therefore a key variable of the equation determining the choice when incompatibility is interpreted to occur.

Again a paradox dissolves. The chances of cooperation are enhanced when stakeholders are motivated to learn about each other, rather than from each other. Especially when differences are brought into play, and when they are compatible, the ‘system’ equals more than the sum of its parts. It is an evident principle practiced by every top coach in team sports.

A stakeholder attempting to learn about one or more other stakeholders will find it useful to abstain from an absolute conceptualization of objects but, rather, treat an object as a – what I call – boundary concept. For one and the same object may show different behaviors, depending on the situation it finds itself in as the subject’s objectification. Now the stakeholder is familiar with the situations he himself contributes as engineered contexts to the overall model. How the particular object is known to him is depicted as signature, with its behavior subsequently modeled as intext. The signature may be connected to one or more other signatures, each referring to the same object in a different situation. He is thereby led to change his focus while starting from what is still familiar and, by definition, corresponding with his interests. From a secure background,¹⁰ he more easily develops new interests. When less agreement is required among stakeholders, it all the more realistic to expect that they increase an understanding about each other.

Actually, the very terminology of stakeholder is well-suited for my anatomy of meaning. Their relationship is primarily characterized by the difference between their stakes, or interests.

13.4 trusted representatives, leadership, etcetera

Every stakeholder may contribute important differences to the overall business information model. However, in complex business processes it is practically impossible to consult all stakeholders. For example, changes in stakeholdership occur. And more and more, information systems cater to open communities. Earlier stakes may dissolve while new stakes are introduced. An example of the latter is government with new regulation.¹¹ Or tomorrow’s, say, customer could be still beyond the business horizon today. This raises the problem of selection. Who are the individuals actually consulted for modeling? What criteria apply for their selection?

The traditional approach is to elect representatives of so-called organizational functions. The underlying assumption is that all employees of for example the department of After-sales Support – even a priori – hold shared meaning. It narrows the problem of representation down to finding the person who best articulates their ‘meaning.’

This assumption is not valid from the perspective of my anatomy of meaning. A particular stakeholder may not at all feel his interests are best represented by a direct colleague. Without direct influence on the modeling process, he especially needs to trust his representative. And it could very well be that an employee doesn’t trust especially his colleagues. They often are, after all, also his competitors, prone to thwart his interests.¹² As far as his interests in the work situation are concerned, more often than not the employee in question will have relationships with persons outside his functional domain that he trusts more. But does one of them qualify for the purpose of the particular modeling exercise?

The predominance of trust emphasizes the political nature of planned change toward getting a new information system operationally used. What politics also suggest, of course, is that the trusted person not always acts in the interests of whoever invested that trust. Again, it is hardly surprising. For every person behaves egoistically, that is, with sometimes regretfully narrow boundaries for empathy. Politicians, say everyone who is in the ‘business’ of representing other individuals, are no exception.

I don’t offer clear-cut proposals for the problem of stakeholder representation. At the present stage I merely point at the risks of uncritically applying functional criteria.

On the concept of trust I remark it actually requires the assumption of differences between persons. Otherwise it doesn’t make sense. For suppose persons completely share meaning, why would they also need trust in their relationship? It really doesn’t serve any purpose under the assumption of identity.

Because shared meaning doesn’t exist, trust helps persons instead to establish and maintain their relationship despite their different interests. One person trusts another person when he believes the other will include him through empathy. It is about respecting and taking into account his interests. The degree of trust reflects the horizon of empathy attributed to the other person in the relationship. Taking a simplistic one-dimensional view for an illustration, the spectrum runs from complete trust to complete distrust.

Trust is an essential ingredient of compliance. Many signs are therefore specifically engineered to, first of all, solicit trust as the compliant reaction to it. When subsequent signs request compliance by the observer that are clearly against his interests, with compliance only promoting the interests of the engineer, it amounts to an abuse of trust. For likewise, it is their mutual differences which participants – aim to – exploit when resorting to power.

