Chapter 3: Concepts, codes and meanings: bridging knowledge and records – Records Management and Knowledge Mobilisation


Concepts, codes and meanings: bridging knowledge and records


Records are a distinct category of information that must conform to strict management disciplines, unlike many other categories. There are two aspects of knowledge and information: as a physical entity, similar to explicit knowledge, and a cognitive process, similar to tacit knowledge, and both can be true. A codebook is necessary to encode and decode working knowledge as information and holds part of the meaning. Knowledge is related to action and in three forms: encoded, embedded, embodied. Records are a kind of encoded knowledge that capture working knowledge. Knowledge mobilisation mediates between working knowledge and action.

Key words


embedded knowledge

information paradigms

knowledge and action

The famous Rosetta Stone is often cited as a metaphor of the problems of digital preservation – what happens to the record when the ability to read the format is lost? – but it offers other lessons for records management too. The Stone was produced two millennia ago in Egypt as a priestly decree with, as is well known, the same text repeated in three languages: Greek, then the language of the modern administrative state; Ancient Egyptian hieroglyphs, endorsing the document with the sanction of the established church, just as Latin is used in Papal encyclicals today; and the demotic language of ordinary people. The copy in the British Museum is one of several made for public display in temples around the country, to be rediscovered by French, and requisitioned by British, troops in nineteenth-century Egypt, and eventually providing scholars with the key for uncovering the lost meaning of hieroglyphics, using the Greek version as a ‘decoding language’.

Decoding, though, was also an issue when the Stone was first made: as a formal record, it is written in three different languages in order to communicate its message clearly to three different communities – priests, administrators and the general public – and conveying the power associated with use of each language. In just the same way, modern records that are produced by specialist communities are often inaccessible, in their full meaning, to others who do not have the knowledge and background to decode them correctly (rather like an EU treaty, as noted in McGregor, 2010). As explicit statements, they must be reinterpreted for different audiences.

Contemporaries would also have decoded the document in different ways. The text records the priests’ recognition of the 13-year-old Ptolemy V as new ruler and god (a necessary qualification at the time), gratefully acknowledging some concessions to them – confirming the religious centre of Memphis, rather than the administrative city of Alexandria, as the seat of formal ceremonies, together with some attractive tax breaks for priests. More subtly, it is a public statement of the adjusted balance of power between church and state on the accession of a less powerful child king. No doubt contemporaries would quickly have worked out the implicit meaning, in the same way that seemingly inscrutable statements of modern autocrats are examined for their implied messages.

There are two kinds of knowledge, and two kinds of meaning, at work in the same record: one explicitly recorded in the language of bureaucracy (though requiring the specialist ‘jargon’ to decode it), the other implied ‘between the lines’ and able to be decoded with an understanding of the current institutional rules and values – knowledge of the way things work in practice. Interestingly, the record was amended on its edge with a later inscription ‘captured by the British in 1801’, conveying the power relationships of a more recent era.

Records and knowledge have an uneasy relationship. Intuitively, they seem to be closely related, but the connection is ambiguous: the record can contain several layers of meaning that require background knowledge to decode. Knowledge is fluid and multi-faceted and only some of those aspects are captured in a record. Records aim to form a well-defined, structured and managed body of corporate knowledge, based on the explicit activities and functions of the organisation. Knowledge is the quality necessary for the members of that organisation to successfully carry through those activities. Both reflect issues of different interests, institutional rules, values and behaviour, but these are often embedded below the surface and take some effort to unpack. The relationship between the two concerns the people associated with each, more than the systems that process them.

This chapter lays the foundations for an integrated model that brings together records management and knowledge mobilisation as complementary aspects, to help think about the relationship between them. While of necessity somewhat theoretical, it does not aim to develop a universal theory of records and knowledge, either by redefining records management or by attempting to navigate the ‘what is knowledge/information’ quagmire – such discussions do not easily inform practical action. Rather, the intention is to develop a working framework for considering pragmatic issues of regulation and innovation, policy-making and delivery, adding value and transforming the public sphere, discussed in later chapters.

Distinguishing records, information and knowledge

Firstly, though, we need to note how the terms records, information and knowledge are used here. These terms (especially the latter two) are often used quite loosely, sometimes interchangeably, and to signify completely different meanings, a practice surprisingly difficult to avoid. The intention here is to clarify the sense in which they are intended in this book, without entering into a debate on which meaning is correct.

