Abstract
To economic historians, double entry bookkeeping served two primary purposes before 1800: control over debt and distant agents, factors, and partners. Few used it to identify overall profit or wealth, doing so only when accountable to others. They reached their conclusions from examination of primary sources within the wider context in which businesses operated. In contrast, the traditional antiquarian accounting history literature using the same sources tells a very different history, dominated by a focus on technique, present-minded teleological research methods, and an absence of context and critical reflection. The ‘what’ and ‘how’ of the bookkeeping is described, but ‘why’ is not asked, leaving motivations unknown, practices neither understood nor explained; and unsupportable assertions are accepted as axiomatic fact. Due to a lack of new enquiry, this literature is becoming increasingly corrupted by misunderstanding. This article considers why this came about and calls for change before it is too late.
Keywords
Introduction
In fact, at the beginning, [accounting historians] studied almost exclusively the technical development of accounting, that is, the invention and development of double entry, the formulation of accounting entries in the ledger in overlapping or opposing sections, the references to the contra entry, the establishment or not of balance sheets, the types of accounts used and other similar formal details. Although these are subjects of great interest, it is clear that with these perspectives the horizon of the [medieval] history of accounting was very limited and, what is worse, it was quickly exhausted. (Hernández-Esteve, 2001: 24)
Few historians would disagree that when investigating material in written form, the historical method involves asking questions of the primary source within what is known of its environmental context, typically using the ‘5 Ws’ of ‘what, who, when, where, and why’, plus ‘how’, in search of understanding and explanation of its form, purpose, and use (Arnold, 2021; Rosenwein, 2023: xvii–xviii). The only exceptions would be antiquarian historians who, by definition, focus primarily on the ‘what’ and ‘how’ and describe what they see (Bann, 1987; Sweet, 2004). Historians seeking understanding, critical historians, need to ask all the questions, especially ‘why’; otherwise, what they are examining cannot be understood; and no possibility of explaining, or of generalising what they discover is possible. Many critical historians who ask ‘why’ use the work of antiquarian historians, who ask ‘what’ and ‘how’, as the foundation for their own research (Sweet, 2001: 188). Yet, as will be discussed later, historians of double entry bookkeeping and financial accounting in periods from its origins to the beginning of the nineteenth century have not been critical. Almost without exception, they have done what is described in the opening quotation from Esteban Hernández Esteve – they have been antiquarian. They have not asked, ‘why’.
At any point in time, historians seek the answers to those questions that most convincingly embrace all that is known of the source under examination and its context. In doing so, most also recognise that what is known can change: that history is never entirely known, that new discoveries and new interpretations arise that lead to revision in understanding and explanation; and, to that end, ‘
Considering the fluidity of knowledge and discovery, historians expect future generations to make new discoveries, and those future generations of historians do what they can to make that happen. It is all part of what is viewed as the role of the historian. However, this is not the case for the historiography of double entry bookkeeping and financial accounting before the nineteenth century. In most cases, accounting historians have not sought to revisit that history or revisit the sources that were used. They are satisfied that their antiquarian predecessors captured everything and that their descriptions are all that is needed to ‘know’ that history. They do not believe there is anything new in yet unexamined sources, or that something new may be discovered if they revisit the sources of earlier accounting historians. Their perspective lacks the spirit of enquiry of historians who ask ‘why’.
An antiquarian perspective
Almost 70 years ago, though he never used the terms, Raymond de Roover implicitly classified the body of research on medieval
1
accounting as ‘antiquarian’ and ‘present-minded’, the habit of reading into the past our own modern ideas and intentions (Barzun and Graff, 1977: 43). In 1956 (p. 114), he wrote that it had a focus on form and procedure, that is, ‘what’ and ‘how’. And in 1958 (p. 42), he wrote that historians of accounting could not understand the practices of medieval bookkeepers because they thought differently then. In doing so, he explicitly acknowledged that accounting historians were being obstructed in their thinking by their own inability to free their minds of the present, and he included himself among them. Switching to another leader of this field in the mid-twentieth century,
2
in none of his publications on this topic did Basil Yamey shift from a focus on form and procedure. Neither did anyone else. ‘What’ and ‘how’ was what mattered. Technical matters dominated the investigation, and the discourse. The question ‘why’ was at most speculated upon, and several did, including Yamey (1959, 1977, 1997), but it was never investigated by, for example, considering the ‘
Beyond medieval times, despite strong recommendations to the contrary from De Roover (1956: 174) and with only a few exceptions, accounting historians have rarely examined account books before 1800. What is known by accounting historians of the early modern history of the technical aspects and uses of bookkeeping and financial accounting is mainly limited to very brief comments on the bookkeeping of, for example, a small number of mainly English businesses of the sixteenth to eighteenth century (Yamey, 1959, 1962, 3 1977); sixteenth-century Florentine international merchant bankers (Matringe, 2016); studies of large businesses, such as the Dutch East India Company (Funnell and Robertson, 2013), the English East India Company (Baladouni, 1986, 1990), the Hudson's Bay Company (Spraakman, 2015), the Pombaline international trading companies of eighteenth century Portugal (Rodrigues and Sangster, 2012, 2013), and French joint stock companies of the eighteenth century (Lemarchand, 1994); and studies mainly of English manuals on early modern bookkeeping (e.g. Edwards et al., 2009; Stoner, 2011), including those of Yamey (1949, 1962). There are a few that go into more detail, such as Vasconcelos et al.'s (2022) examination of the bookkeeping and financial reporting of two eighteenth-century Portuguese Atlantic trading companies, and Spraakman and Wilkie's (2000) study of management accounting at the Hudson's Bay Company between 1670 and 1820, but they are very much the exception. Consequently, the account-books-based accounting history of the period before 1800 is predominately of the period before 1500, and it is overwhelmingly antiquarian and technical. It tells us what the bookkeeping looked like, but not why double entry bookkeeping first came to exist, nor why it was used in particular ways.