Precisely because individuals are different, every individual first of all needs to invest relationships with trust and power to have relationships favoring his own interests, at all. As a consequence of individual differences it is practically inevitable that some of every individual’s trusted relationships are abused through power. Such betrayal of confidence makes every person to some extent careful about necessary and sufficient investments in his relationships.

Many people raise their level of distrust when faced with – the situation of a particular – change to their lives. A majority of individuals usually accept change, not so much by the power of rational arguments, but by following – the power they attribute to – leadership.¹³

Trust and power are a badly neglected concepts in design, development and use of information systems. Emphasis on shared meaning may have contributed to its neglect. The radical, Schopenhauerean concepts of individuality and subjectivity I have presented as underlying subjective situationism including its anatomy of meaning should draw attention to the importance of trust and power, and of dealing with it responsibly, that is, with integrity. Anyone who invests trust and/or holds power expects it to be respected. Through the mechanisms of trust and power an individual holds expectations about the compliance with his interests by another person. When a participant abuses trust to secure commitment from other participants (also read: stakeholders) he actually applies power.

13.5 respectful application of power

Any person involved in modeling thereby enters into a relationship with others who also hold a stake in the new information system. Most obviously, he entertains a relationship with the system’s sponsor.¹⁴

When power – here again viewed simplistically for the sake of illustration – is only attributed a single dimension, the sponsor usually is the more powerful participant in the relationship. This asymmetry should never be ignored or even denied. For example, an employee whose contributions to modeling are subsequently opposed or ignored will surely feel his interests are damaged. As a result he is (even) more frustrated than before. When the sponsor doesn’t intend to grant whatever influence, it is more respectful to be clear about the reality of the relationship and its power distribution than it is to offer other stakeholders the pretense of control. When an employee can take-or-leave a new information system, without any of his own ‘differences’ being taken seriously, sooner or later he prefers honesty.

I don’t believe a sponsor acting with disregard for the interests of other persons is ultimately serving his own best interests. His lack of empathy is certain to make him miss out on opportunities of enlightened team work. Power merges with trust when stakeholders not only mutually recognize constructive differences, but actively promote them.

I loosely conceptualize leadership as precisely such synthesis of power and trust. It regretfully follows that sponsor- or leadership in improving information systems is badly neglected, too. A leader might be characterized as a broker of interests, including of course his own. His diplomatic skills allow him to build relationships others find ‘interesting’ enough to invest them with their trust, too. A leader also has, say, psychotherapeutic skills. By respecting autonomy of every other individual he provides for their freedom to be self-responsible for the management of expectations about fulfillment of interests. It helps other persons to develop, and solicit compliance with, interests that are as realistic as they can possibly be. Importantly too, the leader facilitates actual interest fulfillment.

It may happen that other stakeholders are unnecessarily apprehensive about one-sided application of power. The sponsor may in fact be offering an invitation to a trustful relationship, but it may be declined. In such circumstances it could help to let one or more, or even all, stakeholders make their contributions anonymously. Especially with an expert facilitator they all do trust, enough stakeholders might be sufficiently challenged to participate safely in their own interest, only. Later, the sponsor can use opportunities for showing that expectations of abuse are misplaced. Eventually a minimum level of trust is required between especially the sponsor and all other stakeholders to successfully design, develop and use an information system in complex business processes.

13.6 professional modeler

With some practice, a modeling technique such as the metapattern is simple enough to apply to isolated problems. Often though, and even regardless of the technique used, a relative outsider is necessary to make the modeling process successful for complex problems. For it is usually characteristic of complexity that many stakeholders with a wide range of possibly conflicting interests are involved. An outsider more easily recognizes and respects differences. He thus helps to resolve incompatibilities in an overall model.

A business information model itself only shows the result of resolution. With characteristic contributions to the process leading up to – a particular version of – the model, the facilitator seeks to bridge especially incompatible differences between behaviors of (other) stakeholders. When he comes to believe they are impossible to bridge he redirects his efforts at undoing one or more stakes in the business process.

