List of criteria for imaging and textual conversion for preservation.
Results of a DFG project
English version is available only in electronic form. The German Version is available in printed form. It can be obtained by sending a command to the Archivschule Marburg. The German version contains forms for the identification of intrinsic value, which are not reproduced here.
From 1995 to the end of 1996 the Marburg Archive School carried out, with financial support from the Deutsche Forschungsgemeinschaft (DFG), a project to determine the intrinsic value of archive and library material. The aim was to develop a list of criteria which would allow a speedy and rational selection of holdings requiring, on the one hand, measures to preserve the documents in their original state, or else conversion to durable materials.
The first step was to review the available literature. A questionnaire was then drawn up and sent to a selected group of archive and library experts. At a meeting on 18 April 1996 the reults of the survey were discussed and a provisional list of criteria was made. The final version of the criteria was agreed on by a smaller circle of experts at a further meeting on 7 November 1996.
The participants at the first meeting were:
Prof. Hans Bohrmann, Institute for Newspaper Research, Dortmund; Dr.Nils Brübach, Archive School, Marburg; Dr.Eva-Maria Dickhaupt, Research Centre for Personal Documents, Marburg, Dr.Christoph Drüppel, Esslingen County Archive and Federal Committee of Communal Archivists, Esslingen; Dr.Peter Fleischmann, State Archive, Nuremberg; Prof, Horst Gronemeyer, State and University Library, Hamburg; Dr.Elmar Härtel, Herzog August Library, Wolfenbüttel; Dr.Joachim Jaenecke, Berlin State Library - Prussian Cultural Collections, Berlin; Dr.Robert Kretzschmar, Baden-Württemberg State Archive Administration, Stuttgart; Dr.Angelika Menne-Haritz, Archive School, Marburg (Chairperson); Michael Merchel, Ministry of Home Affairs for Saxony, Dresden; Dr.Werner Röder, Institute of Modern History, Munich; Dr.Helga Unger, Bavarian State Library, Munich; Ulla Usemann-Keller, German Library Institute, Berlin; Dr.Hartmut Weber, Baden-Württemberg State Archive Administration,Stuttgart.
The participants at the second meeting were: Dr.Nils Brübach, Archive School, Marburg; Prof. Horst Gronemeyer, State and University Library, Hamburg; Dr.Joachim Jaenecke, Berlin State Library - Prussian Collections, Berlin; Dr.Angelika Menne-Haritz, Archive School, Marburg (chairperson); Ulla Usemann-Keller, German Library Institute, Berlin; Dr.Harmut Weber, Baden-Württemberg State Archive Administration, Stuttgart.
Work on the project was supported by a work group of students from the Archive School, Marburg. They were responsible for some papers incorporated into the project report and also for taking the minutes at the first meeting of experts. Members of the work group, under the leadership of Nils Brübach were:
Dr.Irmgard Becker, Dr.Peter Exner, Dr.Martin Fimpel, Irmgard Fliedner, Dr.Uwe Grandke, Dr.Thekla Kluttig, Dr.Achim Krümmel, Dr.Thomas Küster, Kerstin Langschied, Dr.Christoph Popp, Dr.Barbara Rößner, Dr.Katharina Schaal, Dr.Andreas Weber, Dr.Norbert Wex, Caroline Wilhelm.
Archive and library material, as defined in this report, is everything which is stored and made available for use according to the function of archives and libraries. It includes texts and physical contexts. The production of printed works, minutes and notes, as well as written communication within administrations in the course of decision-taking, produces concrete material. Statements and messages are put into a form that makes them as easily transmittable and as directly communicable as possible. The formal features employed are partly deliberate and partly pure convention. They serve to emphasize the salient points of a text, often rendering them comprehensible, or can themselves become the subject of research which elucidates their contexts, their production and genesis.
External formal features, i.e. the testimony, of archive and library material express their message by non-textual means. There are many varied non-verbal possiblities. They are unobtrusive and, in contrast to the text, their significance is not immediately recognisable. Their important characteristics become apparent as questions arise in the use of the text. These non-verbal features make the text accessible and must be interpreted in order to understand the text fully.
