Title

Resources for a Christian worldview

BasdenPC

Welcome

Newest Members

 

PRACTICALLY CRITICAL: MAKING THE CRITICAL APPROACH MORE USEFUL

Andrew Basden

Introduction

Critical approaches to information systems (IS) in business may be seen as a paradigmatic response to two previous approaches, which Jackson [1991] calls the hard and soft systems approaches. The hard approach was characterized by a desire for control, with information technology (IT) being the means to effect that control by making business processes more structured and reliable. The hard approach, in its extreme form, attempted to suppress individual human preferences or beliefs, and shared many fundamental beliefs and preferences with the scientific paradigm of logical positivism. The soft approach, epitomised by Checkland's [1981] Soft Systems Methodology, responded by emphasising subjective opinions, values and Weltanschauungen (world views). The soft approach, in its extreme form, denied any possibility of a reality that transcends humanity. It held that we create our own realities by orchestrating 'appreciations' of situations [Bergvall-Kåreborn, 2001].

There are two main problems with both the hard and soft approaches. One is that neither allow for any sound notion of right and wrong, and so gave no sound basis for distinguishing applications of IS that were 'good' from those that are 'evil' or detrimental. When forced to talk about right and wrong, as happens in practice, the hard approach speaks instead about costs and benefits, while the soft approach reduces them to personal preferences; both could be used, for example, to design a more efficient method of genocide. The other is that neither provides any basis for addressing the impact of social structures on the way IT has been developed and used. For example, Adam [1998] argues that IT is shaped to masculine values, and hence is unsuited to feminine applications, and Pacey [1996] that IT is shaped to Western values, and hence is unsuited to application in non-Western contexts. To the soft approach, structures are dissolved in inter-individual relationships while in the hard approach structures are reduced to immutable laws that govern life; neither provides a sound basis for critique of social structures.

The critical approach responded to these by espousing both normativity and societal structures. Two main versions have arisen, both of which give priority to a major norm which is meaningful in terms of social structures and which should guide all our application and development of IT (and indeed the rest of human life); they are 'critical' in the sense of assuming a normativity by which to criticise existing social structures and advocating changing them. In the first, the major norm is that of emancipation: social structures oppress or constrain people and IS should be designed and used in such a way as to overcome this. In the second, the major norm is open communication and discourse by which we come to understand each other perfectly and as a result engage together in 'rational' social action: various factors, including power structures, distort our ability to communicate openly and this results in irrational social action.

However, both of these approaches are problematic, especially when employed in practical situations of everyday life, and in applications outside the professional arena, such as social networking and virtual worlds. The emancipatory critical approach is unable to usefully define emancipation. While it enabled us to critique the 1970s view of IS, of people's lives dominated by large, inflexible organisational computer systems, it is not clear what 'emancipation' means in computer games. The communicative critical approach tends to assume that all social activity must involve rational discourse and critique, and this blinds us to other types of social activity. While it has been useful in analysing business processes of the 1980s and 1990s, in which rational communication is paramount, it is not clear how it can be applied to such applications as social networking today. We need to be critical about the critical approach.

This article examines the nature of the problems that each version of the critical approach has, uncovering problems and inconsistencies at various levels. It then suggests a way to both resolve the inconsistencies and address the problems, to make the application of both emancipation and communicative action easier in everyday IS research, design, development and evaluation. It does this by reference to philosophy. Philosophy is usually seen as a highly abstract discipline, with little direct relevance to practice in everyday experience, but here it will be demonstrated that philosophy not only has direct relevance, but can point the way to practical tools.

The Problem of Emancipation and Power

Emancipation was a central thrust of the early Frankfurt School, partly explainable by its Marxist lineage. It is the major normativity of this branch of the critical approach; it is the 'Good' that overcomes the 'Evil', which is oppressive social structures and their results. The word was originally most closely associated with the issue of slavery, and then became applied to the oppression of workers. Emancipation, in this original sense, has a strong social element, and arose as a major norm because of oppressive social structures, not just because individuals acted in an evil manner. As such, the norm of emancipation transcends individual perspectives and values. Emancipation then became associated with suppression of the rights of those affected by actions and decisions outwith their control. Though still retaining a strong social aspect, it became applied to the indirect impact of inter-individual action.

The notion of emancipation is sometimes associated with, and has often been replaced by, that of empowerment in some discourses [e.g. Allen et. al., 2000; Markus and Bjørn-Anderson, 1987]. There is a difference in connotation: emancipation speaks of the action by one person or group to remove oppressive conditions from another, while empowerment speaks of the action by a person or group to remove oppression from themselves. Walsham [2001] is one example of many who use Foucault's notion of power as a lens through which to view IS use. He does so at several levels, individual, organisational and societal and understands power, not as accorded by status, but as a type of relationship we have with others, which is intimately linked with knowledge. Walsham demonstrates that this understanding is more useful in analysis because relationships are more dynamic and variable than status. As an example, to which we refer below, Walsham [2001:91] comments on how a plotter engineer did not use the IS provided by his company, with "Gary was able to draw on his deep knowledge of plotters to leverage power over Comco". Feminist thought also makes use of the notion of power, but as invested in the more static structure of gender. An example is found in Haraway [1991:196], who views rational knowledge as "power-sensitive conversation".

In relation to IS, the notion of emancipation (or empowerment) leads to 'emancipatory information systems development' (EISD) [Hirschheim and Klein, 1994], to 'participatory design' [Berg, 1998], emancipation as a major driver of management, including organisational use of IS [Alvesson and Willmott, 1992; Tsoukas, 1993]. One concern is how the development and use of information systems entrench social structures that oppress people, and the notion was first applied to IT workers of the 1970s, and (e.g. by Ulrich [1994]) to any who are affected by IT without their having any input. More recently, for example, Ngwenyama and Lee [1997] have applied it to email and Cordoba and Midgley [2003] have applied it to IS planning.

Conversely, IS (including the Internet) can be hailed as a means to emancipation. This contrasts with hard systems thinking, in which IS is seen as a means to efficiency or control, and with soft systems thinking, in which IS is a means to self-expression or self-fulfilment.

But the notion of emancipation (in which we will include that of power unless otherwise stated) exhibits problems and perhaps has proved less useful in practice than had been hoped. Several reasons can be adduced for this. One is that the focus of early emancipation-oriented IS research on workers oppressed by large, centralised IT is of diminishing interest today. Since the 1980s other issues have become much more significant in the use of IS, partly because of the advent of the personal computer, the home computer, and applications like computer games, music, art or social networking. While emancipation might be useful in considering organisational IS, it is less useful for these other types of computing. There is a diversity of types of IS application for which the notion of emancipation has little meaning. Likewise there is a diversity of IS problems that cannot reasonably be treated as emancipation, such as ease of use, affordance, learnability, attractiveness, accessibility, or benefits in use. To do so rather stretches the meaning of 'emancipation' too far. We need a critical approach that, in itself, admits a variety of problems and applications.

A second reason is that emancipation tends to focus on the negative, on the 'evils' from which we should be delivered, rather than on the positive, on 'good' already in the situation or that which IS might be able to bring. Though it could be argued that such good may be treated as emancipation from constraints, this rather distorts the experience of computer users, especially when considering new possibilities. For example, in using a paint package to construct an icon, I do not feel emancipated from inability to construct icons, so much as enabled to use new ways to construct them.