A professional facilitator may face some dilemmas. The first is his relationship to the sponsor. Many sponsors can hardly be called enlightened. They often stop at genuinely granting authority to other stakeholders. How far their immediate coordination and control extends usually counts as their measure of success. In fact, such a sponsor is also his own obstacle. Real, enduring improvements and subsequent creation of further opportunities normally require an adjustment of how power is considered and applied. Again, a sponsor incapable of reconsidering his authority is his own biggest obstacle for improving information systems. When the facilitator comes up against such resistance, and cannot overcome it as a professional, he probably has no choice but to leave. His attempts so far at changing the climate for relationships between the sponsor and other stakeholders have no doubt damaged his own relationship to the sponsor.

Is withdrawal really the only viable option? Suppose the facilitator recognizes the need for a climate change but also discovers he cannot possibly make constructive contributions. Rather than leaving, he may elect to stay on and conduct the modeling process much to the sponsor’s traditional requirements. However, then the result may even lead away from improvements, making it all that much harder to achieve them at some later stage. At least the option of leaving at an early stage doesn’t compromise his relationships with stakeholders whose interests are subsequently frustrated during modeling and all that the model leads to. At some time in the future these participants may build upon the trust established through behaving with integrity. Or, are the other stakeholders now even off worse without his assistance? The facilitator will never be certain but has to act one way or another.

Another dilemma is that anyone worthy of being called a professional modeler, usually designs better models. But it is precisely because he acknowledges differences that he runs the danger of making those invisible in the resulting model. Non-professionals¹⁵ thus remain deprived of important opportunities for focus, and so on. The professional modeler may of course choose to present a less optimal model, relinquishing compactness, abstraction, flexibility, etcetera. The gain is in ready acceptance, the loss in quality.

Actually, I expect this particular dilemma to gradually disappear as business information modeling becomes accepted as a profession in its own right. It allows professionals to draw up their essential models which laymen don’t require to comprehend, too. This is well accepted in for example building architecture. The ‘model’ the architect shows to the sponsor is very different from the ‘model’ he supplies the constructors with. Actually, the sponsor-oriented model of an information system is traditionally called its prototype. However, the only kind of ‘prototype’ that is convincing is the finished information system. This indicates the need for a different approach to prototyping. Information technology directly supporting the metapattern technique just might give every stakeholder the realistic-enough impression his interests are recognized in such a manner as to warrant his trust in the operational information system as the finished result. A metapattern-based display suggests enneadic dynamics which is a closer match for any stakeholder’s experience with signs

I conclude this paragraph on the professional engaged in business information modeling with a few remarks on his essential characteristics.¹⁶ When sponsors lack leadership qualities, the professional modeler should be able to compensate for at least some of the qualities necessary for coordination. But what he as professional obviously doesn’t occupy is the position of formal organizational power. So he must use it to his advantage. His emphasis on personal autonomy and differences can now be authentic. At the same time, he must manage his relationship with the sponsor who is of course also his principal during the period of the professional modeler’s involvement. And as a specialist, the modeler must invest special effort to make himself accountable to all stakeholders.¹⁷

Some persons are different enough from others that, indeed, they qualify as professional modelers. However, I believe qualifications for professional success have not yet received the systematically social-psychological attention they deserve. First of all, both aspiring and practicing professionals themselves have a responsibility to assess their suitability. Secondly, the discipline of business information modeling needs repositioning, away from its traditionally technocratic bias. As I argue with this treatise, it is primarily a motivational discipline. Institutions of research and education should design their programs accordingly.

13.7 modeling as scientific discipline

In this treatise I have deconstructed the concept of shared meaning in favor of establishing the concept of the sign that is universally aimed at motivational inducement of compliance. This perspective may also provide inspiration to reflect on, and subsequently reorganize where opportunities are seen, scientific research and education. I argue for recognition of the importance of a radically subjective psychology. The so-called linguistic turn essentially propagates the modernist values of analytical philosophy. The psychoanalytic annex motivational turn of postmodernism must now succeed it. Subjective situationism is my proposal for an adequate ontology.