At first sight it is therefore easy to overlook the external formal features of a text, to take them for granted and overlook their evidence. However, when the text has to be placed in its geographical and chronological environment in order to ascertain more about its origins and background, it is necessary for these factors to be considered. Conclusions reached about the way a text evolved furnish authentic testimony as to its origins and purpose.
The production of concrete material as an expression of communication, be it knowledge, be it the elements of a decision, places the required texts in a chronologically and geographically definable environment. The external formal features link texts to the material world and thus to their history and their transitoriness. Archives and libraries render their holdings accessible through their history, as revealed by their external formal features. Their condition when they come into the possesssion of a library or archive is the most telling evidence. Every succeeding alteration weakens it. The task of preservation is therefore complementary to that of ensuring permanent accessibility. Preservation must stabilise the present evidential value and prevent any future change.
External formal features play a particular role for archive and library material. They are the result of written communication, independent from the number of copies or the number of editions.
Setting something down in writing is of particular significance in the discussion about intrinsic value. It stabilises the content. As long as subject matter is passed on by word of mouth, or as long as different opinions presented in a discussion group are not set down in the minutes they remain in a state of flux and can be changed unnoticeably or in an underhand manner. Every written record, be it a note in the files, be it the manuscript of a literary or scientific work, stabilises the content expressed in it. The content is linked to certain external features which bind it to material objects. Hand written manuscripts, whose different chronological phases can be apparent through the different colours of the ink, can be sufficient for this. This applies to the production of a book, a journal, a newspaper or to a note made for a file, which makes a particular observation relevant for future decisions.
Writing is a form of using language to describe observations, in order to communicate them. The differing media in which language is expressed give the content their characteristic features. They are the basis for different effects and functions of the content expressed.
Functions become relevant which are latent but usually not consciously used. Language formulates content and gives it a form which distinguishes it from others. This description in its oral form can very quickly change and be forgotten. Writing lends language additional functions by fixing the formulation of the contents, so that they can be remembered and so used again. However, it reduces the expressiveness of the spoken word given by the tone of voice, the speed of delivery and onomatopoeic, not strictly verbal, expressions. Writing both limits statements and makes them more precise. At the same time it extends the stock of signs in contrast to the sounds of speech by means of graphic symbols such as ticks and underlining, by additions such as marginal notes, through script form and the colours of the writing materials, and it achieves an additional, very useful, effect: writing fixes spoken formulation in time. Writing helps memory to externalise. Written notes can be picked up at any time and made use of again through renewed reading.
A written message, in contrast to communication in the reduced oral form, allows a more effective use of language. An oral or written form of communication can be selected according to the degree of directness and frankness desired in the matter to be communicated, or the precision and ease of recall.
In order to achieve communication, in other words, in order that the verbally formulated subject matter can be understood, it is necessary that the message itself as well as the intention conveyed by the form of the message should be understood by the listener or reader. Oral as well as written communication must be interpreted in order to make the subject matter credible or to differentiate it from other statements. A written sentence, as well as an oral statement, can be misunderstood or its source can perhaps be considered unreliable. It requires a supporting context to safeguard it. The deductions made from context about the provenance of a text is such an obvious cultural skill in the employment of language, and so also of documents, that it does not have to be specially learnt or explained. The loss of the physical contexts, and therefore the aid they give in interpreting and understanding a text, as can happen when texts are digitalised, makes their importance particularly clear.
Electronic mail and digital publications extend the choice of available forms and functions beyond those of oral and written communications. They bring a new flexibility to the written word. Electronically written scripts are no longer automatically fixed in time and their formulation is no longer final. Electronic communications can employ greater precision and textual memory aids, but they can also discard them when this is necesssary for a quick and direct reaction or innovative approaches.
