Epistemology and management science: Is accounting history still a legitimate subject of study?

Popper and pragmatism

Popper calls what distinguishes scientific logic from other types of discourse ‘falsification’. In empirical science, a theory should make postulations that can be asserted or refuted by experimentation and observation. We note that Popper does not take a positivist stance which considers that a theory can be considered to be true once it has been verified. In his epistemology, there are no statements that can be said to be unquestionably true. He, however, writes, ‘this does not mean that we cannot use concepts that are ‘true’ and ‘false’ or that their use presents any particular problems’ (Popper, 1934: 280). He does not renounce the idea of science as the pursuit of the truth, but simply considers that scientific knowledge develops according to an asymptotic dynamic, a ‘verisimilitude’. Scientific knowledge can progress but will never come to an end. Popper remains within the realist tradition that defines truth as the relationship between a statement and reality, but he does not consider that the two will ever correspond totally. Scientific logic is the continual search for falsification or refutation. Popper is a critical rationalist.

For a pragmatist like William James, however, a figurehead in this branch of philosophy, ‘truth serves practical interests’ (as quoted in Putnam, 1994: 405). In his conference papers, James declared that truth was measured according to ‘satisfaction’ (as quoted in Cometti, 2010: 169). In other words, what is true is what works and what pleases. For Popper ([1934] 1995), the Pragmatists define the term ‘truth’ in terms of the success of a theory, and therefore its utility (p. 281, underlined in his text), which is contrary to his conception of the truth. One interpretation of these opposing views on how to define truth could be that Popper assimilates it to facts rather than values, while the Pragmatists (James, Dewey and Peirce) refuse ‘to make any difference between questions concerning values and those concerning knowledge and truth’ (Cometti, 2010: 170). ⁵

To recall how we define what is useful, that is, what can be of service to someone or be profitable (thus synonymous with desirable), we can say that Popper separates utility from truth and is opposed to the pragmatist position that conjugates value and truth. Based on the conventional distinction between science and technology, we can consider the Pragmatists to side with technology rather than with science.

Truth versus value

These questions of definition are important and, as we will see, have practical implications for how we conduct management research. But in order to understand them better, we first need to briefly compare value and truth in the light of Comte-Sponville’s (1994: chap. 2) analysis. He identifies three possible relationships:

Science and technology

By adapting this analysis to epistemological thought, as derived from Popper, and by likening the concept of value to that of utility, one being contained in the other (utility does not have the same moral dimension as that of value, but this does not hinder the rational logic to be followed), we can consider that science and technology ⁸ have four possible types of relationship with truth and utility:

This is sophism.

The above relations are presented in the following matrix (Figure 1).

OPEN IN VIEWER

Figure 1.

Relationships of science and technology with truth and utility.

Dead-ends and perspectives

According to Popperian epistemology, should management be considered a science? If so, when and in what context does management research become legitimate? We will see that while much management research slips into dogmatism and sophism, action research offers some real perspectives ¹⁰ and the hermeneutic-historical approach some valuable insights.

The role of history

After Kant defined the limits between metaphysics and the knowledge of phenomena, Dilthey and Rickert asked in the nineteenth century whether it was possible to apply the same methods to moral sciences as to physical sciences (Aron, 1969). Hermeneutics, formerly reserved for the analysis of religious texts, served as a model to find a theory of interpretation, establishing a means for attaining the truth not through explanation, as in natural sciences, but through understanding. This theory studied a form of intuitive identification with other individuals which would make their behaviour intelligible to anthropologists notably via their history.

During the twentieth century, Weber, Gadamer and Ricoeur, in particular, defended hermeneutics as an alternative method to that of modern science for attaining the truth. This raises the question of whether historical hermeneutics is compatible with Popper’s understanding of science. Where Popper and hermeneutics converge is that they both criticize positivism and defend the idea that ‘science thrives on hypotheses and preconceived ideas that have to be constantly tested … there is no real method for finding these hypotheses and that crucial experimentation is mainly negative’ (Grondin, 2011: 409). But their differences are, however, apparent: in hermeneutics, knowledge of the truth is obtained through personal experiences which may be akin to aesthetic impressions. It does not consider the unsubstantiality of hypotheses as a way of differentiating between knowledge and science, which leads Popper (1956) to say,

I am therefore opposed to attempting to raise the method of understanding as a characteristic of the humanities, to the level of a criterion which would allow them to be distinguished from the natural sciences. (p. 286)