Records and information

The term records has a formal definition set out in the ISO15489 standard (ISO, 2006), which lists the particular characteristics – authenticity, reliability, integrity and usability – which records should be able to demonstrate. These are closely related to the idea of evidence; but some might argue that these same characteristics should be displayed by well-managed information in a corporate system and see little difference between the two concepts. Information, though, can serve many needs other than the evidential: sometimes it is sufficient to be indicative rather than precise, ‘good enough’ rather than complete, supportive rather than authentic. A record here is taken as belonging to a category of information which is subject to particularly stringent management disciplines – more so than in most information management systems – because these four characteristics are of over-riding importance: without assurance of their presence, it does not qualify as a record. Many other information management systems do not, and could not afford to, meet this standard.

Both types of system may also, of course, be closely integrated and they both have one particular relationship with knowledge in common, which emerges in two main ways. Firstly, records management systems, if they follow the ISO standard, are structured on the lines of a functional categorisation of record types; records are grouped together according to the function (finance, public relations, childcare, transport strategy, etc.), or a component activity of the function, which produces them. By this means, they outline the perceived anatomy, as it were, of the organisation – the inherent nature which remains constant even through restructuring of teams and units. Structured information management systems can present a similar map of the organisation, for example, in the form of a relational database, which decomposes the surface line operations and workflows into a single, integrating, ‘picture’ of the organisation as data.

Secondly, these same workflows and operational lines of business both take and process organisational information in certain rule-defined procedures that are designed into the system as a whole; they also shape the kinds of records that are produced by them. Both the information processing and record-keeping aspects work with a particular understanding of these rules and norms which define what can be done with them, but the rules are not normally made explicit (except perhaps in a system manual); they are embedded into the information system, either as automated processes or as assumed and unquestioned working practices. Novices absorb them as they learn, and experienced staff do not even think about them.

These embedded procedures have, though, been thought about at some point; they have been developed and refined over time and are based on particular ideas and ways of thinking. Behind the procedures of a routinised decision-making process, for example, are certain assumptions on the way that a decision is made, the criteria and data that are relevant and the available options that result – a model of decision-making. The old, hackneyed adage that information management is ‘delivering the right information to the right person at the right time’ begs the question of who, what and when is ‘right’. There is always an answer (right or wrong) to that question, but it has been given a while ago and is now assumed or unspoken – the model is also embedded as a deeper layer in the system.

Embedded knowledge, then, is present in the assumptions of records management as it is put into practice in any given situation; embedded both in procedures and ways of working, and in the very structure of information in the records. It consists of both the rules for carrying through some activity and a mental model of what that activity involves; and the mental model is in turn built on a set of norms and values that define the features which are important and relevant. Embedded knowledge regulates an activity to ensure that it is always carried out consistently and fairly – in that the same inputs result in the same outputs – without having to think through each case from first principles each time. There are disadvantages, though; when change becomes necessary, it is more difficult, because the change initiative has to challenge not only the rules, but also the underlying model and the assumptions which drive it. New knowledge has to thoroughly unpick the old – to decode, take apart and re-assemble it using different assumptions and values – before change can establish itself. The more deeply embedded the knowledge, the more difficult is the innovation.

Information and knowledge

If records can be seen as a special type of information, distinguishing between information and knowledge is more complex. This is partly because both terms are commonly used with two different meanings, but interchangeably for each. The first meaning of information and/or knowledge conceives of it as a physical objective entity which can be passed from one person to another. Knowledge, expressed as information, is encapsulated within a physical or electronic artefact so that it can be communicated from one person to another; in this sense, we might speak of a textbook, a research paper, a website or a documentary film as containing knowledge which has been articulated by the author and which can be interpreted by many others without any loss of meaning.

Encoded knowledge is expressed in terms of an accepted ‘language’ which is understood (or must be learned) by the recipients – professional ‘jargon’, accepted disciplinary concepts, technical languages such as statistics or the argot of street language – and which is used to decode the meaning. One piece of encoded knowledge does not exist in a vacuum; it makes reference to many other such pieces and takes part of its meaning from that context. Taken together these make up both a body of accepted knowledge on a topic and the ‘language’ which is used to express it. In this view, knowledge is a form of information expressed according to a particular set of rules and meaning lies with the producer; successful communication is determined by whether recipients ‘get the message’ or not.