Flexibility, however, implies complete and continual variability of the text, despite its optical representation, for example on a monitor. This continual variability also annuls the chronological fixation of a text. The numerous versions of a text produced and revised by electronic means are repeatedly rewritten. The text is no longer an entity but rather a continual process, as long as it is electronically retrievable and processable. Interim versions that have been deliberately stored are only steps in the process and cannot completely describe it. Their degree of reliability is, however, dependent on their stability. When this is not given the ability of recall is limited, as there is no possibility for a specific recovery of past material. This would entail the simultaneous availability of all knowledge and experience, which is not possible. The possibility of recovering past material allows the storage of knowledge not immediately required. For this reason the degree of variability above what is currently necessary must be deliberately reduced. This calls for for new methods for fixing texts chronologically and stabilising them in the case of electronic word processing. These methods must be deliberately employable while retaining flexibility.
The digitalisation of documents removes the material influence of the script on the subject matter expressed by it. Electronically stored texts, just as oral statements, can be changed without leaving any trace. The codification and decodification by the processor can give rise to mistakes that are not immediately recognisable. The image on the monitor or the printed page varies according to the software used, or the version of it. Intentional alterations cannot be completely excluded during the course of copying or the regular transfer of material made necessary by the obsolescence of hardware and software. They occur without trace and can only be discovered when versions are compared. Digital material is dependent on numerous technical protocols produced by the operating system and the software used, which are not accessible and which change spontaneously during use. Authenticity is not attainable in electronic, digital representations. Thus in electronic word processing a function of writing on paper is lost that was previously of chance importance and was an automatic byproduct that could be made use of when necessary.
The discussion about intrinsic value also draws attention to the need for contextual features or means for preserving their testimony by stabilising their condition. This cannot be achieved by digital means. The stabilisation of the state of a digital text at a particular point in time, necessary, for example, to reconstruct the grounds on which a decision has been made, can only be attained by transferring it to an analogue representation on microfilm or paper. If the additional technical information is transferred at the same time, the original electronic functionality can be restored when it is redigitalised.
Not until an electronic text has been represented in analogue form on paper or microforms is the content matter of the text stabilised. The stabilisation in analogue form reestablishes the function of script. The electronically formulated and communicated subject matter becomes precise, unequivocal and memorable. In this way memorability is externalised. The facility a script gives in the solving of problems by splitting them into different spacial and chronological aspects is thus also available again.
The functional chain of the different forms of linguistic representation is of great importance for those libraries and archives, whose duty it is to safeguard society's ability to recall its past. Whereas libraries are more concerned with keeping up knowledge and making it available, archives guarantee the availability of earlier experience. Both fields, however, employ the stabilisation and concrete form of the content of library and archive material in order to make it available.
For this reason, it is mainly written and printed works, such as files, written documents, books, journals, maps and plans which are considered for the working out of criteria for the identification of intrinsic value. In the case of these, the loss of testimony is endangered, not only through the physical degeneration of the items, but also through the unconscious destruction of evidence as to the context and circumstances of their origin, which can occur during their conversion and must therefore be prevented by a previous analysis of their intrinsic value.
The external features of written historical material or description are of basic importance for both fields. Isolated information exists neither in libraries nor in archives. The subject matter requires the material form of paper and ink, of film and coating, for its content to be understood, as it is dependent on the concurrent interpretation of the formal features. They present the textual content with its affiliations, clarify intentions and make the contexts accessible. Only within this framework is the subject matter comprehensible.
Because of endogenous paper deterioration, insufficient storage space and increased usage, conservation is becoming of first importance to libraries and archives. The continually increasing recourse to archive and library material gives rise to an almost insoluble conflict with the need to safeguard long-term accessibility.
The archival processing of administrative documents should make them accessible for use. Archival appraisal identifies those items which need to be destroyed so that the remainder can be retained and made accessible as authentic historical material. Archive material should not suffer any alteration, reordering of sheets nor dilapidation in the course of their description, their conservation or their use, because this would prevent a future unequivocal interpretation. Archival processing also calls particularly for material stabilisation at the time of description in order to safeguard the testimony of the formal features of its transmission.
Analog image conversion stabilises the testimony of archive and library material. It maintains the utility of archives and the preservation of printed matter, even when the originals have been altered or destroyed through endogenous or external circumstances. It can also reconcile the contradiction between the demands of preservation and availability if it represents the evidence of the originals in unaltered form. Microfilm is a possibility of preservation by conversion to durable materials and capable of repeated reproduction, even when the original has perhaps been lost.