The author of The Logic of Scientific Discovery does not reject it as intellectual activity, but recognizes its potential:

It is true that we can, and rightly too, say a lot that has no bearing on science, but we must not present our statements as being scientific. (Popper and Lorenz, 1983: 82)

With regard to history, some consider it as a nomothetic science from which universal laws can be deduced. The ensuing predictions and prophesies would then reduce it to historicism or, as Comte-Sponville (1994) calls it, practical dogmatism. Another stance, in line with Popperian epistemology, is to consider it as an idiographic science regarding an individual fact. This would involve using verifiable real-life accounts and archival material to construct hypotheses on the causes of a situation that the historian is seeking to explain (Boyer, 1992). History does not operate like any other form of discourse, but is subject to criticism with truth as the yardstick. It thus offers a means of resisting fashion and sophism.

However, like Janus, it is two-faced. History involves recounting, narration and rhetoric and seeks to convince. If we may attempt an oxymoron, we can say that it is true fiction. We will have the opportunity to come back to this when we examine how some excesses of the Linguistic Turn and the French Theory could threaten accounting history with nihilism.

The question that needs to be asked when history is called upon for management and accounting is that of its utility. Here we cannot avoid mentioning a cause that has been debated so many times, with some opinions in line with that of the historian Paul Veyne (1979: 59) that ‘history is a cultural activity and free culture is an anthropological dimension’. Others, supporting the pragmatist view, pose ‘the question of the “social function” of history in the present world and ask how the knowledge we produce can help ‘“people to have better lives”’ (Noiriel, 1996: 197). ¹³ Both opinions converge when it is argued that

Knowledge is not affected by the different purposes, whether disinterested or practical, each of us attribute to it; these purposes do not constitute it but add to it. (Veyne, 1979: 60)

Here, we have the primacy of truth in the eyes of the historic sciences, as Roger Chartier (1998), a specialist in cultural history and a supporter of the ‘Annales school of thought’, reminds us:

We must argue forcibly that history is commanded by an intention and a principle of truth, that the past which is the object of its attentions is a reality that is external to any discourse, and that its knowledge can be controlled. (p. 16)

Many have argued that hermeneutics can be useful for management and accounting research (Llewelyn, 2003; Martin and Rainelli-Lemontagner, 2002). We cannot be sure whether history can be useful, but as Hobsbawm (1997) argues, we must strive to prove it so:

Unfortunately one thing historical experience has also taught historians is that nobody ever seems to learn from it. Still, we must go on trying. (p. 47)

It is thus that Popper’s (1956) message (quoting Kant, emphasis added) must be understood, because

To give in to all fancies of curiosity, and not allow our passion for research to be suppressed by anything other than the limits of our abilities, this demonstrates an intemperance of the spirit that befits erudition well. But it is wisdom which can be credited for selecting, out of the countless problems that arise, those whose solutions are important for humanity. (p. 60)

It is these varying stances that we will study in the next section, devoted to the virtues and dangers of accounting history. This will use the conceptual framework presented earlier which can be summarized by the following chart (Figure 2). At the bottom left, we have scholastics, which can hinder the innovations generated by experience-based research (bottom right), and at the top, history, which enables us to denounce sophisms and therefore prevent their dissemination as anything other than a fashion (top right).

OPEN IN VIEWER

Figure 2.

Relationships of obstruction and diffusion with sophism and innovation.

Utilitarian

We mentioned in the first part of this article that action research was an interesting concept for a pragmatic approach to management science. History can notably contribute to this by its method of conducting this research and can serve as a basis for introducing new concepts, models and theories (Zimnovitch, 2013). This type of research, which combines theory and practice in a positive dynamic, allows a theory to be expounded and may also contribute to its modification in view of the experience arising in applying the theory in specific contexts.

In the 1970s and 1980s, many academics noted a change in the technological environment of American industry with a risk of American companies losing market share to the Japanese. As early as 1984, Kaplan, Professor of Accounting at the Harvard Business School, criticized the management control methods used by US companies and the inadequacy of the traditional costing methods which, in his opinion, were responsible for the poor performance of US industry (Kaplan, 1984b). To justify his ideas and to find solutions, he investigated what was done with regard to costing in high-performing enterprises. He conducted various case studies with his colleague Robin Cooper (Kaplan, 1986).