Several accounting historians have provided explanations for why there was a shift from simpler methods to double entry in the nineteenth century (e.g. Edwards, 1985); but other than Lemarchand (1994), who considered the 18th century, none have investigated
How accounting history arrived here: A focus on matching twentieth-century practice
From the beginnings of research in this field, the primary focus was the quality of the bookkeeping compared to modern standards, along with its ability to identify movement in net assets in the capital account (Besta, 1916, III: 1–2). To this end, double entry bookkeeping is generally, but not universally defined
4
in the English language literature along the lines of: All transactions [must] be recorded twice, once on the debit and once on the credit side … This principle also involves the existence of an integrated system of accounts, both real and nominal, so that the books will balance in the end, record changes in the owner's equity and permit the determination of profit or loss. (De Roover, 1956: 114)
As De Roover (1956: 141–144) demonstrates, this definition of double entry bookkeeping is of an accounting system that may be used to provide a statement of profit and loss and a balance sheet in which changes in owner's equity/movement in capital can be found. However, those using it did not require evidence that those financial statements were prepared before recognising what they saw as double entry bookkeeping. That is, to them, bookkeeping and financial reporting were separate activities.
The antiquarian analysis conducted in the early twentieth century of medieval Italian use of double entry was led by Fabio Besta and his Italian contemporaries.
5
It gives anyone unaware of Besta's single-minded focus on assets and liabilities the impression that ‘we know everything’ of its use in that period. Besta could see double entries, that is, duality, from the thirteenth century but, similarly to the features required by De Roover's (1956) definition of double entry bookkeeping, Besta's definition (Martinelli, 1974: 217–223) required a bilateral layout in the individual accounts (Martinelli, 1974: 217–218). As a result, he could not find double entry bookkeeping earlier than in the account books of the Commune of Genoa in 1340. To him, medieval double entry unlike private companies, public administrations do not aim to make profits and their main concern is to achieve, often without much success, a balance of revenue and expenditure through tax revenues and state income, unless they have to resort to borrowing to meet extraordinary expenses. (De Roover, 1937: 172)
But Besta was not seeking profit; he was seeking movement in net assets – which is revealed by the movement in the capital account – so that the management or stewardship of the assets and liabilities could be monitored and controlled (Coronella and Maran, 2024).
Others embraced Besta's objective for using double entry bookkeeping, interpreting it wrongly as income-focused rather than asset-focused when seeking to identify movement in the capital account. But his requirement for bilateral bookkeeping, while still being applied by Raymond de Roover in 1937, lost favour in 1950 with the publication by Federigo Melis of I can state with certainty that double entry bookkeeping, believed and declared until yesterday to be of Genoese origin … [was not.] … the method is not conditioned by any form of accounts …. (Melis, 1950: 425, 429)
This freed scholars, including De Roover (1956: 115), to identify instances of double entry bookkeeping earlier than 1340. Alvaro Martinelli's (1974) 968-page summary of the literature published by Besta and his contemporaries on bookkeeping up to 1440, confirms that it contains many instances of double entries and double entry bookkeeping, and no examples of modern financial statements. However, offering explanations for what they found or, in this case, did not find, was not part of their antiquarian goal. Consequently, having answered the questions they sought to answer – ‘what’ and ‘how’ – activity ceased (cf. Hernández-Esteve, 2001: 24); and did so before they had begun to explore wider issues by asking the question, ‘why’.
Before this activity ceased, and as the restricting focus on banks and merchants in particular (Goldthwaite, 2015: 626), encouraged by Besta's purposing of double entry bookkeeping, and its misinterpretation by some as focusing on profit when his focus was on assets and liabilities, may have been expected to wane, Yamey emerged in 1940 followed, almost 50 years later, by Hoskin and Macve. The misinterpreted interpretation of Besta's objective for using double entry as being to identify movement of capital, that is, profit, was preserved. Double entry bookkeeping was ring-fenced as adoptable only by profit-seeking enterprises – merchants and banks.
Basil Yamey
Yamey's publications on double bookkeeping are of several types. All are antiquarian, and they dominate the discourse on this topic in the English language. In some, such as Yamey (2010), an early modern account book is described and illustrated by reference to comments from contemporary manuals of bookkeeping. In others, he discusses the features of different forms of bookkeeping, such as in 1997 when he describes early modern bookkeeping with no account books, single-entry bookkeeping, and double entry bookkeeping. In some, he discusses what manuals teach about bookkeeping, such as in 1967, when he reviewed early manuals from Italy, Germany, and the Low Countries 6 ; and in 1979, when he did so for England. But his most influential and best-known publications concern the claims of Werner Sombart (1924) that double entry bookkeeping was essential for the emergence of modern capitalism.
Readers of Yamey face several problems, beginning with the definitions he uses. Underpinning all his work, he implicitly recognised De Roover's (1956) definition of double entry bookkeeping. This is confirmed when he essentially presents it while discussing how Pacioli viewed double entry bookkeeping (1994b: 62): No one appears to have challenged the view that Pacioli was expounding the double entry system of bookkeeping; and no such challenge is mounted here. The system Pacioli describes satisfies the two central criteria for DEB. First, for each transaction there are two equal but opposite entries in the ledger. Second, besides real and personal accounts, there are also a variety of revenue and expense accounts (which are brought together in a profit and loss account) and a capital account.
However, in his Sombart-inspired investigations of double entry bookkeeping before 1800, which began in 1949, the definition Yamey uses for double entry bookkeeping changes to one that conflates double entry bookkeeping and financial accounting: A particular and influential view of what counts as accounting and what counts as evidence is thus established through Yamey's arguments. … accounting is made equivalent to double entry bookkeeping. (Miller and Napier, 1993: 637)
This is evident in the features he discusses when evaluating the accounts of the English East India Company in 1962, and when he examines double entry ledgers, which he did in several of his publications (e.g. 1959, 1977, 1997). By focusing on what he discusses and highlights, virtually uniquely among accounting historians,
7
Yamey implicitly required the existence of appropriate, twentieth-century
Yamey justifies his use of this different Sombart-inspired, accounting-focused definition for double entry bookkeeping when declaring that both definitions he uses are faithful to the same double entry framework, that is, they are both examples of the same thing, double entry bookkeeping: [Pacioli's 1494]
However, as noted above, in applying his Sombartian definition he was
Unfortunately, there is no recognition in the literature that this is what he was doing. It is left to the reader to deduce which definition he is using. Furthermore, he implicitly uses both definitions when looking for modern features in ledgers that he recognises to be in double entry. This highlights a second problem for readers of Yamey: the terminology he uses, which can leave anyone unsure whether he is referring to double entry bookkeeping, double entries, or something else. In only one article – (1994a) – does he continuously and almost uniquely use the term, ‘double entry bookkeeping’ – 32 times in six pages. Perhaps intentionally, in doing so, he provides guidance to those seeking to understand the synonyms he uses elsewhere. Otherwise, he uses the term occasionally in most but not all articles, preferring to use any of several different terms including:
Any of these could refer to either his Sombartian definition or the De Roover definition. He also used the term ‘modern bookkeeping’ to label the former, which he took from Max Weber (Yamey, 1949: 99), who had similar views to Sombart about double entry bookkeeping. However, he never otherwise used that term and only once used each of the synonyms ‘modern method’ (1947: 271) and ‘modern accounting’ (1994b: 56). Consequently, anyone reading Yamey's work needs to be aware which definition was relevant and that these terms all referred to the same thing – double entry bookkeeping – something that requires an informed and very critical and reflective approach to reading what he writes.