An image conversion in analogue form, such as film or paper copy, cannot, however, reproduce certain kinds of evidence given by the external formal features. An exact definition of the prerequisites which exclude conversion and instead demand the preservation of the original is necessary for the consequential use of conversion procedures. This is achieved by establishing the intrinsic value as a criterion for the non-convertability of the evidential value of library and archive material. The same requirements as for analogue pictorial conversion are set for digital pictorial and textual conversion. However, because this method cannot preserve the evidential value and, in contrast to film, renders the reproduced picture or transferred text flexible, it does not come into question as a means of preservation.
Digital forms of reproduction, however, are often more user-friendly because of their electronic research functions, which are made possible either by means of an additional index in the case of pictorial conversion, or text research for text recognition, and textual conversion. The loss of evidential value and permanent accessibility inherent in digital forms and textual conversion exclude them as a preservation medium. They can only be employed in addition to preservation on film in order to increase the ease of use.
As soon as an oral, written or electronic message formulates contents in order to convey them, it obtains through its expression in a specific medium additional features that are inherent to the medium used. The intention of the message is thus supplemented by the necessarily linked motive in the selection of the medium used and the form of the communication. The text of a foreign minstry memorandum, for example, takes on a different meaning when it is printed in a newspaper. And doctoral theses, at least when they are published in the original version presented for the examination, do not need the endorsement of their scholarly value by appearing with a well-known publisher, as the examiners' report has already established this. When the reputation of the author is enhanced by the reputation of the publisher this is an additional effect that in some cases, however, is to the benefit of the examiner rather than the author. A text published by an author on the internet has a provisional, ephemeral, quality lent by the medium and therefore urgently require the last alteration date.
Once the intrinsic value has been defined it is possible to decide about a conversion for the protection of the original. A separation between preservation and accessibility becomes possible by providing for use authentic reproductions and representations that convey as extensively as possible the evidence of the original. In exceptional cases and on special application the originals can be consulted.
The definition of intrinsic value, together with directions for its application for particular tasks within the framework of preserving collections, is, above all, an instrument for selecting procedures that combat the effects of paper deterioration or contribute to damage repair by converting endangered documents to durable materials. Apart from this, the definition is useful in resolving the conflict between the demands of accessibility and preservation in other fields of archive and library work.
Intrinsic value is based on certain testimonial qualities of library and archive material which support or clarify the content through external formal features, often dependent on the history of the item. External formal features have a non-verbal, visual testimonial quality. The testimonial quality associated with these features often decides on their inclusion in an archive or their purchase by a library before preservation methods are considered. The insight given by these features augments the contents of the text and can for some items even be of major importance for their purchase or appropriation. When certain volumes are chosen by an archive from a number of similar parallel files to be kept for the reconstruction of administrative processes, the evidence they give lies mainly in formal external features and not in the text itself. The appearance of the external formal features of archive material gives testimony about the circumstances which produced them. The identification of the evidential and informational value of administrative records that leads to the decision as to which parts should be retained as archive material is a qualitative appraisal. It determines as to whether testimonial quality exists. It does not discriminate between the different forms of testimony. Archival appraisal and purchase decisions made by libraries take place before the analysis of the intrinsic value. They presuppose that library and archive material have per definition external formal features that first make possible their reading and interpretation, and that these features are coupled to testimony worth preserving. When analysing intrinsic value it is necessary to investigate the form and manner of the physical evidence, that is, its testimonial quality. On the basis of the external formal features the form and kind of the testimony can be determined and a decision taken as to measures for its preservation. This is the establishment of intrinsic value. It investigates the quality of the testimony and sets down in which form it can be realised. Very many testimonial qualities are preserved after the conversion to other materials or other forms. Their transferability is the criterion for employment of different methods to retain the quality of the testimony.
Intrinsic value has up to now played a minor role in archival science and library literature. The discussion about its definition and identification has been enhanced by recent statements of the problem and strategic preservation approaches.