They were joined by Thomas Johnson, Professor of Cost Management at Portland State University, and a doctor of economic history. To extract the ABC method from the specific contexts in which it was used and to generalize it, Kaplan and Johnson made a historical analysis of cost calculation methods, which was published under the title Relevance Lost: The Rise and Fall of Management Accounting (Johnson and Kaplan, 1987), while Kaplan also authored The Evolution of Management Accounting (Kaplan, 1984a). These publications were based on the history of management of the US firms of Alfred Chandler (1977) and showed that control systems would cease to be pertinent if financial criteria ever became hegemonic. This financial approach, which was still widely used at the time of the book’s publication (Johnson and Kaplan, 1987), no longer offered a management solution for controllers in an economic context that was increasingly unpredictable and where overheads were preponderant. The ABC method was the tool which would restore pertinence to these calculations.

If history can put traditional methods into perspective and help propose new ones, as was the case with the ABC method in the 1980s, it also favours

the placing of contemporary discoveries in their longitudinal perspective, namely the accumulated stock of knowledge on a subject over one or more centuries, thereby addressing the questions of how truly significant or new is a recent discovery. (Parker, 1997: 120)

In this respect, history warns against dogmatism and sophism, which is exactly what happened with the ABC method when it fell out of favour in the 1990s.

Many consultants would present ABC as the ‘ultimate solution’. It took advantage of the vulnerability of company managers, their fear of being overtaken by their competitors (Kieser, 2002) and their inability to control uncertainty, a well-known concern in management (Abrahamson, 1996; Clark and Greatbatch, 2002; Clark and Salaman, 1998; Crainer, 1997; Huczynski, 2006). The financial interests of the various exponents of the method (its inventors, Harvard, the CAM-I, the major consulting firms and software designers) were such that the ABC became the ‘one best way’, premised on the notions of ‘measure costs right; make the right decisions’ ¹⁶ and finding ‘the key to future costs’. The ABC method was presented as being ‘more accurate than traditional volume-based systems … they all appear to rely upon a two-stage allocation procedure to indirectly trace the costs of resources consumed in the production process to the end products’ (Cooper, 1989: 1). It was thus that the ABC method became fashionable. What once were academic arguments were now commercial arguments (Lukka and Granlund, 2002). According to Noreen (1987),

The chapters … setting down the authors’ recommendations contain speculations rather than goals to be met. In all honesty, they are like an advert for consulting services. And like all adverts, I suggest that these chapters be read with a good measure of skepticism. (p. 116)

Some historic papers, such as that by Jones and Dugdale (2002), recognized the ABC method and its offshoots for what they were: a passing trend. Likewise, works by French accounting historians such as Bouquin (1997), Lemarchand (1993a), Nikitin (1992) and Zimnovitch (1997) put costing methods back into their context, thus putting the virtues of the ABC method into perspective, saying that a ‘one-size-fits-all’ method is illusory. While the above example is taken from management accounting, there are many more possible illustrations of the usefulness of history to be found in financial accounting. One only has to refer to some of the special thematic issues that Accounting History published in 2000 (Accounting in Crises) and in 2005 (Accounting and Audit Failure within Corporate Collapse) to see the further insight history can offer. The following paragraphs provide further evidence of this.

Intellectual

History also contributes to our knowledge of accounting, by making it dependent on its social and political environment (i.e. the setting of historical events and actors in the local, time-specific context). It succeeds, by breaking away from complacent accounts of successful advances, by heightening our critical faculties, that is, making what we thought we knew appear unfamiliar. The history of depreciation in accounting studied by Yannick Lemarchand (1993a, 1993b, 1994a, 1997) is an example of this. The same applies to the structuring of the accounting profession and the debate on value.

For his PhD, Lemarchand (1993a) studied the development of the concept of depreciation and its practices in France over a long period of time, from the early seventeenth century to the start of the twentieth century. The history of deprecation thought and practice, as he presents it to us, is not that of a technique whose successive improvements are studied according to a Darwinian vision of constant and irresistible progress, selecting only the best practices. The author preferred to put the technique into its economic, socio-political, legal and cultural context, by considering such things as production techniques, forms of raising capital, sources of financing, the institutional environment and state intervention, none of which followed a purely linear process. While industry was confronted with the accumulation of capital as early as the eighteenth century, for two centuries there were nevertheless many conceptual uncertainties regarding the loss of asset value and how to account for it. This included the cost of raising capital declared as a fictive interest, accounting for wear or depreciation and estimating its impact on the cost of products, theoretical leases paid in obtaining the use of a fixed asset, forecast renewal and putting the necessary cash into a ‘reserve fund’ and so on. The highly diverse reasons for these various actions, such as self-financing, the policy for distributing dividends (or not), concealing losses during difficult times and, later, tax issues, also needed to be considered. During the nineteenth century, the technique for accounting for a loss of asset value became known as ‘depreciation’. The acceleration of technical progress, which brought with it equipment obsolescence in addition to wear, and state intervention in accounting practices favoured the spread of this technique. On the other hand, contrary to a linear vision of advances in accounting, depreciation was sometimes ignored by certain types of firm, such as most of the railway companies in France (Dakkam, 2014; Lemarchand, 1993a). In the absence of appropriate accounting regulations, it was court rulings that first started to codify evaluation and depreciation methods, which were quickly taken over by the tax administration once profits became taxable during WW1 in France. Owing to the many grey areas surrounding the concept of depreciation and the various possible translations of the term, its history shows us that an accounting technique may not be seen as an isolated set of rules and techniques that can be used to solve all problems, but on the contrary, as a response to a problem posed by a specific environment.