In the second meaning, knowledge (and information) is seen as a cognitive, sense-making process rather than a product. Information is an inherent, or potential, quality in something and must be interpreted to make sense. New information is processed by the recipient in the context of an existing system of personal knowledge, beliefs, values and assumptions; the reader of the physical artefact tries to make sense of the information which it contains by fitting it into the frame of what they already know or believe. The success of this process of sense-making may depend on whether the information makes a ‘good fit’ with the recipient’s existing mental structures, or perhaps on whether it effectively challenges and changes existing beliefs and assumptions. As Raber (2003) suggests, information is a matter of impact – how it is used; how we decide to put it to work says something about our view of the world around us and our place within it. Meaning lies with the consumer; embodied knowledge is in the mind of the individual.

Do we have to choose between these two seemingly contradictory ideas, or can both be true? In the context of information, Ellis (1998) suggests that they can: we do not have to choose between these two paradigms, as he calls them: it is an and/both, not an either/or, situation. He draws a parallel with the example of modern physics, in which rival theories on the nature of light, as either a stream of particles or a wave-form, were eventually resolved by a realisation that it has a dual nature – it behaves as both a wave and a particle stream – and is consistent with both the Newtonian and quantum physics paradigms. In a similar way, information can be understood to possess a dual nature, as both an entity that can be passed from one person to another and also a mental activity inside the mind. Each way of thinking about information or knowledge highlights different aspects and uses different sets of criteria to draw out insights; what is important is to know which is being used at any given time.

Knowledge and action

This is somewhat similar to the common distinction between explicit and tacit knowledge: that some forms of knowledge can be made explicit, by eliciting, capturing and encoding; but others are inherently tacit in nature and cannot be captured in this way. Tacit knowledge underlies the skilful performance of an expert, but she is not aware of it and cannot articulate it: it can only be learned by doing. Therefore, in this view, attempts to formalise and transfer such knowledge are doomed to failure. But this is not quite the concept as originally put forward by Michael Polanyi (1958). He gives the example of an expert swimmer, who stays afloat without being consciously aware of how; nevertheless, the principles of buoyancy that he is unconsciously using can be formally captured. What is explicit for one person may be tacit for another, depending on the context and degree of expertise.

The tacit/explicit dimension was initially formulated to describe forms of personal knowledge, and it is particularly helpful in understanding how skills-based learning takes place in organisations. In doing so, it makes clearer the link between knowledge and the idea of action. Knowledge is distinguished from information by the ability to put something into practice: to use the information in order to achieve something. It seems intuitively obvious to say that these two are linked: we choose to do one thing rather than another – study a particular topic at college, follow a career, choose a pension plan – because of the knowledge we have about it: not just the published facts, but also our own beliefs and assumptions on what the right course of action is, what is expected of us by family and friends, what is acceptable within our own culture or religion, among other factors. And yet many models of information management suppose otherwise: they assert that individuals make rational decisions by gathering and processing all relevant information, evaluate the options and make a choice to maximise the personal use value.

This is the crux of the problem:

 Is knowledge mobilisation a matter of capturing, ordering and formalising, and making explicit as a physical entity, the personal and collective knowledge of some, so that it can be transferred and independently assimilated by others? Initiatives attempting to record the knowledge of key staff leaving a post by asking them ‘What do you know?’ so that the new recruit can learn from it, would answer: yes. The issues are then about improving the capture and encoding process, and bringing about the culture change that encourages people to use the resource.

 Or is it a matter of building an environment which encourages individuals and groups to exchange knowledge in ways that are meaningful to each other, that they can make sense of in the context of working and professional cultures and institutional expectations? A new recruit will see things in a different way (possibly why they were recruited); a face-to-face handover might be more productive for both. The issues here are about how to make or augment effective connections between people in ways that focus action on objectives and outcomes, and make the knowledge they embody more productive.

 Is records management a system for capturing the objective truth about organisations as corporate memory, aiming to encompass the full complexity of activities and regulate their proper application? Or does it represent one, necessarily selective, view of the organisation incorporating both embedded knowledge about values and assumptions that will change over time and encoded knowledge fixed into documents?

This book suggests that both are, in fact, all of these things, in different measure, and that the relationship between records and knowledge is one of balance and appropriateness in a period of rapid institutional and governmental change.

Knowledge in organisations is of three kinds:

 embedded in institutional rules, organisational procedures, information systems and legislation;

 encoded in physical and digital documents and products of various sorts;

 embodied in the members of the organisation and their associates.