Measures taken for preservation of library and archive material, the conservation of the originals or their conversion, run into the danger that external formal features of files and printed works can be overlooked or falsely evaluated when their importance for archive or library consultation, their accessibility for reading and interpretation, is not known. The ordering of pages in files of loose sheets, the bindings, the pencil notes of one author in the book of another, which was in his library; all these features are clues which give testimony to those who can interpret them about physical context and causality. This testimony is non-verbal and depends solely on appearance. This makes it direct and authentic but also dependent on the understanding of the person looking at it.
Paper deterioration, which has threatened industrially produced paper since the middle of the nineteenth century, makes conservation measures necessary in dimensions far beyond the usual spectrum of conservation and restoration measures. Decisions of great import regarding the economic employment of present possibilities have to be made.
The best and most economic method for avoiding the danger of deterioration is to forego the use of paper containing acid or lignin in the production of books and written documents. Preservation of the original is also technically possible for paper threatened by endogenous deterioration. The measures developed for doing this comprise the deacidification of the paper in conjunction with stabilising measures including paper splitting.
Deacidification prolongs the durability of paper by means of buffers and makes sense when paper is still sufficiently strong. Mechanical strengthening procedures such as paper splitting also help to stabilise paper that is already damaged. Mechanical procedures are available for both cases.
A large number of threatened items cannot be preserved in their original form for economic reasons. Endangered information has for a long time therefore been transferred economically and effectively to stable microfilm. Quite apart from the development in costs of mass deacidification programmes, transfer of documents and media to film is suitable for material of no intrinsic value since it also saves storage space.
Serious testimonial loss can only be avoided by conversion to durable information media in image or textual form. It must be established for which testimonial features of the item in question the potential loss of information during conversion lies beyond what can be justified professionally. Conversion does not come into question for such items. They must be preserved in their original form regardless of cost. Exact identification of these items in archive and library collections is necessary.
This can be done by establishing which items retain their testimonial value after an image conversion with perhaps an additional textual conversion. Closer investigation shows that a large number of items held by archives and libraries can be converted without information loss. To identify these, certain rules must be followed which can be worked out with the aid of the criteria for intrinsic value.
The definition of intrinsic value is based in the first place on the presence of such features whose appearance offers clues as to particular intentions or circumstances concerning their origin or transmission. This evidential value conveys information about an item to the user or reader. Both the content and the formal features of the written form offer new insights.
The safest way to retain authentic testimony is to preserve the original. However, when the originals deteriorate for endogenous reasons new ways must be looked into. Consultation in reading rooms or offices, which wears the item and contributes to its continued degeneration, also stands in direct conflict to its conservation. However, items are preserved in order to be used so that this conflict must be resolved. The development of conversion forms offers a solution for utilisation. The conversion is made on to copiable materials from which cheap duplicates can be made. These can be made available for use so that the film negatives or master copies do not suffer wear and tear. Current conversion methods are filming, imaging and digital text recognition.
Microfilm reproduces a document with an image in analogue form. The pictures are taken on film rolls or sheets (microfiche). Conversion thus takes place through image representation on to another medium. This image representation can be duplicated as often as desired and so that copies of the original can be made available for use in unaltered form (Weber, 1992).
Roll microfilm is a typical sequential imaging means, while microfiches facilitate the deliberate selection of individual pictures. Microfiches can be produced by optical rearrangement of microfilm rolls.
Roll film can be equipped with so-called blips for research purposes, which enable a deliberate selection of individual pictures. Their use is closely associated with an additional description and retrieval, which makes electronic research possible.
Originals can be digitalised in image form and electronically stored and processed. Imaging is also the first step in converting a text to a visual form, in which the digital image is compared with the characteristics of the script and, where they corresspond, transformed into the appropriate code. The text can then be electronically processed.
Because of the quality differences between analogue and digital images as well as the uneconomic costs of the required digital quality the final report, "Digitalisation of endangered library and archive material" presented by the workgroup "Digitalisation" of the DFG subcommittee on collection preservation recommends the scanning of microfilms (Dörr and Weber, 1997). This procedure can be repeated as often as desired without endangering the original.
Digital imaging is not suitable for permanent storage. For working purposes, however, it is very user-friendly because electronically stored pictures can be linked to features which enable research into them. The main advantage lies in the fact that digital images can be made available anywhere via electronic nets. Particular requirements must be met for representation on a monitor, which have also been listed in the previously mentioned final report of the "Digitalisation" workgroup.