The history of the accounting profession as a form of socio-historical research (Poullaos, 1994) was begun by Willmott (1986). ¹⁷ He sought to go beyond the official history of these organizations which presents them as an advance (see, for example, Carnegie and Napier, 1996), both for all professionals and for the accounting profession, by putting them into their historical context and taking into account their identity, culture and system. A study of how accountants formed associations based on the aspirations, ideologies and interests of some individual players (Carnegie et al., 2003) considers the political aspects of these individuals engaged in conflicts and acting out of self-interest. While these authors used the rhetoric of the greater good, they continued to pursue corporate interests. Carnegie et al. (2003) reveal the alliances and struggles between professional bodies and how they functioned so as to allow their members to obtain ‘higher remuneration and prestige for their labour’ (Willmott, 1986: 559). In the same field, there were also biographies and prosopographies which highlighted the importance of certain individuals in this endeavour. Indeed, contemporary accounting cannot be understood without reference to some key personalities who have contributed to accounting development (Bocqueraz and Walton, 2006; Carnegie and Napier, 1996; Edwards, 1994; Previts et al., 1990).These works show the diversity of practices, types of professional accounting bodies and the economic and social role of accountants. They also serve to identify reasons why previous attempts at mergers failed so as to enhance our understanding of differences in structures and value systems and, thus, make it more probable that a future merger might succeed. On the other hand, the researcher might wish to demonstrate, by uncovering the reasons why separate bodies emerged in the first place, that merger attempts may prove otiose (Carnegie and Napier, 1996).

The recent debate on the question of the value of assets, ranging from the supporters of historic costs through to those who uphold fair value, also illustrates the intellectual usefulness of accounting history for tackling complex problems. We do not wish to present here the history of the different ways of presenting the depreciation of assets in a balance sheet, but to show why it is important to put them into their historical context. If we understand the different ways assets have been valued since the nineteenth century, it is easier to see their advantages and their limitations. When we speak of net asset value, it is for the needs of the creditors, while historic cost is used to monitor the industrial performance of managers and fair value to inform the money markets (Richard, 2005). While these developments do not lead to a concrete solution and we cannot even go so far as to say they are useful or precisely measure their influence, we can say that they stimulate personal reflection, contribute to forming an opinion and favour intellectual activity.

Besides what has been done already, much remains to be done, especially to ‘unlock the role of accounting in the social, and seek to discover its functioning in phenomena such as: the construction of social relationships, the deployment of social control, the solidification of social structures and the formation of social identity’ (Walker, 2015: 4).

The risk of dogmatism, sophism and scholasticism

Accounting history could have been tempted to adopt the naive economic determinism of such scholars as Littleton and Garner, to name the best-known accounting historians of the last century. In the last 40 or so years, however, many academics who did not agree with their ideas have adopted multi-historiographical methods, as evoked by Stephen Walker (2008) in observing

signs of mounting discomfort among accounting historians with dogmatic adherence to singular analytical frameworks. This was particularly the case among scholars associated with economic rationalism. (Walker, 2008: 299)

If the discipline has escaped the lure of dogmatism, can the same be said with regard to sophism? We have seen how history can play a beneficial role in opposing past events that are simply a passing trend. The legitimacy that the accounting profession can draw from its history could tempt some academics to embellish reality in order to achieve this legitimacy. Indeed, there are some business histories that are somewhat contrived, but we have no knowledge of such cases in accounting history written by academics.