A third reason is perhaps deeper than these two: emancipation lacks useful definition, relying instead on an intuition of its meaning being shared by us all. Emancipation has become almost synonymous with 'avoidance of problem', becoming a generic super-norm, a general Good over against a general Evil. As such, hardly ever is it questioned or its nature explored, so it gives no hint as to what might be considered evil. From what should we expect to be emancipated? What means can we use? How do we recognise the need for emancipation? How do we know when it has been achieved? To none of these practical questions is a good answer forthcoming in most literature. Many books espousing a critical approach fail to include emancipation in their indices. A notable exception is Ulrich [1994], where we find 23 index entries, as well as a number of indirect references, and he lists eight things from which we might be emancipated. However, he gives no basis for selecting others, nor does he discuss the other questions. Perhaps a definition of emancipation might be found in Habermas [1972], to which he frequently refers as introducing the idea of emancipatory interest. When we turn to Habermas [1972], we find no index entry of emancipation, and again its meaning is assumed rather than explained. So we must discuss briefly here what emancipation is and how it might apply.

In some cases what emancipation means is reasonably clear and its importance as a norm is uncontentious, for example an over-weening management, which demands that employees sacrifice their home and family life for the organisation. In many other cases, it is far from clear. It could be defined as 'freedom from unwarranted constraints'. This definition seems initially useful to us in that IS are very good at embodying constraints (e.g. as IF-THEN rules) or facilities that help users overcome constraints. however, this merely shifts the need for clear understanding to three other concepts: 'freedom', 'unwarranted' and 'constraint'.

The fact that something is a constraint does not necessarily imply freedom should be sought. For instance, gravity is a constraint, but since we can do nothing about it, we tend to accept it, and oftentimes use it to our advantage. One might reply that 'nature' is to be excluded, and that the emancipatory paradigm only applies in human social situations. Yet even here these notions are less than helpful, because there are many human 'constraints' that we might find arduous and yet are good. In writing managerial reports or writing poetry, for example, we might be expected to use a certain style of writing. Should we be emancipated from the demand for such a style of writing; is this an unwarranted constraint?

Perhaps emancipation should be restricted to issues that are more important than language style (though workers' strikes have been called for less), such as issues of justice? Unfortunately, we then meet another problem, that one person's constraint is another person's justice. For example the norm of justice to workers in the two-thirds world is a constraint on our purchasing of goods ('fair trade'), the norm of justice to future generations calls us to radically change our (Western) lifestyles so as to curb climate change, and the norm of justice to animals suggests that we should not allow more experiments on animals for cosmetics? When are such constraints warranted, and when unwarranted?

These examples illustrate a number of things, which we also find in our everyday experience. One is that it seems necessary to recognise there are fundamentally distinct types of warrant or constraint, and should respond differently to each. If it is to be useful, the notion of emancipation must be enriched, to acknowledge different types of warrant, constraint and freedom. Emancipation can no longer be assumed to be a single, undifferentiated norm.

Another is that constraint and freedom are intertwined. Constraints on us might be required by the freedom of others, and our freedom might impose constraints on them. As Alvesson and Willmott [1992:448] put it, "Critique and liberation from old dogmas is then followed by new dogmas." The relationship between constraint and freedom can become more nuanced than this, however. The attempt at emancipation from one 'evil' might lead to also losing a 'good': "women emancipating themselves from dominant socialization patterns and gender roles may reduce their interest in, and capacity for, caring" (p.447). If we assume that constraint and freedom are mutually exclusive of each other, then we end up with paradoxes of this kind.

Wilson [1997] discusses one particular paradox within emancipatory IS design (EISD) at greater length, what might be called the paradox of 'enforced emancipation'. His argument is that though we are supposed to be emancipated from positions that constrain us, to judge whether a system is emancipatory or not presupposes a position from which to make such judgements. He asks (p.15)

"How do the proponents of EISD know that what they urge is in the service of all mankind and not merely a function that they themselves happen, at the moment, to desire?"

He points out (p.20) that,

"the use of facilitators to ensure 'that everyone contributes and is listened to' takes on slightly menacing overtones when it is suggested that their deployment of 'emancipatory methodology' will be used to overcome 'wilful unresponsiveness by an individual'."

It may be that the notion of empowerment is less afflicted by this paradox than is emancipation. It does exhibit its own problems, however. Like most lenses, the power-lens tends to determine what we see and often makes us blind to factors not explainable in its own terms. That it can distort analysis and misrepresent a situation is illustrated in Walsham's [2001] study of the plotter engineer, Gary, mentioned earlier, as "leverag[ing] power over" his company. However, if we examine Walsham's actual report of Gary's working [p.68-9], we find not a hint of power leverage. Instead, we find someone who believes in the company, feels closely associated with it, and defends it. Gary is willing to go beyond the call of duty by, for example, keeping his own stock of spare parts and his own log book instead of relying on the IS, because he wants to give a good service to the customers. Gary's generosity is 'reinterpreted' by the lens of power as its very opposite, self-interest. It is not uncommon to find a blindness to generosity and self-giving love in power-centred literature. For a more detailed discussion of Walsham's approach, see Basden [2008]. A similar blindness to these aspects may also be found in some feminist literature that likewise focuses on power relations, such as those that sought emancipation from male hegemony or empowerment for women during the 1970s to 1990s.

However, a more nuanced version of power-centred critical approaches, including feminism, seems to have emerged around the turn of the century. Walsham [2001] himself recognises that other conceptual tools are needed, and cites Knights and McCabe as finding that those affected are not so concerned with an issue of power so much as maintaining professional self-identity. The feminist writer, Adam [1998], argues for a more sophisticated form, eco-feminism, which seeks to restore aspects of reality that have been overlooked by 'masculine' assumptions when technology was developed rather than focusing on emancipation or power. Artificial intelligence in particular has been developed according to masculine assumptions of the supremacy of logic and mind over intuition and body. In her conclusion [p.180] she says, "The way that a number of aspects of knowing are not reducible to propositional knowledge, but rely instead on some notion of embodied skill, points to the role of the body in the making of knowledge." We will return to these more nuanced versions later, and first consider the root of the above problems in emancipation and power.

The Root of the Problem: Dualistic Ground-motives

These problems with emancipation and power may be seen as arising from a deeper presupposition. Eriksson [2006] argues that the three main systems thinking approaches, hard, soft and critical are all expressions of a deep presupposition known as the nature-freedom ground-motive (NFGM). 'Ground-motive' denotes a "spiritual driving force that acts as the absolutely central mainspring of human society" [Dooyeweerd, 1979:9]. A ground-motive governs thinking in society for centuries about the nature of reality (what constitutes it), diversity (what types there are), normativity (what is good and evil) and how to obtain the good (how to solve problems and what is ethical conduct). Dooyeweerd discussed four main ground-motives, the form-matter motive, which governed the thinking of the ancient Greeks, the creation-fall-redemption motive, which governed that of the Hebrews and early Christians, the nature-grace motive, which emerged as a synthesis of these and governed the thinking of mediaeval Europe, especially Scholaticism and Roman Catholic thought, and the nature-freedom motive, which has governed thinking from the Renaissance to this day. (Habermas [2002] also recognises the form-matter and creation-fall-redemption ground-motives under the names 'Athens' and 'Jerusalem'.)

Three ground-motives, those of form-matter, nature-grace and nature-freedom, are dualistic in nature. Down the centuries when a dualistic ground-motive is in force, theoretical thought swings dialectically between the two poles. At any time in a given community of thought, one pole is seen as real and 'Good' (to be sought), while the other is seen as unreal and 'Evil' (to be avoided).