Digitalisation, where scanned-in pictures are converted to text by means of OCR software, creates a text which can be electronically processed. The various procedures for fully-automatic or trainable text recognition and their differing strongpoints are also described in the above-mentioned DFG report. The conversion produces processable texts, without reproducing all contextual factors. It produces an electronic text with the advantage of full text research and the disadvantage of stability loss. This makes it more user-friendly, but unsuitable for sole and authentic storage. In order to test the authenticity of the text a back-up analogue representation, such as the representation on microforms, is necessary.
Digital publications, offline on CD-ROM or online, are, at least since general use of the internet, usable independent of their ownership. Whereas in the case of handwritten or printed material their availability for use is dependent on their physical existence, this does not apply to originally, online electronically available texts.
Libraries used to be the main distributors of information as they made books and printed works available. This function of libraries gives rise to the task of preservation in order to ensure permanent availability. However, the preservation of books does not alone ensure their informational quality. Knowledge can become obsolete or can be reworked and incorporated in other scholarly works. Both of these can reduce the informational quality of printed works to the characteristics of book production, and, as this testimony may be available elsewhere, make the physical preservation of individual copies of printed works superfluous. The purpose is the availability of the information. However, the logistic treatment of material objects such as books is normally the prerequisite for the availability of their evidential value. This is not so in the case of online accessible original digital texts.
Originally digital texts have different characteristics than handwritten or printed texts. The internet signals provisionality. It therefore creates the need for the publication itself to be externally authorised. This occurs either through the quasi-physical connection to an authorising homepage of areas of the server accessible only to the homepage managers or, as in the case of a doctoral thesis, through the examination of the thesis itself. Both of these are a form of reliability audit and replace, as already in the case of doctoral theses, the scrutiny by a publisher's reader. This does not apply where texts are made electronically available to closed groups, or for internet publications for which the author alone is responsible.
The accessibility of a publication is dependent on the organiser of the homepage or on the author. This means that its preservation is no longer the prerequisite for its use. An external distributor of the information is no longer necessary. However, the contents only remain permanently available if their preservation is deliberately ensured. The task of libraries is different for electronic publications than for printed matter. Preservation becomes the priority. This can only take place through conversion to a stable medium. Only when libraries want to preserve electronic publications, for example by a print-out, can they ensure their permanent accessibility even when they are no longer available on their original server. For this the additional features for the reliability audit must also be converted. This, however, raises the question of the rights, and perhaps duties, of the author and the operator of the server. Does copyright also include the right to destroy the contents?
The stabilisation and permanent availability of originally digital publications can only be achieved through long-term programmatic, but very expensive migration or by conversion to analogue representation, be it paper or microfilm. Electronic functionalities which cannot be represented in an analogue fashion must be represented in such a way that redigitalisation restores the same functionality as in the original publication. The HTML codification common to the internet could offer the possibility of reconvertability of electronic functionality after analogue storage.
The conversion methods which can be used for the preservation of testimony of archive and library material fall into the categories of image and textual conversion.
Analog image conversion has the advantage of stabilisation and authentification. The representation is more stable than the original. Conservation authorities and researchers into art history make use of earlier pictures when reconstructing the original state of buildings and artworks. The traces of usage can be very precisely made out on filmed archive material converted at an early stage. Even were it at some time to deteriorate through use, the original condition can be recognised. Image conversion can reproduce certain testimony of the prototype. However, when it is digitalised it loses its stabilising function. Digital image conversion should therefore be undertaken from microfilm. This makes it permanently available for double-checking.
Pictorial analogue conversion guarantees the stabilisation of the original, and is therefore suitable when the intrinsic value is considered insufficient to warrant its inclusion in a library or archive.
Textual conversion is open to greater danger of alteration. When a text is copied errors can be made or the text deliberately falsified. On editing a text not all external features, such as edletions and additions, can be mentioned and notes explaining the state of the original may perhaps be misunderstood by the reader. Moreover, digitalisation of a text renders it alterable without trace and can, solely through its technical processing, result in discrepancies between the original and its reproduction.