I have remarked above that scientists and professionals should be less technologically oriented. But it would indeed be counterproductive when they do so at the expense of understanding technology. In fact, its tool nature can only be properly understood when relevant situations of application are firmly grasped, too. It is the primary challenge to develop information science as a characteristic synthesis of motivational and technical orientations. It requires an eclectic approach.¹⁸

The discipline of business information modeling is conceivable as a high-level situation, as are many other disciplines. Next, business information modeling can ‘enlist’ whatever other disciplines are deemed relevant for it. Any such other discipline, but now in a different situation, is attributed different behavior. It pertains to its application in the situation of business information modeling, only. Figure 13.7.1 sketches this eclectic principle. It includes some examples of ‘related’ disciplines. For the sake of simplicity I have omitted that the discipline of business information modeling in return influences disciplines from which it borrows and for which it establishes a particular extension. Of course, a healthy scientific climate supports such continued cross-fertilization.

Figure 13.7.1.
Eclectic extension of scientific disciplines.

For this treatise I have adopted such an eclectic approach. It involves recurrent choices about the extent of research into another discipline. I propose the practice of due diligence. It is a concept I borrow from auditing. The researcher must exert sufficient effort in order for his results to be presented, as a sign, as a sincere request to observers to comply their objectified realities with them.

There should be no constraint on other disciplines investigated to promote development of the particular discipline of business information modeling.¹⁹ Removal of a priori constraints helps educational institutions to put an essentially interest-oriented curriculum into practice. It of course means taking the interests of students seriously. Can a student (still) be directed in developing the interests as the institution requires? Or should his (further) education be completely tailored to fit the interests he initially brings to the educational institution?

It is clear a particular student feels optimally motivated when his studies fully match his interests. But doesn’t a student also need specific direction for developing his interests? One way to keep education efficient is promoting the entry of students with similar interests. An educational offering may then be considered suitable for a whole group of students. But at some time during the course of his development the student must be offered an individualized program. Actually, he should optimally offer it to himself, with the educational institution acting mainly as coach.²⁰

Respecting differences and recognizing the essential individuality of students surely have a beneficial influence on research. Innovation and originality should be supported. A student pursuing even what at first looks an unpromising angle deserves benefit of the doubt. Given time, he may after all make a significant contribution. And when unsuccessful, he will also have learned from his efforts. It is in this respect equally necessary to maintain an individualized perspective on the institution. Its employees are individuals with particular interests, too.

As any situational object, business information modeling is only knowable as objectified reality. It is therefore interpreted differently by every individual. Its measure of success as a discipline and profession lies in the ultimately subjective illusion it supports of directed empathy and coordinated behavior. I believe subjective situationism – always fully respecting its essentially speculative nature – can promote different persons focusing on as behaviorally compatible illusions as possible of business information modeling.

notes

1. The proper attribution of different modes of causation is useful for classification of sciences. I believe it also to be more fundamental than, for example, the distinction H.A. Simon makes in The Sciences of the Artificial (1981). Underlying his concept of the artificial I recognize that of the sign, inducing motivational (re)actions. He seeks to explain how some sciences (p xi) “are concerned not with the necessary but with the contingent—not with how things are but with how they might be—in short, with design. The possibility of creating a science or sciences of design is exactly as great as the possibility of creating any science of the artificial. The two possibilities stand or fall together.” Actually, I don’t support the distinction simon starts from, that is, between “understanding the natural and artificial worlds.” All modes of causation, and subsequently their effects, are natural (or real).

2. Of course, the engineer and the observer may be one and the same person. But the distinction between engineering and observation remains useful.

3. What from a technological perspective are usually known as communications systems should more aptly be called transmission systems. For an early fundamental treatment from a semiotic perspective see The Meaning of Information (1972) by D. Nauta.