The nature-freedom ground-motive leads thought to assume a mutual antithesis between the principles of nature and freedom: that what is controlled or operates by deterministic laws cannot be free, and vice versa. Disciplines like management, politics and IS have been experienced a swing from the nature pole (around 1970s) to the freedom pole (1990s). Though in everyday experience both freedom and control are interwoven, our theoretical thinking (via science, philosophy, etc.) presuppose either a determined cosmos, or one constructed by an absolutely free human ego, and have no way of truly integrating both {FOOTNOTE fn-cr}. The influence of NFGM may be seen in many of the common dualistic oppositions of today, such as nature-culture, fact-value, rationalist-irrationalist, science-humanities, modernism-postmodernism, positivism-interpretivism. Thought influenced by one pole finds thought influenced by the opposing pole meaningless or inferior, as exemplified perhaps by the attitudes of each of positivist and interpretivist researchers, respectively.

Eriksson argued that hard systems thinking is centred on the nature pole (systems are objectively 'out there' and the aim of hard systems thinking is to control), and soft systems thinking is centred on the freedom pole (systems are a matter of subjective interpretation, and the aim of soft systems thinking is to allow interpretations to surface). The critical approach attempts to embrace both poles (systems are socially constructed but, transcending this, they impact on people and are subject to the norm of emancipation). Laws that transcend our freedom (such that they govern us whether we recognise them or not) have traditionally been recognised only in pre-human nature (physics, biology), and thus the notion of something that transcends us has strong connotations of the nature pole of NFGM. Emancipation is therefore interesting: the word itself suggests the freedom pole, but that it transcends us suggests the nature pole. This is the root of the paradox.

Dooyeweerd [1984], on which Eriksson based his thought, argued that the dualistic ground-motives of Western thought like that of nature-freedom are deeply problematic and will always end in such paradoxes since they force apart two aspects which in reality are intertwined (especially as that reality is experienced in everyday life). Dualistic ground-motives hide rather than reveal the structure of reality, including normative reality. To understand everyday experience requires a way of understanding that can embrace both poles of each dualistic ground-motive, but this cannot be done by any approach that is within the confines of the ground-motive, because it presupposes an absolute antithesis between them. A dualistic ground-motive forces our attempts at understanding to focus on one of its poles or the other. If they attempt integration, they will be found initially to contain an antinomy, such as 'enforced emancipation' [Wilson, 1997]), which can only be resolved by eventually moving to one pole or the other. Marxism gave priority to the nature pole of NFGM; Frankfurt-School critical thought gives priority to the freedom pole.

Dualistic ground-motives are also problematic in the sense of hindering critical thought, in that both poles and also their presumed opposition are protected from critical interrogation. We noted earlier that precisely this has happened with emancipation. Since freedom and constraint are seen as aligned with opposing poles, there has been little attempt to seek to understand the different types of constraint or freedom. 'Unwarranted' is largely meaningless under both poles of the NFGM because under the nature pole it is reduced to a mere label for deterministic processes and under the freedom pole it is a label for value-judgements; the reason it appears in the definition of emancipation is because we already have an intuition of what it means, outwith the confines of the NFGM. 'Warranted' speaks of 'due', of doing justice to something, but neither the nature nor the freedom poles offer a sound basis for recognising this.

To summarise,

  • We need to expand the critical approach so that it acknowledges the 'good' as well as criticises the 'evil'.
  • We need a way of revealing more clearly the nature of emancipation, as diverse and as relevant to twenty-first century applications of IS.
  • We need an understanding of reality that allows both freedom and constraint or control, which can resolve paradoxes like 'enforced emancipation'.
  • We need a way of acknowledging power that does not obliterate other aspects like generosity.

To make further progress requires moving to a non-dualistic ground-motive; this is discussed below.

Contribution from Habermas: Communicative Action

A second critical perspective, centred on communication, has become important over the last two decades, based on the thought of Habermas. Habermas, while initially emphasising emancipation [1972], re-focused on communicative action as the deeper and more important issue [1984, 1987]. Communicative action is rational social activity that seeks mutual understanding between people via open dialogue. Habermas contrasted it especially with instrumental action, which seeks to control others and closes down dialogue. Open (or 'ideal') dialogue involves discussion and critique both of the content of what is said, by processes of argumentation, and of its claims to validity (which Habermas classified as intelligibility, propositional truth, sincerity and social and moral appropriateness). It is assumed that the force of the better argument always wins, so rationality is an integral part of communicative action. Various things distort open communication, including lack of information, subjective bias, unwarranted assumptions, conflicts of interest, power relations or unjust social conditions. The ideal therefore is communication that is free of all such distortions; is a counterfactual ideal, though Habermas holds it up as a useful aspiration.

In Habermas' theory of communicative action (TCA), ideal dialogue replaces emancipation as the major norm. It has close links with emancipation, in that it is believed to lead to emancipation from oppressive conditions because it enables all participants to express their views without distortion, and it itself is emancipated from distortions by critique of validity claims.

"Communicative action" says Habermas [1987:120] "relies on a cooperative process of interpretation in which participants relate simultaneously to something in the objective, the social, and the subjective worlds". In this cooperative process, participants draw upon "a reservoir of taken-for-granteds, of unshaken convictions" [1987:124], which Habermas calls the lifeworld, a concept he adapted from Schutz and Luckman [1973], who in turn adapted Husserl's [1970] concept of 'life-world'. The lifeworld was to Habermas a mysterious thing that dissolves as soon as we try to take it up with our thinking [Ray, 1993]. We can think and talk about things in the three 'worlds' of objective things, inner feelings and social/moral norms, because we direct ourselves towards them from the outside, as it were, but our relationship with the lifeworld is that we orientate ourselves within it [Habermas, 1987:126].

Normativity in TCA is complex. The lifeworld is a stock of shared understandings that includes meaning and norms. Habermas argued that these both guide communicative action and also emerge intersubjectively from critique of validity claims. Normativity is, therefore, not ignored, as in the hard approach, nor reduced to subjective values, as in the soft approach. At first sight, TCA admits of no transcendental normativity, of the kind emancipation was in the earlier critical approach, in that all norms emerge by the processes of communicative action. A closer examination reveals, however, that there is a transcending normativity, a deeper norm which pertains regardless of what actually emerges from communicative action, which is the norm of communicative rationality or ideal dialogue, the norm of sound argumentation and critique of validity claims.

Habermas' theory of communicative action has obvious relevance to IS and has been applied both directly and indirectly. It is used directly, for example, by Lyytinen and Klein [1985] to form a theory of IS, and by Heng and de Moor [2003] for understanding and evaluating practice on the Internet, on the grounds that Internet use might be seen as discourse. It is used indirectly via the Language Action Perspective [Goldkuhl & Lyytinen, 1986, Weigand, 2006], which incorporates Habermas' ideas. The Language Action Pespective has found use in analysing and modelling business processes and business conversations, such as that between supplier and customer [Denning and Dunham, 2006], Goldkuhl, 2006], and in making a proposal for new research directions in data and knowledge engineering and natural language processing [Basden and Klein, 2008].

There are problems in using TCA, however, which will be discussed in order of increasing depth. One problem is whether ideal dialogue (to the extent that it is feasible) will in practice lead to emancipation from oppressive conditions, as has been assumed, and occasionally argued in some of the literature surrounding TCA. This belief rests on the assumption that language as employed in argument is sufficient to capture all meaning in life perfectly and fully. This assumption may be questioned, even in the ideal, and comes about as a result of the following problems.