It could be argued that any accounting historians using Yamey's material on this topic as their source would read it carefully enough not to be misled. But, if so, why has no one pointed out how Yamey can be misread? In the style it is written, exacerbated by his use of many synonyms for double entry bookkeeping – many of which may be interpreted differently by those unaware of the meaning he attached to them – it is easier to uncritically interpret his criticism, which is continuous, as being criticism of the bookkeeping.
For those unaware of what Yamey was doing, which a lack of recognition of these features in the literature suggests is the majority, this double definition, or homonym, results in what appear to be contradictory conclusions. He recognised that non-Sombartian double entry bookkeeping was used ‘
Perhaps that conclusion was influenced by his disbelief that double entry bookkeeping without value adjustments had any merit. He could not explain why double entry bookkeeping was used before modern times (1940: 342, 1949: 113); speculated inconclusively about the features of the non-modern bookkeeping practices he saw in the early modern double entry ledgers he examined (1959, 1997); and, unable to accept an alternative perspective (Lane, 1977; Yamey, 1975), he was convinced that double entry bookkeeping served a meaningless role until appropriate value adjustments were made in modern times because other simpler methods of bookkeeping could satisfy the same demands (1949: 105, 1962: 854–856, 2005: 84–85).
In contrast, he did virtually the opposite in his conclusions concerning his Sombartian form of double entry bookkeeping, generalising from his analysis of the same very small set of early modern ledgers, plus less than 4 per cent of bookkeeping manuals published before 1800
8
: ‘
Not only does Yamey's generalisation of his findings in these ways appear less than impartial, use of them by other scholars appears to be selective. Despite his 1949 article having been cited more than 400 times, his recognition that double entry bookkeeping was used ‘
Overall, this quite extensive body of literature is remarkably confusing and ambiguous, but there is one clear conclusion that can be drawn from it, apart from what Yamey wrote about the claims of Sombart. De Roover (1956: 174) concluded that accounting did not develop further until the nineteenth century, which is what Yamey also found. Neither Yamey nor De Roover ever wrote that double entry bookkeeping was rarely used before 1800.
Hoskin and Macve
In possibly the most cited and influential publication in the accounting history literature of the past 50 years because it introduced use of Foucauldian theory to accounting history, Hoskin and Macve (1986) adopted a much simpler approach, presenting far less technical detail than Yamey, and vastly less than is to be found in the earlier literature. In a few lines near the beginning of their article and using no sources to support their assertions, they summarised the literature on medieval and early modern accounting history. In complete denial of, and contrary to the findings of De Roover and Yamey, they created a five-century-long period of drought of little interest to accounting historians, stating that First it is now clear that full-scale systematic accounting
9
(… especially in the new double entry systems) is developed in the thirteenth and fourteenth centuries … . Second it is clear that it remains sporadically used until the nineteenth century….
In doing so, they gave no one who would read their article any motivation to discover what double entry bookkeeping was used for before 1800 – if it was not being used, why investigate its use? If they were believed, none would. Neither those convinced, like Besta, that the only reason to use double entry accounting rather than simpler forms of bookkeeping is to identify movement in assets and liabilities; or, following Yamey, to identify movement in capital.
Overall
The inflexible present-mindedness combined with an antiquarian focus that pervades much of this literature until the 1980s is the reason very little work has been done by accounting historians into explaining why double entry bookkeeping was used before 1800. This was especially the case after Hoskin and Macve's unfounded and misleading assertion made in 1986, which surely discouraged even critical accounting historians from revisiting this field of enquiry. Accounting historians are satisfied that the work of Italians in the first half of the twentieth century was sufficient to inform them what they sought to know about use of double entry bookkeeping by medieval Italians, and this is still very much the case today (Costa, 2017: 40). What Yamey wrote, despite being almost entirely about bookkeeping in England, fills much of the gap from that medieval literature to the literature that focuses on the nineteenth century.
This body of research from De Roover to Yamey and Hoskin and Macve never considered what motivated use of double entry bookkeeping when financial statements were not produced. To those scholars, unable to overcome their present-day mindset (c.f. De Roover, 1958), doing so was, in any case, pointless because they mistakenly believed they knew from Besta that the purpose of double entry bookkeeping was to identify movement in ‘capital’, when Besta's focus was actually on the assets and liabilities, not the profit.
Inexplicably, despite the overwhelming lack of pre-modern evidence to support this misinterpretation of Besta's focus, it has never been considered by any accounting historian that it may have been wrong, that the capital-focused motivation identified in the twentieth century for using double entry bookkeeping mainly by those writing in English might not be relevant to earlier times. Besta's focus on net assets was relevant, but the focus on profit that others interpreted it as, was not.
This literature is also replete with other myths, misunderstandings, and misconceptions, including the implicit assumption promoted by Hoskin and Macve (1986: 108) that Pacioli's (1494) description of the Venetian approach based on a ledger paired with its journal reflected how double entry bookkeeping was done everywhere in medieval times. Even a cursory glance at artefacts originating elsewhere, or simply at the catalogue of account books in a Tuscan archive, reveals instantly that it was not. 10 More to the point, the uniqueness of the Venetian paired journal and ledger is evident to anyone who looks beyond the descriptions of regional differences in the accounting history literature by reflecting on what that literature reveals. 11 Despite Pacioli (1494) mentioning it, this had not been noted as a distinct feature in the literature because the focus of this branch of accounting history to this point has been relentlessly antiquarian.