The nihilism that surrounds certain excessive positions of the Linguistic Turn movement has presented a more serious risk. We admit that history should be considered as much a science as an art, and that its narrative dimension contains a certain desire to engage the reader, but certain positions taken by the Linguistic Turn are more radical: ‘the basic idea of modern theory of historiography is the denial that historical writing refers to an actual historical past’ (Iggers, 1997: 118). To take the example of French Theory which has been widely quoted by accounting academics (Chiapello and Baker, 2011), we can question the ideas that specific authors can really offer on the subject – such as Barthes (used by Davison, 2011), who ‘argues, there is no difference between truth and fiction’ (Iggers, 1997: 121), and particularly Foucault, ¹⁸ considering his great influence on what is now called new accounting history. ¹⁹

It would be irreverent to sum up in one paragraph how Foucault relates to the concept of truth in his work. May we simply say that his ideas have disconcerted more than one person, including his peers at the Collège de France:

If we considered, like my colleague and friend Michel Foucault, that we must once and for all dispel the myth that there is a difference between truth and falsehood, all philosophers would immediately lay down their arms. They would say that if we refuse to distinguish between truth and fiction, as philosophers they would no longer know exactly what they mean. (Aron, 1969: 145–146)

Some among the community of historians have even criticized Foucault for engaging too much in speculation to the detriment of the real facts:

Foucault excels in concrete knowledge as much as in theory; he even presents his abstract exaggerations convincingly enough to make us want to agree with him without taking the time to look at the ‘real facts’; what is plausible, if wrapped in literary connotations, can almost be more convincing than the truth – a bland scientific fact. (Léonard, 1980: 17)

In the arena of accounting history, we find in the same proportion certain historians who criticize an approximated view of facts: the neo-classicists will criticize the Foucauldians for supporting theories that are insufficiently well-informed (Tyson, 1993) and in which they will not let themselves become trapped. What is specific to this debate is the differing opinions as to the usefulness of management accounting. For the Foucauldians, and it is in this that we think the risk of critical history lies, labour costs only make any sense if related to a more general grammar, that is, an ‘episteme’ which makes it possible to consider an individual as a measurable object, rather than information calculated by a manager to increase the efficiency of his or her personnel or to help him or her make financial decisions (Ezzamel et al., 1990).

That we must treat with circumspection any claim in accounting that costs are true, this is after all the purpose of many accounting conventions; to say that it has no economic utility is for us a form of nihilism, but is nevertheless a subject of debate (Fleischman et al., 1995). Management history already has a hard time justifying its utility among accountants (Parker, 1999), but here it is battling on a second front on its own side, a balancing act between hyper- and anti-intellectualism.

The nihilistic risk of Foucauldian historiography applied to accounting is nevertheless contained. The power–knowledge pair is used in a highly watered down manner, verging on psittacism, compared to its philosophical and critical meaning which derives from Nietzschean genealogy (Macintosh, 2009) and, more generally, as Walker (2008) says,

The new accounting history is not as yet some wayward post-modern history libertine which has come to wreak improvident havoc. The different frameworks of interpretation used by new accounting historians … have not meant a complete repudiation of the conventions of traditional history. (p. 300)

If we can believe anthropologist David Graeber (2015), the reduction in credibility faced by all the social sciences in the United States started around 30 years ago: ‘We have been reduced to the equivalent of medieval scholastics, writing endless annotations of French Theory’ (p. 60). Over-specialization, introspection, a fall in the numbers of readers from other disciplines and subjects too far removed from contemporary issues: these are the risks of scholasticism. In 1999, regarding what was being said about company histories, Lee Parker (1999) wrote,

The editors specifically complained that company histories were mostly ‘paralysingly narrow and dull’ and that academics concerned with scholarly respectability were often ‘cautious, jargon-ridden and out of date’. If accounting and management history are to take their place ‘in the sun’ in the new millennium, they would do well to heed the lessons being offered by business history critics. (p. 27)

After the roaring 1990s (Fleischman and Radcliffe, 2005), in the noughties, it was argued that it was the turn of accounting history. Accounting historians have lost their interdisciplinary bite and have become ghettoized (Walker, 2008). This is a feeling that subsists in the 2010s (Carnegie, 2014a, 2014b; Gomes et al., 2011). ²⁰ The decline in accounting history is already perceptible in the United States (Fleischman and Radcliffe, 2005), ²¹ and it will not be long we fear before it will affect other countries (e.g. France?). The epistemologist Lakatos, a student of Popper, to whom we owe the analysis of scientific development through research programmes, makes no bones about what happens to those who degenerate: they disappear from the scene. ²²