A second problem concerns the notion of ideal dialogue. In proposing that all issues may be resolved (in principle) by rational argument, where the 'better argument' will prevail, it depends on there being a single universal type of rationality. But Winch [1958] argued for multiple rationalities. "In science, for example," suggested Winch [1958], "it would be illogical to refuse to be bound by the results of a properly carried out experiment; in religion it would be illogical to suppose that one could pit one's strength against God's." He went on to argue that the rationality of the social sciences is unlike that of the physical. Practical life turns up many examples of different rationalities, for example, when financial indicators suggest one course of action while legal or moral reasons suggest another. While we can expect seamless reasoning within a rationality, we cannot do so between different rationalities. Any attempt to argue from one to the other involves category errors. In decision-making, when a choice must be made between them, that choice is supra-rational. There is no 'higher' rationality that tells us which type of rationality should prevail over the other. This means that TCA's notion of the 'better argument' prevailing is inadequate {FOOTNOTE fn-hbwinch}. While it might help prepare the ground within each rationality, it cannot ultimately help us make choices between rationalities. TCA's call for openness of dialogue must be reinterpreted to include supra-rational choice between rationalities. The hidden prejudices or presuppositions that distort communicative action are not always a problem of rationality but can be a denial of other virtues like honesty and goodwill.

This leads to a third problem. The focus on communicative rationality fails to recognise that it is frequently very difficult to express what is relevant to the situation being experienced. On the one hand, much knowledge is tacit, and on the other, some people are less able to express their views and thoughts than others are (for example because of Asperger's Syndrome). So, TCA-based proposals that we should aim to improve expression of relevant ideas and views in open dialogue merely favours the already-articulate and that which is easily expressed in words.

A fourth problem is related to this. There are some types of social activity in which argument or critique of validity claims as such is not appropriate. Most attempts at applying TCA have been in business decision support, where rational communiction is presumed to be central. Recently, Internet applications relying on such types of social activity have burgeoned, such as social networking, social virtual worlds and computer games. Users of applications like Facebook 'poke' each other or send 'growing things' and 'snowglobes' to each other for fun, not to make business decisions. Though this might be described as communication, it is not governed by rationality (not even of multiple types). It is for the purpose of enjoyment and fun, not to enhance mutual understanding (communicative rationality), nor even to control (instrumental rationality). The same applies to computer games and social virtual worlds, and many applications outside the professional arena. Even within the professional arena there are many applications in which communicative action is not primary, such as physical simulations, mathematical calculations, route-finding software and computer music and paint packages. Recognition that critical researchers should look beyond professional applications is also reflected in Fraser's [1989] feminist critique of Habermas, cited by Ray [1993:68]: she emphasises the household. That we should also look beyond the communicative aspect is echoed in her emphasis that women's unpaid child-rearing is a 'dual aspect activity' that involves a 'material' aspect and cannot and should not be reduced to the single communicative aspect.

Root of the Problems in TCA: Reduction

The root of most of these problems may be traced to a tendency to reduce all aspects of life to a logical-communicative aspect, though the nature-freedom ground-motive discussed above also plays some part. Reduction can take several forms, including methodological reduction, in which we narrow our focus in order to simplify research, and value reduction, in which only a limited range of aspects has importance for us. These are usually unobjectionable as long as they are not taken too far and we are always aware of other aspects outside our focus. However, there are types of reduction which Clouser [2005] argues are objectionable, not only in their practical consequences but also philosophically. He identifies four main types, two strong and two weak. The weak reduction that Clouser calls causal dependency pervades much of Habermas' TCA. Clouser explains [2005:359],

"Causal dependency. The nature of reality is basically that of aspect X (or of aspects X and Y). It is the Xness of things which make possible the other kinds of properties and laws true of them. So while other aspects are real and can be proper objects of scientific investigation, there is a one-way causal dependency between the non-X aspects and aspect X. The non-X aspects could not exist without X, while X could exist without the others."

The X aspect to which TCA reduces others is the logical-communicative aspect. Most kinds of social action are made possible by X (Habermas [1991] explicitly reduces at least dramaturgical and normatively-regulated action types to the communicative). The entire lifeworld (with its diverse types of meaning and normativity) could not exist without X. Though Habermas does not go to the extreme in reduction, TCA does exhibit a tendency to reduction. We will first look at the extreme version and then draw back from it.

Clouser argues that reduction is objectionable both philosophically and in practice. It is objectionable philosophically because reduction is self-performatively inconsistent, by which Clouser [2005:85] means:

"a theory must be compatible with any state that would have to be true of a thinker, or any activity the thinker would have to perform, in order to have formulated the theory's claims."

This means that to formulate TCA as a theory of social action Habermas (or any other thinker) would necessarily function in aspects that cannot be reduced to the logical-communicative in the way assumed possible under TCA. By 'aspects' Clouser is referring to basic kinds of properties or laws that we experience and function in pre-theoretically, among which he includes such things as faith, justice, morals and aesthetics, as well as logicality and communication {FOOTNOTE fn-hbvision}. According to Clouser, this criterion for theories is relatively new, having been first defined and deployed by Dooyeweerd about fifty years ago, but since then advocated independently by other thinkers like Foucault. Clouser's argument that reduction is self-performatively inconsistent [2005:192 ff] is complex and need not be rehearsed here.

What concerns us more is that reduction is objectionable practically. Clouser explains [2005:187] "it lowers the status of all other aspects by making them products of, and thus less real than, the aspect(s) it favors". This is what has led to some of the problems identified above. The fourth problem arises from TCA assuming that logical-communicative dialogue should normally be central, and that other types are either negative (for example instrumental action) or need not be given the same amount of attention as is communicative action; in Clouser's terms, those types of action are "less real". As will be seen below, each aspect defines a distinct type of rationality; that TCA has difficulty in recognising types of rationality other than the logical-communicative (and the instrumental which it treats as deficient) is explainable by causal dependency reduction, in that it assumes that the other rationalities arise from its favoured one.

Clouser can take us further, to address the first and third problems above. He argues that reduction, when taken to an extreme, accords a status of absolute self-dependence to the favoured aspect (he uses the term 'divinity' to denote this). This is what leads to the assumption lurking at the root of the notion of ideal dialogue, that language use is, in principle, able to capture all meaning, leaving nothing tacit and that as a result all emancipation will be overcome by rational dialogue. Clouser [2005] and Dooyeweerd [1984] claim this is a false assumption.

However, Habermas does not in fact go to the extreme in reduction, in that he recognises instrumental action as distinct from communicative action, and he recognises that there are some things that cannot be couched in dialogue: the "stock of taken-for-granteds" that is the lifeworld. His recognition of instrumental reason as a distinct aspect is, however, only partial. In seeing it as an 'evil' that distorts communicative action it is thus defined mainly in relation to communicative action, rather than as a truly distinct aspect. One might also detect the influence of the modern ground-motive of nature-freedom here, which opposes control (instrumental action) and freedom (from distortions resulting from instrumental action). Though Habermas recognised the tacit nature of the lifeworld, this sits uneasily with the notion of ideal dialogue, because if we cannot talk about the contents of the lifeworld there is much that is omitted from our dialogue and hence it is no longer ideal. TCA offers no way to allow the contents of the lifeworld, including its diverse meaning and normativity, into our dialogue without distorting them.