While the antiquarian nature of this literature is obvious for all to see, it is only through adopting a hermeneutic approach (Reiter, 2006), reading it critically and extensively, and repeatedly that the reality of what it contains or, perhaps more accurately, what it
How others view this
In contrast to many accounting historians, historians of other disciplines, from their use of books of account and business records in their own research, do understand how double entry bookkeeping served the needs of business and other forms of entity in earlier times, and they recognise its use. They dismiss the medieval accounting history literature with contempt (Goldthwaite, 2018; Lane, 1944, 1977; Pilla, 1974). Compounding that criticism, some accounting historians have also voiced concern over its Whiggish nature (Funnell, 1996: 54; Vasconcelos et al., 2022), in how it views history as inevitable progress and improvement, seeing the present and looking for signs of progress towards it in the past. It is a situation that 40 years ago, some senior accounting scholars believed applied to accounting scholarship as a whole, one of whom also drew attention to the lack of interest in challenging current beliefs: Much of the corpus of traditional accounting doctrine and practice rests on myths, propositions that are treated as not corrigible by recourse to observation and experiment. There are myths or quasi-myths in abundance, in respect both of substance and method. Adhocery, polarities and indeterminacies persist for lack of definition and critical analysis as well as for lack of empirical investigation. To escape from dogmatism takes time.
The rest of accounting research has moved on. With very few exceptions, research on double entry bookkeeping and financial accounting before 1800 has not. It continues to be antiquarian, still clinging to research methods and ill-informed beliefs first established over 100 years ago, reflecting a lack of awareness of what merchants and others were most interested in doing with the records kept in their account books; of what motivated them to use double entry bookkeeping; and of what did, and did not motivate them to prepare financial statements. In addition to its antiquarian nature and the problems of generally believed myths, misunderstandings, and misconceptions afflicting anyone working with it, other factors add another layer of confusion to this literature.
Further problems in this accounting history literature
Yamey's implicit use of two definitions for double entry bookkeeping highlights one specific problem: terminology. Another concerns the variations that exist in the definitions adopted, such as Fabio Besta's demand for bilateral ledger accounts, which resulted in its earliest use originally being recognised in governmental bookkeeping in 1340. Elsewhere, Geoffrey Lee's (1977) outstanding scholarship, which established without any doubt the use of double entry bookkeeping in 1299, defines double entry bookkeeping (p. 85) as requiring:
The firm as a distinct and separate entity to which all the entries relate; Duality in all entries; Use of a single money of account; Capital as the balance between assets and liabilities; Profit or loss as the net increment to capital; The existence of periodicity that provides a time-period over which profit is measured.
To these, he later added (p. 91): ‘
Lee's first and sixth criteria assumes that all medieval Italian double entry was done as it was done in Tuscany which, as previously mentioned, it was not (Antinori, 2004; De Roover, 1937; Lane, 1977; Pilla, 1974). Furthermore, if this definition were applied to all the early modern ledgers examined by Basil Yamey, none would be classified as double entry bookkeeping, nor would any fifteenth-century Venetian double entry ledgers, or even Pacioli's description of the method.
The differing views in the literature of what double entry bookkeeping is, each in its own way impacts conclusions and understanding (Lane, 1977: 185–187; Winjum, 1971: 334–335). The meaning of the term used needs to be spelt out, as Lee did, if what is written is to be understood and correctly interpreted. Unfortunately, that is rarely done, and statements such as ‘they used double entry bookkeeping’ or, ‘double entry was not used’ are treated as unambiguous when, lacking any definition, this is never the case. Furthermore, Besta's requirement for bilateral accounts before recognising double entry bookkeeping was embraced universally before 1950, when it was dropped by Federigo Melis. Thus, anyone reading the literature published before 1950 needs to appreciate that when bookkeeping is labelled in that literature as ‘single-entry bookkeeping’, it may have been because it was in paragraph form, which after 1950 would have resulted in the same example being labelled as ‘double entry bookkeeping’.
The uncritical self-informing nature of the literature
These confusing elements in this literature are almost completely unrecognised, perhaps because antiquarian accounting historians are not, by definition, critical researchers. However, other accounting historians who would profess to be critical, such as Hoskin and Macve, have also failed to recognise the problems. Being critical accounting historians does not appear to extend to considering how sources and the statements or phrases within them, including ‘double entry bookkeeping’ and Yamey's many synonyms, are used.
This is evident to anyone drilling down through the sources used in the accounting history literature of double entry bookkeeping before 1500 and of later periods that use the medieval accounting history literature, passing through the sources and the sources of the sources. Anyone doing so will recognise that later scholars trust their predecessors. They do not heed the aforementioned advice of Barbara Merino (1998: 607) to embrace ‘
The result is an increasing undetected level of confusion, and an accounting history literature on double entry bookkeeping and financial accounting before 1800 that encourages, in addition, a view of uniformity and progression. But as anyone will confirm who, for example, examines the 3,000 medieval and 2,000 sixteenth-century Tuscan account books held in Italian archives (Bettarini, 2021; Goldthwaite and Spallanzini, 2018) and the hundreds of others that have survived from elsewhere, including those examined by Basil Yamey: there was no uniformity in practice beyond a superficial level, even in individual firms; no Darwinian progression of the method as refinements were added.
The corrupting impact of this uncritical literature
Two recent examples illustrate how, alongside these problems, a new fantastic historiography of medieval double entry bookkeeping and financial accounting is developing that is being validated by reviewers and editors lacking the specialist expertise to separate fact from sometimes extreme, fiction.
Example 1
The text relating to a table recently published by an accounting historian, and the table itself (Matringe, 2022: 764), typifies how far this situation has corrupted the historiography of medieval Italian accounting. The table charts the evolution of the accounting concept of offset from the thirteenth to the fifteenth century, which the article claims (p. 766) did not exist in early paragraph format bookkeeping, the earliest example of which dates from 1211 (Lee, 1973). The table contains four equally spaced vertical columns representing the passage of time over that period, and arguably, more than 20 misstatements. Assuming that each column represents a span of 75 years, the qualities in the bookkeeping that existed in well-known pages from that ledger from 1211 are indicated as first occurring as much as 200 years later, despite the ledger being included as an exemplar of the items listed in the first column, representing the earliest period. These qualities include existence in that ledger of a section for debtors and another for creditors, which the table locates in the second of the four columns; and cross-referencing to the contra entry, which is shown in the third column. Double entry bookkeeping is shown in the final column representing the fifteenth century while economic historians might recognise existence of double entry bookkeeping in that ledger from 1211; and accounting historians who were not adherents to Yamey's implicit need to see ‘appropriate’
It is also claimed in the final column of the table that examples exist from the fifteenth century showing use of the balance sheet equation (equity = assets – liabilities), supported by asserting that this can be seen in the account books of the Tuscan international merchant Francesco Datini in the late-fifteenth century. Datini died long before that, in 1410, and none of his more than 600 surviving account books and financial reports contain any known representation of that equation. Furthermore, there is no known example of a balance sheet that formally presents the equation until long after the end of the fifteenth century – not even the English East India Company did so until the 1780s (Baladouni, 1986, 1990), despite by then having used double entry bookkeeping for over a century. Yet, using only sources from the accounting history literature, this table has been compiled and supported sufficiently by appropriate sources to be accepted as valid by reviewers and the editor.