All three are related to action, and the difficulties essentially lie in understanding the inter-relationships of the three forms and the range of influences which each are subject to. Without some sort of framework to map these issues, to help devise strategies which encourage a synergy between the three kinds and to select the tools and techniques to put the strategy into practice, initiatives can easily be misdirected.

Role of the codebook

To encode knowledge, to take what an individual or group knows and make it explicit, by writing it down, say, requires some form of code for it to be produced in. Both the producer and the recipient must have a common understanding of, and access to, this code for the meaning to be transferred intact. This is more than just words on the page; most writing does not explain everything from first principles, but rather in a kind of shorthand which assumes a level of understanding about existing ideas and concepts, explaining new ideas in terms of them. Specialist language works like metaphors to refer implicitly to this background understanding, without stating it; and this context-dependent knowledge is necessary in order to interpret the words correctly. This is the well-known problem of the ‘stickiness’ of tacit knowledge; it works in one context, but does not easily transfer to a different one.

Writing about the economics of knowledge, Cowan et al. (2000) develop the idea of a ‘codebook’ to help explain this problem. A codebook contains the shorthand, or the contextual background, necessary to produce the encoded knowledge; producers need to make use of it and recipients need to have access to it. Many government documents are written in this way – policy briefs, project documentation, case reports, statistical analyses – where the unspoken is as important as the spoken, sometimes perhaps deliberately to restrict understanding to a small circle. For most outsiders, professional discussions in science and technology are impenetrable without understanding of the relevant codebook and this can take much time and effort – perhaps years of training – to acquire. The transaction costs, then, of transferring explicit knowledge are more than the effort which it takes to produce or distribute it; although text can be duplicated as information very cheaply, each recipient must make sense of it for themselves by mobilising their own cognitive resources (Foray, 2006). The extra costs of putting it into practice are associated with each and every recipient.

Cowan and Foray also point out a further complication: that the encoded knowledge itself adds to the codebook. Each new piece of knowledge in a topic area adds something to the overall body of assumed, contextual knowledge: as the discipline progresses, the codebook keeps changing, incrementally. It is not simply a matter of translating into a different, fixed language, because the translation itself changes the language. Professional communities have developed robust and well-established systems for coping with this problem – networks of seminars, conferences, specialist journals, peer review, formal education – which rely on informed debate and dialogue to filter new ideas and affirm new knowledge. These systems are backed up with commonly shared values and norms of behaviour, accepted formal and informal rules which make it work: defining, for example, acceptable research methods, collaborative spaces, opportunities for mentoring and development.

Four categories of situation are identified by Cowan et al. (2000):

 where knowledge is encoded and the codebook is present and accessible: the knowledge is well established and regulated, and there are mature distribution channels reaching out to a wide community (such as manuals, textbooks, standards, etc.);

 where knowledge can be encoded, but the codebook is implicit: the knowledge is accessible to a limited and experienced community, but for the outsider, dialogue is much more difficult, or the meaning just opaque, across the community boundary; the use of professional ‘jargon’ is a typical example;

 where knowledge is not yet articulated, although it could be in principle, and no codebook (yet) exists: typically, in the emergence of new knowledge areas or where innovation produces rapid change;

 where the knowledge cannot be articulated in principle.

The point to note is that the problem of knowledge transfer is not just about extracting embodied knowledge and encoding it as a physical entity; in fact, misdirected efforts may only make the situation worse. If, as the trends in government transformation suggest, successful policy outcomes result from coordinating interaction between different communities participating in the process, then the problems are to do with building understanding between different kinds of knowledge, different encoding mechanisms and different sets of assumptions and values. Knowledgetransfer activities should be more concerned with making the connections between communities that support collaborative knowledge development. This will involve developing shared codes, shared meanings and shared values.

For members of the public, a common experience of government is the need to deal with several different branches for aspects of what, to them, is the same issue. For example, families with children who have been classified as having special needs must deal with teachers and the school authorities, social workers and healthcare workers, each separately. Each of these professional groups projects their own particular slant on the child in their environment, to some extent overlapping, to some extent inconsistent with each other, each seeking information from forms and interviews. The answer to a ‘joined-up’ response is not to link all these elements together in one, integrated but unchanged, information process: this simply compounds the problems of disjunction and puts the burden of resolution on the service user. A joined-up response should try to present the user with an integrated knowledge process, expressed meaningfully to all concerned. Table 3.1 summarises the comparative features of the physical and cognitive knowledge paradigms.