To summarise,

  • We need to counteract reductionist tendencies in TCA, so that it can give due regard to the non-rational types of action in which communication might occur but is not of primary importance.
  • Especially, we need to recognise multiple rationalities and provide a way to facilitate supra-rational openness, such as involving honest and goodwill.
  • We need a way to give due respect to difficulties in expressing thoughts and meanings while not abandoning rational dialogue.
  • We need a way to be able to bring the contents of the lifeworld into dialogue in a way that does not overly distort them.
A basis for meeting these requirements, as well as those of the emancipatory approach, may be found in the thought of Dooyeweerd.

The Contribution of Dooyeweerd: Enriching the Critical Approach

Dooyeweerd's major work, A New Critique of Theoretical Thought [Dooyeweerd, 1984], first published as four volumes in 1953-1955, shows Dooyeweerd to be both a critical and a positive philosopher. He both deconstructed Western philosophy and then accepted the challenge of constructing something that could itself be criticised. The latter, known as Cosmonomic Philosophy or Philosophy of the Law-Idea, should not be seen as a wholesale replacement for Western philosophy, but as a means of reforming it. Here we will review a little of Dooyeweerd's critical approach, before briefly looking at sufficient of his positive proposals to allow us to address the problems above. A fuller explanation is available in Basden [2008] or Clouser [2005], and a critical discussion of how Dooyeweerd's thought 're-integrates' social theory, in Strauss [2006].

Dooyeweerd contended that we cannot start, in philosophy, by presupposing the possibility of taking a theoretical or critical attitude of thought, but must first establish what makes it possible. An extensive immanent critique of 2,500 years of theoretical thinking revealed that it is never neutral, because what we take to be 'theoretical' is itself determined by our religious presuppositions. By 'religious' Dooyeweerd meant an ultimate commitment about the nature of reality, of good and evil, and of knowledge.

The nature-freedom ground-motive, mentioned earlier, is one such religious presupposition, which makes the antithesis between freedom and control absolute, with the result of shielding both from critical examination. In Dooyeweerd, such polar antitheses are questioned and he allows us to probe the nature of freedom and control. This can overcome the paradoxes inherent in the emancipatory critical approach.

Dooyeweerd held that all theoretical thinking, including his own, rests on presuppositions {FOOTNOTE fn-khi}. Rather than seeking presupposition-less thought, we should make our presuppositions explicit. He made his own clear: the creation-fall-redemption (CFR) ground-motive. His positive philosophy might be seen as a philosophical (not theological) exploration of what follows if we presuppose the cosmos is created. This is not a return to the mediaeval dominance of the religious over the secular as occurred under the nature-grace ground-motive, but an exploration of implications for philosophy. Presupposing CFR enabled Dooyeweerd to tackle diversity and coherence and to begin with the lifeworld; see Basden [2008] for discussion of this.

So at the very start of Dooyeweerd's New Critique, we find, "If I consider reality as it is given in the naïve pre-theoretical experience ..." [1984,I:3]. He invites his readers to do likewise, to adopt a pre-theoretical rather than theoretical attitude of thought. In the theoretical or critical attitude, we 'stand over against' some aspect of reality (Dooyeweerd used the term 'Gegenstand'), whereas in the pre-theoretical attitude, we engage with all the aspects instead of standing over against them. In the theoretical attitude, the chosen aspect becomes, to us, a distinct 'world' seen, as Habermas' would say, from the 'outside'; Habermas' three 'worlds' (objective, social, subjective) might correspond approximately to three aspects of reality which Dooyeweerd delineated (see the analytical, social, psychic aspects below). In the pre-theoretical attitude, we are usually unaware of the aspects - echoing Habermas' notion of being 'inside' the lifeworld. A pre-theoretical attitude to reality accommodates all the aspects thereof whereas a theoretical attitude separates aspects from each other conceptually. The aspects resist such separation, however, because each is inextricably intertwined with the others, so that what theoretical thought delivers must never be treated as 'truth'. It was perhaps this resistance that Habermas observed when he said that the lifeworld dissolves when we try to think about it.

The notion of aspects is not unique to Dooyeweerd as we have already seen in Adam [1998]; Habermas also uses the term [1987:126]: "... singles out above two aspects of ... the teleological aspect .. the communicative aspect ...". Most thinkers speak of aspects when they want to make distinctions among things that cannot be reduced to each other. Thinking aspectually is so natural that we usually take it for granted.

Dooyeweerd, however, addressed the nature of aspects, their intertwinement and the possibility of separating them out conceptually without undue distortion. He was convinced that the aspects are not mere categories, but have a modal character, as spheres of meaning and law. This enabled him (and us) to do what Habermas was hindered in doing, namely to probe the contents of the lifeworld in such a way that it does not dissolve in the process. (Dooyeweerd never used the term 'lifeworld', but it is equivalent to his notion of pre-theoretical attitude of thought.) The challenge is to determine what different types of meaning and law there might be. This cannot be achieved by theoretical thought alone, because of the Gegenstand relationship prevents us experiencing the lifeworld from the inside. However, the kernel meaning of each aspect may be grasped with the intuition in a way that does not involve theoretical thought {FOOTNOTE fn-int}. This can help us distinguish the types of meaning we encounter in everyday life; for example, we might be aware of justice or frugality (see the list below) as we engage in life. Once this has occurred theoretical thought can be employed to make various checks, for example for possible antinomy. From a lifetime of sensitive reflection (together with some theoretically-based checks, described in Basden [2008]), Dooyeweerd delineated fifteen such aspects:

  • Quantitative aspect, in which amount and quantity are meaningful
  • Spatial aspect, in which continuous extension and space are meaningful
  • Kinematic aspect, in which flowing movement is meaningful
  • Physical aspect, in which energy, matter, forces are meaningful
  • Biotic aspect, in which life functions are meaningful
  • Sensitive aspect, in which feeling and response are meaningful
  • Analytical aspect, in which distinction is meaningful
  • Formative aspect, in which creative power, instrumentality and technology are meaningful
  • Lingual aspect, in which symbolic signification is meaningful
  • Social aspect, in which social interaction and structures are meaningful
  • Economic aspect, in which frugal management of resources is meaningful
  • Aesthetic aspect, in which harmony, delight, fun are meaningful
  • Juridical aspect, in which what is due (rights, responsibilities) is meaningful
  • Ethical aspect, in which self-giving love is meaningful
  • Pistic aspect, in which faith, vision, commitment are meaningful.

Each aspect is irreducible to all the others in respect of its meaning. So, for example, to speak of self-giving love is 'meaningless' when thinking economically or quantitatively, and vice versa. However, all aspects are equally important in principle, so one cannot say that the meaning of one aspect has a priori priority over that of another, nor can any be set aside. Because of this, analysis should normally consider every aspect; analysis that ignores an aspect distorts our view. Aspectual irreducibility is the key to defending ourselves against reductionist tendencies.

The laws of the earlier aspects (e.g. physical law of gravity) are determinative while those of the later aspects (e.g. lingual laws concerning syntax, semantics, pragmatics) are non-determinative in the sense of allowing some freedom. The laws are normative in the sense of providing guidance for what is 'Good' in the widest sense (aspectual law is not an authoritarian command - "Do X, or else!" - so much as a promise - "If you do X, then Y will tend to result"). In this way freedom is consistent with the idea of transcendental normativity, as required by the critical approach. Aspectual normativity must not be confused with social norms; the latter are socially constructed while the former pertain regardless of us and though we might be able to go against them we cannot escape the repercussions of doing so. For example, the syntactic laws of any actual language are socially constructed, but their very existence reflects the aspectual law that syntax is important and beneficial.