Example 2
Another recent publication (Yuryevich, 2024) has introduced new definitions for single-entry bookkeeping and double entry bookkeeping that restrict the label of ‘single-entry bookkeeping’ to refer to a bookkeeping system where some or all of the following classes of items are not included: personal accounts, a cash account, accounts for goods, and nominal accounts.
The author also broadens Fabio Besta's focus on identifying movement in assets and liabilities as the reason double entry bookkeeping is adopted, which is misinterpreted in the literature as identifying movement in net assets/capital, to equally apply it to his version of ‘single-entry bookkeeping’. In the eyes of the author, they are not different bookkeeping systems, Inventory of all assets and all liabilities at the end of the reporting period (as of the balance sheet date). Goods are recalculated according to actual availability, obligations-according to the data of primary settlement documents and reconciliation of the status of settlements with counterparties. On this basis, the capital at the end of the reporting period and the profit for the reporting period (year) are calculated. … profit [is defined] as the difference between the values of capital at the end and beginning of the period (adjusted for the amounts invested by the owners in the business and withdrawn from the business in the form of dividends). (Yuryevich, 2024: 5)
The final (fifth) level of this ‘single-entry bookkeeping’ embraces the inventory of assets and liabilities plus maintenance of a cash book, and personal and commodity accounts, but no nominal accounts. ‘Double entry bookkeeping’ is described as that final level of ‘single-entry bookkeeping’ plus nominal accounts (Yuryevich, 2024: 6). The manner of this conversion to ‘double entry bookkeeping’ that results solely from the addition of nominal accounts makes it implicit that all entries in the cash book, personal accounts, and commodity accounts are, or could potentially be double entries, otherwise the addition of nominal accounts would not convert the system to double entry. That is, ‘single-entry bookkeeping’ as defined in this article does not comprise of single entries, except perhaps at the first ‘inventory’ stage. Its sixth and final stage is, however, consistent with De Roover's (1956) definition of double entry bookkeeping. But labelling all prior steps as ‘single-entry bookkeeping’ is misleading, not least because this is a term widely used for simple notation as, for example, is done when a small shopkeeper records the amount owed by a customer on a scrap of paper.
Whilst it is an intuitively appealing proposition to conveniently explain the thirteenth-century emergence of double entry bookkeeping in this way, it reflects a reality much of which historians of medieval Italy would not recognise; and it cannot be supported objectively by the evidence. The author's identification of reporting periods of one year ignores the fact that Venetians did not close their ledgers until they were full. There is also no evidence that they had any interest in calculating their overall profit by preparing an inventory of the type described by the author; and the form of trade they conducted, with its reliance on ventures, which included consignments of goods sent to agents, that could last months or even years before completion, made doing so meaningfully impossible (Lane, 1977: 188–189; Yamey, 1962: 838–839).
With respect to medieval Tuscan bookkeeping, some early 14th century partnerships prepared summary reports of assets and liabilities, such as the Alberti, and the Bardi and Peruzzi international merchant banking firms; and they distributed the difference between the totals to the partners as if it were profit (Sapori, 1934: 14; De Roover, 1956: 123–130). But this was several years after the first confirmed instances of double entry bookkeeping in 1296 (De Roover, 1963b) and 1299 (Lee, 1977), not before.
The earliest known to have drawn up such reports annually was Francesco Datini in the late fourteenth century, almost a century after the first confirmed instance of double entry bookkeeping in 1299 (Lee, 1977). However, Datini's reports were similar to a single-column trial balance, listing separately all the debit and credit balances, one following the other, neither of which lists was organised into any particular order. Unlike the inventories described by Professor Yuryevich, they were not used to determine profit; and they never became nor represented one of his six stages in progression towards double entry bookkeeping. They were not compiled from evidence. They were compiled directly from the ledger accounts. Furthermore, in the several thousand account books kept in Florentine archives into the seventeenth century ‘nowhere … is to be found a single inventory of one's total net worth of the kind Pacioli lays down as the very first step in the process he sets out to describe’ (Yuryevich, 2024: 28), which is implicitly what Professor Yuryevich appears to be describing.
Concerning calculation of their net worth, the largest Italian firms of the fourteenth century – the Peruzzi and the, Bardi – the smaller Alberti firm, and the fifteenth-century Borromei bank (Corsani, 1922: 57–58), Datini's original firm in Avignon, and the Medici bank all kept secret ledgers containing ‘a highly synthetic presentation of partners’ equity and division of profits over a number of years’ (Yuryevich, 2024: 17). But none of them contained any information on investments, such as land, not recorded in the ledger (Yuryevich, 2024: 28). While they could provide information to the authorities on properties held because they kept separate records for them, these were not embraced in the bookkeeping system from which they identified profits to be distributed.
Further fantastic propositions in the article relate to the author's assertion that ‘single-entry bookkeeping’ and ‘double entry bookkeeping’ competed for superiority from the thirteenth century to the twentieth century (Yuryevich, 2024: 16). There is no evidence that single-entry bookkeeping, however defined, and double entry bookkeeping as defined by the author ever ‘competed’ with each other. Accounts were kept using the method perceived to be most suitable for the needs they served, and that was not necessarily either the ‘single-entry bookkeeping’ described by the author, or double entry bookkeeping. Most businesses until the nineteenth centuries that did not use double entry bookkeeping used simple bookkeeping methods, normally labelled ‘single entry’, that do not embrace or include double entries, unlike the version of ‘single-entry bookkeeping’ described and defined in the article, which implicitly does.