Table 3.1

Knowledge paradigms

Physical Cognitive
Encoded/embedded Embodied
Being informative Being informed
Focus on source producing new knowledge Focus on destination producing new knowledge
Knowledge of external codebook required to interpret Tacit knowledge or implicit codebook required to assimilate
Rule-based/empirical Pattern-based/experiential
Collective knowledge Personal knowledge

Reconciling structured records and fluid knowledge

How, then, to reconcile the structured world of records with the fluid world of knowledge? Records management aims to capture the essence and purposes of an organisation as a formal information model, a ‘deep structure’ that is sustainable over time and independent of surface organisational change. It is a syntactic view of the organisation in both senses: showing the rules and policies through which the functional elements are combined to form the whole and conveying the meaning inherent in the relationship between those two.

Knowledge mobilisation aims to support the generation, sharing, assimilation and practical application of the most effective knowledge available in order to improve the outcomes that are the purpose of the organisation. It is concerned with the semantics of the organisation: it turns up the focus to show the more specific meanings understood by the programmes and activities, and puts these in the context of the culture, norms, values and behaviour of the organisation. Both are concerned with the same aspects, but stress different facets.

Records structures are the warp of the organisation, the tightly strung threads that form the basic structure in the loom; and knowledge practice is the weft that weaves through these threads to form many and varied designs across the piece.

Knowledge is an action-based, interactive sphere in which individuals collaborate with each other to produce the desired outcomes, drawing on a range of ideas, interests and the institutional ‘rules of the game’, the formal and informal ways of working. These individuals struggle with outcomes that are hard to pin down, trying to make a workable relationship between the framework of what they know and their understanding of the results they are trying to achieve – to work out how to make things happen. This we might describe as working knowledge. Experience gained from attempts to put this knowledge into action, to produce the desired outcomes, adds new working knowledge for individuals and groups. This is the space of the reflective practitioner who is engaged in analysing practical experience, questioning established practice, developing new ideas and thinking, incrementally building up the knowledge base and finding new ways to translate it into action. It is also the space where the encoding process begins, where knowledge embodied in practical action and experience starts to become encoded and the codebook itself evolves as part of the same process. The fundamental challenge for knowledge mobilisation lies in finding ways to mitigate the tensions between working knowledge and knowledge in action, and to help improve the quality of the fit between knowing and doing.

Records aim for solidity and a tangible presence, capturing the fluid and messy struggle of working knowledge at fixed points in time and freezing these into an official narrative. Records are themselves struggling to create structure and order from these changeable knowledge processes. In this sense, records management systems themselves make use of a kind of codebook, an overarching structure which gives meaning to each separate part of its contents. Given an overall classification or category structure, each new record has to be located somewhere within it, and in part takes its meaning from that location. New classes or categories may be created as new sets of records emerge, but these too must fit within the overall arrangement. Influence also works in the opposite direction: new records tend to be created with an eye on fitting within the existing structure. If it is influential and taken to heart, the knowledge structure of a records management system has the effect of channelling and restricting – encouraging into pre-formed categories – the working knowledge of its users.

Although records and knowledge intuitively seem to be complementary, they can also find themselves in opposition, depending on the quality of fit between the two at any one time. If the records management codebook, oriented towards regulation and governance, and the working knowledge codebook, oriented towards practical action, make a poor fit with each other – if each is structuring the world of action in a different way – then serious disconnects result in user frustration and culture clashes. Initiatives which attempt to extract embodied knowledge from the right-hand side of Figure 3.1, asking staff ‘what do you know about?’, hoping to encode responses as inert information directly into the record on the left, ignoring the codebook issue, are unlikely to succeed. Working knowledge is the meeting ground for records management and knowledge mobilisation. To be able to see the two together, we need to build an integrated model in which the complementarities, and the tensions, can be contrasted and balanced. An integrated model does not resolve all issues; rather, it provides a framework for comparing perspectives: between getting things straight and getting things done.

Figure 3.1 Records, knowledge and outcomes

Key principles

 Records are a strong form of information, more stringently managed than most forms.

 Knowledge is information in action, characterised by the ability to put it into practice.

 Both records and knowledge are context-dependent; they take their full meaning not just from content, but also from the organisational, institutional and mental structures and norms in which they are located.

 Encoded knowledge = coded content + codebook; some of the meaning is in the codebook.

 Both have layers of meaning for different communities, in different contexts, using different codebooks.

 Building bridges between codebooks is central to an integrated approach to records and knowledge.