As spheres of law, the aspects provide a suite of basic types of 'Good' and 'Evil' that are irreducibly distinct, defining what is 'warranted' or 'unwarranted'. That which is in line with the norms of an aspect is warranted from the point of view of that aspect. Table 1 gives examples of this for each aspect (in the most determinative aspects 'evil' is not possible since it is impossible to go against their laws).

Table 1. Aspectual Good and Evil
Aspect Good ('Warranted') Evil ('Unwarranted')
Quantitative Numeric order, precision (determinative)
Spatial Simultaneity, extension (determinative)
Kinematic Movement, duration, change (determinative)
Physical Persistence, causality, unity of whole cosmos (determinative)
Biotic /
Organic
Health, vitality, thriving Disease, death
Psychic /
Sensitive
Sensitivity, responsiveness, recognition, memory Insensitivity, inability to respond, sensory deprivation, forgetting
Analytic Clarity, non-contradiction Obfuscation, false reasoning, inconsistency
Formative Achievement, planning, construction, structure Laziness, confused plans, destruction, messiness
Lingual Understandability, expression of meaning Poor pragmatics, semantics, syntax, lexics, etc.
Social Respect, friendliness Disrespect, enmity
Economic Frugality, management, prosperity Waste, excess, mismanagement, poverty
Aesthetic Harmony, interest, surprise, fun, playfulness, leisure Uniformness, boredom, over-seriousness, over-work
Juridical Justice, responsibility Injustice, irresponsibility
Ethical Generosity, self-giving Selfishness, self-interest
Pistic /
Faith
Vision, faithfulness, commitment Low morale, disloyalty, refusal to commit, hidden agendas

This can give crisp, diverse meaning to 'emancipation'. When seen as freedom from unwarranted constraints full emancipation occurs when the only constraints we experience are warranted in every aspect. In any concrete situation the constraints upon us might be warranted within one aspect (e.g. the economic) but unwarranted in another (e.g. the juridical). So emancipation becomes, not freedom from constraints in general, but modification of the aspectual structure of constraints such that they become warranted in all their aspects. This helps us, in practical situations, decide which constraints we can accept and which, not. This view also provides a basis for analysis by which we can recognise the 'good' in the situation, so as not to jeopardise it.

Is this possible in principle? Or might some aspects inherently work against others (for example, will being ethical always, inherently jeopardise prosperity)? Van der Kooy [1974:40-41] argued that simultaneous realization of norms (of every aspect) is possible in principle, because there is no inherent conflict between norms of different aspects if one presupposes the creation-fall-redemption ground-motive. This gives a hope and vision to aim for in IS development. It does not preclude, however, conflict between concrete (socially constructed) norms, such as the requirement mentioned earlier of a certain writing style for reports.

The paradox of enforced emancipation as Wilson [1997] introduced seems at first sight to be a clash between freedom and control (as seen under the nature-freedom ground-motive). But does not all application of emancipation to others involve some control? All attempts at emancipation could therefore be deemed 'enforced'. Such a view is untenable. So we need a better basis for differentiating when emancipation is wrong from when it is valid. This may be provided by reflecting that, in this context, 'enforced' is not merely formative functioning, but it involves at least an attitude that transgresses the norms of the ethical aspect, namely one of unconcern for the other.

To the power-oriented critical approach, Dooyeweerd can offer both affirmation and critique. By asking which aspects are most important in power, we find the formative aspect (control), then the juridical (concern that rights are being suppressed by such control and should be reclaimed by gaining control), and then some of the pistic (in its link with identify [Walsham, 2001]). This reveals that power ignores certain aspects, such as the ethical aspect of self-giving which, as mentioned above, is distorted into its opposite when viewed through the lens of power. The utility of Dooyeweerd's suite of aspects is that it can act as a multiple lens, making every aspect visible and thus precluding such distortions. It is especially useful in conjunction with (eco-)feminism to reinstate aspects overlooked my a 'masculine' worldview (see Basden [2008]).

Now we can turn to communicative action. As a sphere of meaning, each aspect defines a distinct way in which things can 'make sense', or not - a distinct type of rationality. Aspectual irreducibility explains why the rationality of one aspect is irreducible to that of another. The two types differentiated in our earlier quotation from Winch [1958] are the physical and pistic rationalities. Instrumental and communicative rationalities, as differentiated by Habermas [1984], may now be seen as those of the formative and lingual aspects. In this way Dooyeweerd already recognises multiple rationalities, and challenges us to find supra-rational ways to bring rationalities together in the situations of dialogue that we face. Ultimately, Dooyeweerd argued, such decisions have religious roots (in the sense of ultimate commitments) and it is good if these religious roots can be laid bare.

The aspects can assist us towards open dialogue among multiple aspects in three ways. Aspectual analysis of situations provides a basis for disentangling distinct basic kinds of meaning. Secondly, it was suggested earlier that open dialogue involves other virtues than rationality, such as honesty and goodwill; these can now be seen as 'good' in aspects other than the analytic. Thirdly, some distortions of dialogue arise from undue commitment to one aspect at the expense of others. Analysis of which aspects people think are important can expose this.

This has practical relevance for IS, because it would suggest that if an IS funnels its users' thinking down one path of reasoning, however good, then it provides a disservice rather than a service. In the case of an Electronic Placement System for insurance [Walsham, 2001], usage of the system remained low for many years because of its focus on the economic aspect, which overlooked what was more important in insurance trading, namely the pistic aspect of Utmost Good Faith. This was despite offering benefits in terms of communication (lingual aspect) and time-savings (economic).

Human activity is multi-aspectual in the sense that it involves functioning in all aspects, but in many human actions one aspect is more important than others (the 'qualifying' aspect). This can open up Habermas' notion of action types approximately as follows:

  • instrumental action: qualified by formative aspect
  • communicative action: qualified by lingual aspect, with strong analytical flavour
  • strategic action: qualified by social aspect, with strong formative flavour
  • dramaturgical action: qualified by aesthetic aspect
  • normatively-regulated action: qualified by juridical aspect.

Since aspects are fundamentally irreducible to each other and all are of equal importance, if this view is valid, then all action types are of equal importance in human life, and it can help TCA avoid reduction. It can also perhaps suggest other action types not discussed by Habermas - such as ethical and faith-based.

That all human activity is multi-aspectual means that we will find, for example, some lingual functioning (i.e. communicative action) in almost all activity. To call this communicative action, however, is misleading when the qualifying aspect is other than lingual. This makes it necessary to understand how aspects are intertwined in human activity. One type of inter-aspect relationship is dependency, and this can explain the apparent reduction of dramaturgical and normatively-regulated action to communicative. Action in any aspect depends foundationally on action in earlier aspects, and also depends on later aspects to gain its full meaning (anticipatory dependency). Social action, for example, cannot occur without lingual, but conversely lingual action has very little meaning if it is not carried out to enable social action. Likewise, economic, aesthetic, juridical, ethical and faith-based action depends foundationally on social, and thus on lingual action {FOOTNOTE fn-calf}. Thus dramaturgical and normatively-regulated action (aesthetic and juridical) depend foundationally on communicative action. However, they are not reducible to it, because they involve other things that do not make sense in terms of the lingual or analytic aspects alone, such as surprise and restitution. Habermas did not differentiate foundational dependency from reduction in the way Dooyeweerd did.