Also concerning the definitions used, it is asserted on page 13 that ‘almost all small enterprises’ – declared to include
These two recent articles both assume that bookkeeping developed in stages until the records kept were consistent with one of De Roover's definitions of double entry bookkeeping, either that of 1937 (p. 270) or that of 1956 (p. 114), both of which required bilateral accounts. It may have done, but only in individual firms, and at very different times. However, with respect to offset, examination of the entries in the bank ledger from 1211 that the first article cites reveals that offset was already a fully developed process. It is difficult to perceive how, by its nature, involving the transfer of an amount from one account in a ledger to another, offset could have developed in stages. Neither of the sources cited by Matringe (2022) – De Roover (1937, 1956) – suggest that it did.
Both these examples illustrate how the accounting history literature is being corrupted by uncritical use of secondary sources that serves to compound the errors, misunderstandings, and ambiguities within it concerning the past. From another perspective, fantastic stories such as these being weaved and published in highly respected outlets highlight the lack of critical thinking and reflection in the pre-modern accounting history literature of double entry bookkeeping and financial accounting; and, in some ways even more concerning, among some who are writing and reviewing articles before publication that include those topics.
If Anthony Hopwood was correct when, emulating Lane (1944: 153), he wrote in 1983 (p. 289) that accounting history lacks consideration of organisational context, and in 1987 (pp. 207–208) that it lacks critical thinking and reflection, what would he write now? Hopwood also dismissively observed (p. 207) that an antiquarian approach dominated the accounting history literature, and that it was time for this to change. The response was already underway, and it took accounting history in a new direction.
A possible solution, or a false start? – Socio-historical accounting history
The response to Hopwood's comments was an acceleration of the emergence of ‘contextualising accounting’ (Napier, 1989: 244) which had first begun to appear in the mid-1980s in the form commonly known as ‘new accounting history’, or ‘socio-historical accounting history’. This shift in focus immediately impacted where accounting history was published including, not surprisingly, the generalist journal Hopwood edited,
The emphasis in socio-historical accounting history is on theory generating answers, not the account records; with the goal of developing the theory prioritised. Compared to the predominately antiquarian history of accounting up to that point, it directly addressed the question ‘why’; and, as analysed by Napier (2006), did so in multiple directions involving primary sources previously not considered within the remit of accounting history, broadening the sense of accounting to embrace a wide and diverse field of interest. Contrasting the two approaches, traditional antiquarian accounting history is criticised for its focus on description, not understanding or explanation and, as noted by Carnegie and Napier (1996: 15), the emphasis in that literature on the economic decision-usefulness of accounting. Similarly, socio-historical accounting history may be criticised for its theory-driven, narrowly focused and relatively context-and-content-limited nature. Perhaps the most comprehensive comparison between these two branches of scholarship is Funnell (1996), in which both the criticisms against them and their positive aspects are discussed.
It is not the purpose or role of this article to take sides. Each of them can be legitimately criticised by its detractors, but some of the criticisms are directly relevant to the focus of this article. In particular, answering the question, ‘why’. For example, an accepted truth of historical research ‘is that all evidence is subject to multiple interpretations’ (Walker, 2008: 299) which leads to criticism of the socio-historical approach, pointing to risks inherent in its emphasis on relations of power, privilege and domination (Ciambotti, 2009: 138; Hernández-Esteve, 2001) when many other factors and perspectives may provide other, perhaps more convincing answers to the question ‘why’ (Ferri et al., 2018: 192). As with all scientific research, 14 having identified what is to be investigated, the socio-historical method begins with the selection of a theory based on a preconceived perception of which one would be the most appropriate to apply to what is being examined, which is itself a subjective choice, and one on which there may be considerable disagreement. Variables to consider are then selected. If supported by the findings, the theory is shown to be applicable, but in a contextual vacuum, that is, in a ‘perfect world’ where nothing else is relevant beyond what has been recognised. Evidence that does not fall within the focus of the theory can be ignored while an ‘objectively’ identified conclusion is reached. Whether the conclusions would still hold if all the knowable environmental context were considered cannot be determined because other contextual variables are not sought, recognised, or known. Furthermore, other explanations for what is being analysed can be ignored or remain unidentified without interfering with findings.
Funnell (1996) sets out the criticisms of its proponents against traditional accounting history who may, for example, point to selectivity bias whereby what can be examined is dependent on what has survived, which is impacted by why some things have survived and others have not. They may also point to the traditional accounting historian's subjectively informed selection of what is examined – for example, one account in a ledger, rather than another, or one account book rather than another – and to the antiquarian, context-lacking, predominately economic decision-making assumption, or focus, underlying much of that historiography.
From another perspective, there is no motivation for the socio-historical researcher to fully understand or examine in detail the artefact, evidence, or event being investigated. Consequently, some fear that the depth of analysis of the accounting itself might be too superficial, that is, our knowledge and understanding of the accounting might not be adequately enhanced by this approach: A danger of socio-historical accounting research is that practitioners do not probe deeply enough into the actual accounting that they claim is affecting organisations and society. (Napier, 2008: 40)
Some, such as economic historian of Renaissance Florence, Richard Goldthwaite, believe this danger has materialised (2018: 249–250, bold added). Over the last generation or so historians of accounting have been much agitated over, on the one hand, what influences the ambience from which double entry emerged had on its formation, and, on the other – taking up the older concern of Max Weber and Werner Sombart – how to measure the impact it had on the rise of capitalism. The discussion of influences going back into the ambience proceeds more by way of speculation than research and tends toward argumentation by analogy with rhetoric, algebra, geometry, socialization, networks, ‘panoptic surveillance’, or whatever other fashionable analytical tool shows up. Going forward to impact, the point of departure is [
Intuitively, the medieval period in northern Italy, in which the transformation from feudal to municipal society occurred – a transformation with merchants and moneychanger-bankers at its head, the emergence of guilds, the emergence and growth of legal systems, both commercial and Canon law (Fortunati, 1996; Greci, 1988) – a period during which the societal upheaval alone ought to have attracted proponents of a socio-historical methodology with their emphasis on the social rather than the economic (Funnell, 1996: 39) or, as some believe it transpired, the social, political and institutional rather than the economic (Bisman, 2012: 8), along with its focus on relations of power, privilege and domination (Hernández-Esteve, 2001: 31) – but it has not. Their focus has been mainly on the nineteenth-, twentieth-, and twenty-first centuries, and generally not directed towards the account books that are the core of the discipline and practice of accounting. This is not just the case with socio-historical accounting history. Accounting history in general since the mid-1990s has focused on modern times (Fowler and Keeper, 2016: 410; Walker, 2008: 314).