4. K.E. Weick already writes in Sensemaking in Organizations (1995, p 188): “Shared meaning is difficult to attain.” So, he “points to a different glue that can be attained. Although people may not share meaning, they do share experience. […] If people have similar experiences but label them differently, then the experience of shared meaning is more complicated than we suspect.” I even relinquish any suspicion about shared meaning. I believe it is a false – in the sense of behaviorally counterproductive – assumption altogether. Weick is not so consistently radical but he does stress the importance of equivocality, a concept he applies earlier in The Social Psychology of Organizing (1969, 1979). The same concept appears in Making Strategy (1998) by C. Eden and F. Ackermann. They place its usefulness, however, during facilitating, only. And its danger lies in (p 67) “backing away from clarity as clarity begins to emerge.” In Group Model Building (1996) J.A.M. Vennix is also tolerant of individual differences. His orientation is nevertheless at overall consensus, too (p 4): “The purpose is to support a decision making group in structuring a messy problem and designing effective policies to deal with it. [… p 5] A[n] important goal of the intervention should be to foster consensus within the team which, by the way, should not be confused with compromise. Consensus refers to unanimous agreement about a decision while compromise alludes to a settlement reached by mutual concessions.” My position is that relevant differences should be as unequivocally clear as possible. It is also that every individual always makes his own decisions. The concept of group should therefore be applied with care. And from the pervasiveness of politics in human relationships it follows that participants will settle for some degree of compliance with their individual interests. Then, consensus is merely a compromise where each participant does not feel any insurmountable loss of compliance.

5. The importance of compatible interpersonal differences for improved group performance is convincingly analyzed by R.M. Belbin in Management Teams (1981).

6. I don’t find it necessary to take a detailed position on the issue of nurture and nature. What counts for my exposition is that stakeholders have different biographies, for whatever reasons.

7. It can be easily computed for any number of stakeholders. With n stakeholders overall, the number of stakeholders involved in a particular model overlap may be denoted by p. The value for p ranges from 1 to n. The number of different combinations of p elements from a set of n is (n over p), or n!/(n-p)!p!. The exclamation mark refers to permutation. For example, 5! is the permutation of 5. It is the multiplication of the natural numbers from 1 up to 5: 1x2x3x4x5=120. The outcome of 0! is set by definition at 1. All different combinations from n, i.e., any subset, requires addition of the outcomes for every possible value of p. For n=3, the calculation is (3!/2!1!)+(3!/1!2!)+(3!/0!3!))=(3+3+1)=7.

8. This must not be confused with registration of users. For administrative systems it is standard practice for securing an audit trail. The auditor who has previously been involved in modeling an information system may not be explicitly shown in the model.

9. For specific models I refer to my books Aspecten en Fasen (1991), Informatiekundige ontwerpleer (1999) and, especially, Metapattern (2001).

10. The Russian psychologist and educationalist Lev Vygotsky (1896-1934) proposes two areas of intellectual development: actual and near. Actual development is accomplished independently by the learning individual. With proper guidance, however, the individual can learn beyond its actual area, that is, into his area of near, or future, development (J.F. Vos, 1977).

11. Government could equally well be an example of the former, that is, through deregulation.

12. On purpose I sketch a caricature. It brings out more clearly the risks involved when ignoring the individuality of interests.

13. I highly recommend E.M. Rogers’s Diffusion of Innovations (1962).

14. I refer to § 7.2 for an informal inventory of types of stakeholders involved in information systems.

15. Here of course, I only refer to stakeholders who are not professional information modelers. They may very well be highly professional in other areas.

16. In Informatiekundige ontwerpleer (1999) I pay special attention to the professional modeler’s preferred personality traits, as I see them that is. The radical recognition of subjectivity provides the ground for individual accountability.

17. See the second sentence of note 12, above.

18. I point at similarities with Die Modularität der Wissenschaft (1991) by András Kertész.

19. An example of an especially relevant discipline, but one that I have not at all treated explicitly here, is hermeneutics. A synthesis with semiotics under the heading of subjective situationism certainly looks promising. An overview presents K. Mueller-Vollmer (editor) with The Hermeneutic Reader: Texts of the German Tradition from the Enlightenment to the Present (1985).

20. See note 10, above.

table of contents; previous chapter; corresponding prelude.