Instrumental action is, under Dooyeweerd, not wholly negative, but has a valid place alongside communicative. Habermas' insight that instrumental action distorts communicative action may be accounted for by the imposition by us of one aspectual norm (formative) on action that should be guided by another (lingual). It is not instrumentality as such that is the problem, but our attitude to it. Dooyeweerd's insight into this enables us to extend TCA to see the imposition by us of any aspect on activity that should be guided by another as detrimental. This frees us from imposing communicative action on all software design or use and allows us to recognise that Facebook 'pokes' are meaningful in the aesthetic aspect of fun rather than the analytic aspect of argument or the formative aspect of control.

Dooyeweerd's thought can, therefore, account for some of the problems encountered in emancipatory and communicative critical approaches, but also ameliorate the problems and enrich the approaches. How it does so is summarised in Table 2.

Table 2. How Dooyeweerd enriches critical approaches
Issue in critical approach Dooyeweerdian enrichment
Should acknowledge the 'good' as well as 'evil' Aspects define types of 'good' as well as 'evil'
Nature of emancipation Aspects provide diverse normativity
Paradoxes, e.g. enforced emancipation Move away from nature-freedom ground-motive and Consider ethical aspect of self-giving
Power-lens obliterates certain aspects Give due regard to all aspects, especially that of self-giving
Reductionist tendencies in TCA Recognise multi-aspectual nature of human action
Open dialogue with multiple rationalities Each aspect defines distinct rationality; considering aspects ensures no aspect overlooked, and exposes hidden agendas
Problems in expressing ideas (Not explicitly addressed)
Dialogue including contents of lifeworld Intuitive grasp of aspectual meaning

Situating the Critical Approach

As we noted earlier, Eriksson [2006] argues that hard and soft systems thinking express the nature and freedom poles of NFGM, and see themselves as opposing each other. Critical systems thinking tries to embrace both poles but will ultimately be unsuccessful in this and migrate to the freedom pole. This gives a picture of the three approaches being always incommensurable with each other. But is this correct? In good practice, for example in IS development, we find elements of all three.

According to Dooyeweerd [1984, volume I] the opposition between nature (control) and freedom is false, when viewed from outwith NFGM. So the presumed incommensurability between the approaches might also be false. If this is so, then on what basis may we differentiate between them? One answer is in terms of which aspects each deems important. Table 3 attempts to show this for each type of systems thinking. Hard approaches focus on the quantitative, analytic, formative and economic aspects, insofar as they try to quantify and conceptualise an objective reality and control it with reference to costs and benefits. Soft approaches focus on the sensitive, social and pistic, insofar as they pay attention to what participants feel, to their social roles and their Weltanschauungen. Critical approaches focus on the social, juridical and lingual aspects, insofar as they recognise the social and societal element of IS, and either promote emancipation or ideal dialogue. Feminist critical approaches add the biotic aspect of the body (as an antidote to overly mental approaches) and power-oriented approaches add the formative. This analysis is not meant to be exhaustive, but rather to provide an overall view that shows approaches as complementary rather than antagonistic. It also might suggest there are other aspects which none of these approaches give much attention to, such as the aesthetic and ethical.

Table 3. Aspectual focus of systems approaches
Aspect Hard systems approach
(Instrumental)
Soft systems approach
(Subjective)
Critical perspective
(Emancipatory or Communicative)
Quantitative Quantification
Spatial
Kinematic
Physical
Biotic /
Organic
(Feminist: body)
Psychic /
Sensitive
Feelings
Analytic Requirements analysis
Formative Planning, control, achieving goals (Power)
Lingual (Documentation) Communicative action,
Ideal dialogue
Social Group awareness (Intersubjectivity)
Economic Resource management
Aesthetic
Juridical Oppression,
Emancipation
Ethical
Pistic /
Faith
Perspectives,
Weltanschauung
(Power: identity)

Some Reflections

This article has examined some of the difficulties encountered in trying to turn the critical perspective into practical guidance for IS research and practice. It has to some extent blurred the boundaries between the two main versions of the critical approach, namely emanciptory and communicative, and given attention to some issues at the expense of others. Nevertheless, it offers a critical affirmation and enrichment of both. Problems in the two versions have been highlighted and discussed. Both versions were seen as too vague to be of full practical use, and need a basis for acknowledging diverse normativity and types of application, and diverse rationality and types of human action.

These problems were addressed first by exposing their root, in the currently-active nature-freedom ground-motive and a tendency to reductionism. These are philosophical issues, and so a philosophy was appealed to which had discussed these at length, the cosmonomic philosophy of Dooyeweerd. Not only can it help us critique the current ground-motive but it also provides the basis for understanding diversity as experienced in the lifeworld. This led Dooyeweerd to delineate a number of aspects (spheres of meaning and law), and his suite is employed here to discuss normativity, rationality and social action. This offers a rich, multi-aspectual picture, which accords with the lifeworld of ISD and IS use. It is possible, in principle that an alternative suite of aspects could be employed, as long as it treats aspects not just as categories but as spheres of meaning and law.

It has been demonstrated that philosophy can have direct relevance, and can point the way to practical tools. The reason it has been able to do this is because the philosophy employed (that of Dooyeweerd [1984]) self-consciously adopts a pre-theoretical (everyday, lifeworld) attitude first, rather than presupposing that a theoretical attitude is the superior route to knowledge.

This article has not sought to work out the relevance of this for critical research in IS in detail. Rather, it has sought merely to raise some initial questions about critical approaches and to point out some directions in which research might proceed in order to seek answers. To draw these pointers together, the following suggestions may be made for various types of critical research in IS, most of them employing Dooyeweerd's suite of aspects as a checklist.

  • To research that is critical of social structures, Dooyeweerd's suite of aspects offers a tool for analysis that is intuitively grasped and serves to make explicit what has been overlooked. Winfield [2000] worked out a practical method for 'multi-aspectual knowledge elicitiation' {FOOTNOTE fn-make}.
  • To research that focuses on emancipation, the normativity inherent in Dooyeweerd's aspects offer a usefully diverse and well-founded definition of emancipation.
  • To power-oriented research, Dooyeweerd would offer aspects that are often overlooked when using power as lens, especially the ethical aspect.
  • To feminist research, Dooyeweerd's aspects offer confirmation of the desire to reinstate long-overlooked aspects, and helps feminism avoid the counter-productive strategies of mere reaction.

A benefit of using Dooyeweerd's aspects is that it focuses attention as much on the positive as on the negative, so that critical action research does not end up destroying the former.

We seem to have presented Dooyeweerd's suite of aspects as a universal solution. It should not be seen so. Dooyeweerd himself argued that while the aspects transcend us, our knowledge of them does not because it involves analytical thinking to differentiate them, and in this respect "theoretical thought has never finished its task" [Dooyeweerd, 1984,II:556]. Dooyeweerd was not only being critical but was being critical of criticality itself, recognising that even criticality is not absolute but itself invites critical analysis. This is what we have tried, in a small way, to undertake in this article.

Footnotes

FOOTNOTE fn-cr. The stance known as critical realism claims to integrate constructed knowledge with reality, but raises other problems; we do not discuss it here.

FOOTNOTE fn-hbwinch. Habermas [1984:53-67] does recognise the importance of this question and discusses Winch, but, arguably, the outcome of his discussion is flawed. In the end, claiming that Winch's argument was too weak, he reduced the number of types of rationality that he would discuss to two, namely instrumental and communicative rationality. He apparently resolved the problem of which type of rationality should prevail (communicative), but did so on the basis of the dogma that communicative rationality is superior. That he gathered reasons to support this dogma does not remove its status as dogma, because the reasons were largely provided by communicative rationality itself.