Consequently, the emergence of socio-historical accounting history has made virtually no impact on knowledge and understanding of the history of bookkeeping and financial accounting in earlier periods, and no impact at all on the period between 1150 and 1500 when the foundation emerged upon which modern accounting stands. The historiography of medieval and early modern double entry bookkeeping and financial accounting, constructed through the uncritical and unenquiring antiquarian scholarship of earlier generations of accounting historians, continues to act as the foundation for any use made of that historiography. And, as illustrated earlier, this literature is becoming increasingly corrupted as it is misused and little understood by those who uncritically embrace it, including, as mentioned in the above quotation, adherents to a socio-historical methodology. Another solution is needed if the increasingly muddled and confusing historiography of pre-modern double entry bookkeeping and financial accounting is to be addressed; and it needs to deal also with another issue that has arisen following the rise of socio-historical accounting history.
Other considerations
Before reflecting on how this situation may be tackled, any other approach adopted needs to address the current noticeable lack of respect from several senior accounting historians for any form of historical enquiry by accounting historians that is not socio-historical. Failing to do so will signal a catastrophic endgame for accounting history feared by, for example, Funnell (1996: 59) and Hernández-Esteve (2001: 32). That there is a problem is indisputable. From the beginning of socio-historical accounting history, journal editors have perceived its use of theories as indicative of research validity and quality, which was obviously welcome as it improved the status of accounting history as a field of enquiry. However, some socio-historical journal editors publicly express the view that knowledge of the past, in itself, has no relevance unless it can be used to develop understanding of theories from the social sciences.
15
Others implicitly reject traditional history in any form as a valid approach if done in isolation of theory, believing that only through application of existing theory can a contribution be made: Papers published in this journal are expected to make strong theoretical and empirical contributions to our understanding of accounting as a social, institutional or behavioural practice. (Senior Editor of top-ranked journal, 2023, personal communication)
To the keen observer, this statement raises another even wider issue: why accounting history research of any form is not considered worthwhile if it focuses upon accounting as an economic practice, which is where the emphasis lay before the advent of socio-historical accounting history. No one would surely argue that accounting exists, if not entirely, at least in major part for economic reasons. Yet, it appears it is no longer considered valid to investigate this major theme in its history which, as discussed in this article, is something it has barely begun to do. The contest between the social and the economic, summarised by Funnell in 1996, has apparently run its course to the destruction of the latter, at least in the eyes of this senior editor: New accounting history raises to prominence interpretive frameworks which owe more to social theory, the relevance of which has been previously underplayed, than to economics. Accounting history becomes less econocentric and more sociocentric. In this way, the rise of the new accounting history may also be seen as a contest between the rule of economics in accounting research and social theory. (Funnell, 1996: 39)
The way forward: Critical traditional accounting history
[The historian] whose breast is oppressed by a present need and who wants to cast off his load at any price has a need for critical history, that is, history which sits in judgment and passes judgment. (Nietzsche, 1874)
To summarise the present situation, the emergence and current dominant status of socio-historical accounting history has not addressed the unanswered ‘why’ of the traditional antiquarian accounting history that some believed it sought to address. It has created a new field of enquiry, but its exponents have not applied themselves to improving what is known and understood by extending the work of earlier generations of antiquarian accounting historians. They have not done so because they have not revisited the sources. Not because such improvement is not possible. Socio-historical enquiry has never used the Italian account books that were and still are at the heart of medieval accounting history and, when it appears to refer to the conclusions of those antiquarian accounting historians who studied them in the past, which it seems to accept unquestionably, it short-circuits reference to medieval bookkeeping practice by referencing instead the elementary 16 bookkeeping system described by Luca Pacioli in 1494 (Goldthwaite, 2018).
Consequently, at a macro-level, the beliefs reflected in the antiquarian literature on pre-modern double entry bookkeeping and financial accounting history continue to be the result of a combination of a self-confessed dominant present-minded teleological research method, which involves starting from today and reasoning back, explaining things from the past based on their present-day purpose (De Roover, 1958: 42); 17 a lack of awareness of business problems (Lane, 1944: 153) and business context (Hopwood, 1983: 289); a technical focus combined with an absence of critical thinking and reflection when primary sources are analysed and described (Hopwood, 1987: 207–208); and a belief in the Whiggish concept of continuous improvement (Funnell, 1996: 54). In other words, a surfeit of teleologically justified antiquarianism and a failure to ask, ‘why’. A similar situation was faced by economic historians in the 1920s, and they found a different solution.
Until that time, their field had been largely left to economic theorists to explore for more than 75 years without the aid of historians and, as a result: unsupported and unrestrained by research, [they] inevitably produced notions unrelated to historical facts. … To them epochs of history were successive stages in the uninterrupted ascent of mankind … [Thus,] Karl Bücher [1901], … regarding the institutions of our civilization as the product of a growth extending over the whole of European history, … was bound to assume that in the earlier stages credit could only play a minor part in economic life. … Bücher's general theory of economic evolution now [in 1928] forms one of the axiomatic assumptions of historical research; and his views on credit were naturally adopted by [economic] historians … [Yet,] the abundance of [evidence] … clearly demonstrates [that this is incorrect:] credit commonly entered into the commercial practice of the middle ages. (Postan, 1928: 234–235, 238)
Since then, economic historians and historians of other disciplines have long ceased using Bücher's teleological Darwinist approach, looking at the present and searching the past for evidence of progress towards that present. They have acquired awareness of context, and the necessary historical knowledge and understanding from the evidence that has survived which, after the twelfth century, is increasingly to be found in books of account (Ciambotti, 2009). They became critical traditional historians.
If accounting historians were also to embrace the methodology of critical traditional historians, they could begin anew to investigate the creation, adoption, use, purpose, and dissemination of double entry bookkeeping up to the nineteenth century and the forms of financial reporting during that period – the reasons why double entry bookkeeping began; the reasons why it was used; the needs of business it served and those of other organisations and society; 18 and why, how, and by whom it was done, with variants in practice investigated and explained, looking past the method to the benefits its use brought, whether social, economic, political, institutional, or otherwise, and the needs it and any reports prepared from it addressed.