Textual conversion is not suitable as the sole replacement for the original. It can make the processing of texts easier, but here also requires a pictorial analogue representation as a permanent validation.
The concept of intrinsic value, as it is used here, was much influenced by the ideas of professional American archivists. It was employed above all in the National Archive in Washington for the preservation of the collection and the development of strategies to do this.
Intrinsic value is given when the testimony of the original relevant for its consultation is not completely preserved on conversion to a different medium. Intrinsic value is based on features whose testimony is dependent on the form of the original and can therefore not be converted. As prior archival appraisal or purchase by a library have established this testimony as being worth preserving and relevant for consultation, the possible loss of testimony on conversion requires the preservation of the original. The intrinsic value is therefore the characteristic that excludes a conversion of archive and library material. The establishing of the lack of intrinsic value means a decision must be made as to what kind of conversion is suitable.
Criteria for the establishment of intrinsic value can help in selecting that part of a library or archive collection that is suitable for certain conversion measures. They should give guidance for practical use and the filling out of approraiate forms should make it possible to apply them quickly and efficiently.
Duplication, the production of copies or transcripts, or the printing of texts have always been a way of making them more readily available and better ensuring their preservation. Even where only a few copies survived the dangers of time and history, the purpose of the production and preservation of the original has been achieved. It still serves for information or for symbolic use, as when oaths are sworn by laying a hand on a bible.
Up to now libraries have profited from this side effect in the improvement of availability through duplicate copies. Printed books which appeared in an edition of several hundred had a relatively high chance of surviving in sufficient numbers. Even when no particular measures were taken the size of the edition has always been a relatively good guarantee for published material. Unica and collections of handwritten documents have always enjoyed particular protection.
The phenomenon of paper deterioration endangers the entire production of a book which has been printed on the same paper, and therefore applies to all the copies of an edition, also depending on the external factors of storage and use. Relying on the large number of copies is only of limited help in such cases. Regardless of the size of the edition, not a single one of the printed items has a chance of survival in its original state.
In comparison to an author's manuscript, the physical production of a book brings additional information through its own characteristics and features. The publisher's name lends it a certain authority and establishes a connection between the contents and other productions of the same publisher. The place and date of publication, the ISBN, the form of binding, the number of pages, all give additional information which can strengthen or weaken the content of the text. The book produced by a publisher is therefore the result of the text as produced by the author and the publisher's production of the actual book. This in its totality has a testimonial quality made up of the contents and the outer form of the book. Form and content complement each other. Both together are only available where at least one copy of the edition survives.
Uniqueness or unicum character have up to now increased the danger. However, the threat of paper deterioration is the same for large editions of books as for one single copy. Libraries therefore often find themselves in the same situation as archives, which are concerned with unique documents and sources of information. Libraries are being increasingly confronted with this problem. Paper deterioration, which eliminates the automatic safeguard of a large edition, brings libraries and archives into the same situation of having to concern themselves with the preservation of the actual testimonial quality or quasi-unique features.
Even more than with books, archive collections are concerned with the different aspects of testimonial possibilites, with different forms of reading and research into them. Archive material is unique in more than one way. It has a unique evidential value as to its genesis. It originates from a unique, usually cooperative process of decision-making and the physical composition of the documents in the individual files is unique. All these unique features tell us something. The page order demonstrates the chronological order of the working stages and can even elucidate a connection between question and answer. The colour and labelling of the file cover, insofar as these have been preserved for contextual reasons, can offer clues as to the working process. The colour of notes made on individual pages show who intervened. Even the order of the files in a register can offer clues as to internal connections, which would no longer be recognisable were this feature to be lost. And copies of the same manuscript become unique specimens when they have been made use of in different working processes or when they have influenced various decisions and so left varying traces of the work process. As opposed to manuscripts, files are not created to be duplicated; their purpose has been achieved when they are complete, when they have been closed. They are not, like a manuscript, prepared by a publisher, printed and sold. Thus they do not profit from the side-effects connected with distribution, that is their safeguarding and preservation through multiplication.