FOOTNOTE fn-hbvision. For example, one can detect in Habermas' writing about TCA [1984, 1987] a strong vision for, and a almost aesthetic delight in, its ability to account for many things in human and social life and to resolve problems left by previous thinkers.

FOOTNOTE fn-khi. Dooyeweerd's notion of ground-motives is a level deeper than Habermas' [1972] notion of knowledge-constitutive interests, which are more equivalent to Dooyeweerd's notion of aspectual world-views.

FOOTNOTE fn-int. Basden and Klein [2008] suggest the lifeworld might be composed to two types of shared background understanding, of which intuition of aspectual meaning is one. It is a shared understanding of what Dooyeweerd called the law side. In addition to this, we possess a shared background understanding of actual things and happenings, which are subject to the law side, which Basden and Klein call subject-side intuition. Law-side intuition is trans-cultural, but is sometimes obscured by subject-side intuition, which is culture-dependent.

FOOTNOTE fn-calf. Communicative action is more than simple lingual functioning. In saying, "But communicative action designates a type of interaction that is coordinated through speech acts and does not coincide with them" Habermas [1984:101] may have been referring implicitly to what Dooyeweerd called inter-aspect dependency.

FOOTNOTE fn-make. Winfield and Basden [2006] might be a more accessible source.

References

Adam, A. (1998). Artificial knowing: Gender and the thinking machine. London: Routledge.
Allen, D. K., Colligan, D., Finnie, A., Kern, T. (2000) Trust, power and interorganizational information systems: the case of the electronic trading community TransLease. ISJ, 10, 21-40.
Alvesson, M. & Willmott, H. (1992). On the idea of emancipation in management and organization studies. Academy of Management Review, 17(3), 432-64.
Basden, A. (2008). Philosophical Frameworks for Understanding Information Systems. Hershey, PA: IGI Global.
Basden, A. & Klein, H.K. (2008) New research directions in data and knowledge engineering: a philosophy of language approach. Accepted for Data & Knowledge Engineering.
Berg, M. (1998) The politics of technology: on bringing social theory into technological design. Science, Technology and Human values, 23(4):456-90.
Bergvall-Kåreborn, B. (2001). The role of the qualifying function concept in systems design. Systemic Practice and Action Research, 14(1), 79-93.
Checkland, P. (1981). Systems thinking, systems practice. New York: Wiley.
Clouser, R. (2005). The myth of religious neutrality: An essay on the hidden role of religious belief in theories (2nd ed.). Notre Dame, IN: University of Notre Dame Press.
Cordoba, R. & Midgley, G. (2003.) Addressing organisational and societal concerns: an application of Critical Systems Thinking to information systems planning in Colombia. pp.159-208 in Cano J.J. (ed.) Critical Reflections on Information Systems: A Systemic Approach, Hershey, PA: Idea Group.
Denning, P.J. & Dunham, R. (2006) Innovation as language action. Comm. ACM, 49:47-52.
Dooyeweerd, H. (1979). Roots of Western culture: Pagan, secular and Christian options (J. Kraay, Trans.). Toronto, Ontario, Canada: Wedge. (Original work published 1963).
Dooyeweerd, H. (1984). A new critique of theoretical thought (Vols. 1-4). Jordan Station, Ontario, Canada: Paideia Press. (Original work published 1953-1958)
Eriksson, D. M. (2006). Normative sources of systems thinking: An inquiry into religious ground-motives of systems thinking paradigms. In S. Strijbos, & A. Basden (Eds.), In search of an integrative vision for technology: Interdisciplinary studies in Information Systems (pp. 217-232). New York: Springer.
Fraser, N. (1989) Unruly Practices: Power, Discourse, and Gender in Contemporary Social Theory. Cambridge: Polity.
Goldkuhl G. (2006) Action and media in interorganizational interaction. Comm. ACM, 49:53-8.
Goldkuhl, G. & Lyytinen, K. (1982) A Language Action view of information systems. pp. 13-30 in M. Ginzberg, C. Ross (eds.), Proc. of the 3rd International Conference on Information Systems, Ann Arbor, Michigan, December 1982.
Habermas, J. (1972). Knowledge and human interests (J. J. Shapiro, Trans.). London: Heinemann.
Habermas, J. (1984). The theory of communicative action: Vol. 1. Reason and the rationalization of society (T. McCarthy, Trans.). Cambridge, England: Polity Press.
Habermas, J. (1987). The theory of communicative action: Vol. 2. The critique of functionalist reason (T. McCarthy, Trans.). Cambridge, England: Polity Press.
Habermas, J. (1991). A Reply. In Honneth, A. & Joas, H. (eds.) Communicative Action, pp.214-264.
Habermas, J. (2002). Religion and rationality: Essays on reason, God and modernity (E. Mendieta, Ed.). Cambridge, England: Polity Press.
Haraway, D. (1991) Simians, Cyborgs and Women: The Reinvention of Nature. London: Free Association Books.
Heng, M. S. H & de Moor, S. A. (2003) From Habermas's communicative theory to practice on the internet. Information Systems Journal, 13, 331­352.
Hirschheim, R., & Klein, H. K. (1994). Realizing emancipatory principles in information systems development: The case for ETHICS. MIS Quarterly, 18(1), 85-109.
Husserl, E. (1970). The crisis of European sciences and transcendental phenomenology (D. Carr, Trans.). Evanston, IL: Northwestern University Press.
Jackson, M. C. (1991). System methodology for the management sciences. New York: Plenum Press.
Lyytinen K J, Klein H J, (1985), "The critical theory of Jürgen Habermas as a basis for a theory of information systems", in Mumford E, Hirschheim R A, Fitzgerald G, Wood-Harper A T (eds.), Research methods in information systems, pp.219-231, North Holland.
Markus, K. L. & Bjørn-Andersen, N. (1987). Power over users: its exercise by systems professionals. Comm. ACM, 30, 498-504.
Ngwenyama, O. K. & Lee, A. S. (1997). Communication richness in electronic mail: critical social theory and the contextuality of meaning. MIS Quarterly 21(2):145-67.
Pacey, A. (1996). The culture of technology. Cambridge, MA: MIT Press.
Ray, L.J. (1993). Rethinking Critical Theory; Emancipation in the Age of Global Social Movements. London: Sage Publications.
Schutz, A., & Luckmann, T. (1973). Structures of the life-world, volume I. Evanston, IL: Northwestern University Press.
Strauss, D.F.M. (2006). Reintegrating Social Theory. Frankfurt-am-Main: Peter Lang.
Ulrich, W. (1994). Critical heuristics of social planning: A new approach to practical philosophy. Chichester, England: Wiley.
Van der Kooy, T. P. (1974). De gereformeerde wereld en de sociologie [The Calvinist world and sociology]. Anti-Revolutionaire Staatkunde, 2, 37-56.
Walsham, G. (2001). Making a world of difference: IT in a global context. Chichester, England: Wiley.
Weigand, H. (2006) Two decades of the language-action perspective. Comm. ACM, 49:45-6.
Wilson, F.A. (1997). The truth is out there: the search for emancipatory principles in information systems design. Information Technology and People, 10(3), 187-204.
Winch, P. (1958). The idea of a social science. London: Routledge and Kegan Paul.
Winfield, M. (2000). Multi-aspectual knowledge elicitation. Unpublished doctoral thesis, University of Salford, England.
Winfield, M.J. & Basden, A. (2006) Elicitation of highly interdisciplinary knowledge. pp. 63-78 in S. Strijbos, A. Basden (eds.) In Search of an Integrated Vision for Technology; Interdisciplinary Studies in Information Systems. Springer.