As previously mentioned, some of the writers of this historiography, such as Basil Yamey, have speculated about these issues, but very few use the environment or surrounding context to reach their conclusions or even, in some cases, as identified by Goldthwaite (2018), books of account themselves. Studies founded in deep context are absent; and knowledge of context is fundamental to a critical methodology. It may be obtained from the literature of other disciplines, such as history, business history, economic history, social history, legal history, and political history. It is also obtained from merchant manuals and manuals on bookkeeping, of which respectively thousands and hundreds exist from 1470 to 1820. 19 Context is obtained far less directly from the account books alone, or from what is known of individual firms, governments, or other entities of which the account books have survived.
In engaging with accounting history critically, previously studied primary sources – account books and other related documents – would need to be revisited and the reliability of secondary sources would need to be reviewed. Compared to earlier generations of accounting historians, a more contextually informed, objective, and critically reflective approach would need to be adopted. In this way, critical accounting history could provide what the current purely technical antiquarian approach to pre-modern periods has not: insights from the past that may be relevant to the present, and to the future.
The benefits of doing so go beyond a rewriting of accounting's past. It would not simply mean that we might learn the lessons from the past of which we are currently unaware, making future developments in accounting thought better informed, less likely to emphasise the less relevant or the less important, and more likely to include the previously unconsidered it reveals. It would also mark the beginnings of respect for a discipline from those in other fields of history who believe accounting history currently has little to contribute to knowledge and understanding of medieval and early modern business practice.
The consequent improvements in knowledge and understanding of business practice and of the emergence, adoption, use, purpose, and dissemination of double entry up to the nineteenth century would add to and complement the work already done by historians from other disciplines, and by socio-historical accounting historians. It would also provide a route map back to medieval times enabling better informed evaluations of, for example, how original and innovative eighteenth-, nineteenth-, twentieth-, and twenty-first century bookkeeping, accounting, costing, and auditing practice was compared to what had been done before; and identifying previously undetected changes in practice and priorities, so setting research agendas for future scholarship.
In doing so, a critical history of accounting would emerge of a type that both Hopwood and Chambers were advocating 40 years ago, one that challenged, compared, contrasted, evaluated, and assessed, providing the depth of knowledge and understanding that the historiography of accounting before the nineteenth century lacks. Its emergence would open doors to research and to its publication in this field that over several years have become increasingly closed. Most importantly, it would set aside the misunderstanding and confusion in the current literature, providing a much sounder foundation on which others, including socio-historical accounting historians could rely.
Conclusion
In order to follow this path towards a critical traditional history of accounting, some adjustments would need to be made to how accounting historians describe what they examine. Yamey's conflation of bookkeeping and financial accounting (Miller and Napier, 1993: 637) would need to be set aside, with the method of double entry recognised as separate and distinct from accounting – more precisely, financial reporting – as it was to Fabio Besta. Terminology used would need to be defined, so that meanings are clear in all cases. Secondary sources would need to be critically read, drawing out what is unambiguous and relevant. In many cases, it may be necessary to declare why some sources are not being used. If used as sources, bookkeeping manuals would need to be treated not as Yamey did – primarily as surrogates for account books – but as both supporting evidence of context and as diagnostic tools when investigating the thousands of account books and other business records that have survived from pre-modern times. Context and evidence would need to be recognised for what it is, and what it tells embraced. But that is not all.
Those who accept the challenge of becoming critical traditional medieval and early modern accounting historians would need to read more widely, embracing the currently largely ignored historiography of other disciplines for both the context they reveal and the uses of medieval and early modern account books they describe and analyse. Doing so would facilitate more meaningful dialogue and greater synergy between accounting historians and historians of other disciplines (Goldthwaite, 2018: 251), boosting levels of knowledge and understanding in those related fields, as well as in accounting history. These scholars would also need to cease to ignore anything written in any language other than their own, which is currently evident to anyone examining the lists of references across a sample of English language publications in this field. Only through a more interdisciplinary and multi-lingual approach that recognises these needs will a robust and demonstrably evidenced history of bookkeeping and financial accounting before 1800 appear to replace the ahistorical myths, misunderstanding and confusion that have been established upon a foundation of knowledge of ‘what’ and ‘how’ over the past 120 years. Doing so, should make inroads into the current rejection of traditional accounting history by journal editors who would be able to see the manner in which a critical approach has rejuvenated this field, though that may take considerably longer to materialise than it took for the current contrary situation to be established.
Finally, with respect to all those who believe there is no need to either know more or understand what is known about pre-modern double entry bookkeeping and financial accounting, with few exceptions and compared to what we could know, we know very little and understand almost nothing, because we have never asked ‘why’: it is time to start again.
Footnotes
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
The following updates were made to this article:
1.
The original sentence “It gives anyone unaware of Besta's single–minded focus on assets and liabilities the impression that ‘we know everything’ of its use in that period. Besta could see double entries, that is, duality, from the thirteenth century but, in addition to the features required by De Roover's (1956) definition of double entry bookkeeping, Besta's definition (Martinelli, 1974: 217–223) required a bilateral layout in the individual accounts (Martinelli, 1974: 217–218).” has been changed to “It gives anyone unaware of Besta's single–minded focus on assets and liabilities the impression that ‘we know everything’ of its use in that period. Besta could see double entries, that is, duality, from the thirteenth century but, similarly to the features required by De Roover's (1956) definition of double entry bookkeeping, Besta's definition (Martinelli, 1974: 217–223) required a bilateral layout in the individual accounts (Martinelli, 1974: 217–218).”
2.
The original sentence “Neither those convinced, like Besta and Yamey, that the only reason to use double entry bookkeeping rather than simpler forms of bookkeeping is to identify movement in net assets, or capital. Nor those convinced that double entry bookkeeping has benefits Yamey, in particular, did not recognise as justifying its use.” has been changed to “Neither those convinced, like Besta, that the only reason to use double entry accounting rather than simpler forms of bookkeeping is to identify movement in assets and liabilities; or, following Yamey, to identify movement in capital.”
3.
The original sentence “The author also broadens Fabio Besta's insistence on identifying movement in capital as the reason double entry bookkeeping is adopted to equally apply it to his version of ‘single–entry bookkeeping'.” has been changed to “The author also broadens Fabio Besta's focus on identifying movement in assets and liabilities as the reason double entry bookkeeping is adopted, which is misinterpreted in the literature as identifying movement in net assets/capital, to equally apply it to his version of ‘single–entry bookkeeping’.”
Notes
Correction (October 2025):
Article updated; for further details please see the Article Note at the end of the article.