In the archival processing of administrative documents for evaluation uniqueness is the deliberate aim. The utilisation of files as archive material requires the realization of a new purpose that can only be undertaken with completed, closed files. This occurs through appraisal, the aim of which is to identify uniqueness. Only uniqueness ensures authenticity. A number of items with the same information and external features can, in the course of future use, develop in different directions. Forgeries, impairment of readability and later additions can be clearly identified when the uniqueness and therefore the authenticity of the original is assured. Duplicate copies, contributory files, unprocessed copies and forms are destroyed. Uniqueness of action and uniqueness of information are the two criteria for archival appraisal. It condenses the documents selected by the administration to a concentrated form and thereby creates archive material freed from repetition and ballast. The safeguarding effect of redundancy is, other than in libraries, not linked to identification and description, that is, to the genesis of archive material. It is one of the measures taken for preservation and storage in archives and must be an additional and deliberate aim. After establishing whether an item is worth keeping in an archive and thus becoming part of the collection, measures for its preservation are taken. Producing of redundancy is one of several possible measures for preservation, after authenticity is established and stabilized.
The utilisation of image and textual conversion for preservation is therefore traditionally of great importance to archives. It is used deliberately and is not an automatic waste product of working copies. The danger that information is lost or changed by duplication is always present and must be weighed against the preservation effect.
The prerequisite for the determination of intrinsic value is the unequivocal identification and definition of the item under examination. While this determination is unequivocal for monographies and journals acquired by libraries as well as in the taking over of private papers, in the case of archival acquisition of official redords it requires archival appraisal and description. This refers to clearly defined archive and library material that can be unequivocally identified. The determination of the intrinsic value as criterion for the use of preservation measures cannot replace archival appraisal. It can only follow it. Archival appraisal and library purchases decide what is to be preserved. The determination of intrinsic value decides on how it can be preserved.
The analysis of the intrinsic value decides on the testimonial quality of the documents, which themselves, on the other hand, have to fulfill the criterion of uniqueness. The intrinsic value determines to what extent the external formal features of an item are informative for the reader or researcher. This concerns the purpose of the utilisation of library and archive material, and that is to gain information.
Two categories of measures to be taken come into question, which currently give rise to differing costs:
The decision as to what measures should be taken must be made primarily according to professional criteria, taking the costs into account. The yardstick of intrinsic value is the basis for discriminating between different prerequisites for both categories of preservation measures, as it is oriented on the testimonial quality of the item concerned and is thus definable for the work of archives and libraries.
The use of conversion instead of preservation of the original depends on the kind of features of testimonial quality. An item can have one or more external formal features of testimonial quality. What counts above all in analysing intrinsic value is to what extent the testimonial quality of these features can be reproduced by image and textual conversion.
Roll microfilm, when properly produced, automatically preserves the already existing order. It needs to be read picture by picture. Microfiche film is particularly suitable for a directed selection of individual pictures. As opposed to three-dimensional objects, the information which can be deducted from the space taken up and ascertained by specific actions, such as turning over pages, is no longer accessible in the case of imaging. The ease of use changes.
Whether an item can be reproduced depends not only on the item itself but, above all, on the those features of testimonial quality which make up the intrinsic value. Thus the reproducability of certain features can lead to their being filmed if these can be preserved in this way though other features are lost.
The choice of the method is linked to the costs. Economic considerations force the choice of method. Professional criteria, the kind of features and their testimonial quality should above all influence the decision.
There are several groups of intrinsic testimonial quality on the basis of determinable features:
These testimonial characteristics can be divided into those that can be reproduced by image conversion and those that cannot. Clear quality standards must be maintained. They include the order of conversion as the documents are read, also in the case of written items with many cross-references. They require continuous numbering and the reproduction of a scale. The quality of the reproduction must take account of the intensity of the original and requires an acceptable format. Uniform standards are necessary. For this reason the quality criteria should be based on those of the DFG.
Testimonial characteristics can be divided into the following groups according to their reproducibility:
The decision whether to preserve the original or to convert it must take into account the kind of features which influence the convertibility of the testimonial characteristic to be preserved. The decision can be based on the following table.
|1. Imaging with particular procedures||
|2. Imaging with standard microfilm and microfiches||
|3. extra textual||