What Are Sources?

In historical studies, sources are records which grant us insights into the past. Texts, images, interviews with contemporary witnesses, sound recordings and physical artefacts, among others, can all serve as sources. Crucial for any historian is an understanding of the basic difference between historical sources and secondary literature. In everyday speech, the word “source” is often synonymous with the term “source of information”, an expression which encompasses all potential persons or things from which knowledge can be obtained (conversation partners, books, television programmes, old documents, etc.). In history, however, a distinction is drawn between primary sources and specialist academic literature (also known as secondary literature), which also relies upon sources when investigating the past.

The dividing line between secondary literature and sources is sometimes unclear, especially in relation to more recent history. Whether a text should be classified as secondary literature or as a source depends not only on its quality or age, but above all on the perspective from which we view it. For example, a study on Basel church history will use Rudolf Wackernagel's book Beiträge zur Geschichte des Basler Münsters (Basel 1881-1885) [Contributions to the History of the Basel Cathedral, Basel 1881-1885] primarily as secondary literature. A study on regional historiography, however, will view Wackernagel's work first and foremost as an example of historiographical representations of the late nineteenth century, and thus as a source.

The term “(water) source” evokes an image of naturalness and purity, thereby suggesting that a “return to the sources” (ad fontes) can allow the past to be accessed directly and understood immediately. In reality, however, this is impossible for two reasons. First, each record of the past is produced in very specific circumstances for specific purposes by specific persons. Second, each source comes with its very own history stretching right up to the present during which it would typically have undergone a number of transformations. As such, a source does not simply speak “for itself”; to enable it to do so – i.e. to be able to understand it and accurately interpret it – we must know the context in which it was created as well as its subsequent history as an artefact. Establishing these facts is the function of source criticism.

In principle, all artefacts from the past which have survived into the present can be regarded as a source. In practice, however, a record from the past only becomes a source when it is used to answer a historical question or to glean information about the past. Source selection is a critical step in this process; the sources which you consult will strongly influence the nature of the claims which you can make about the past.

Source Typologies

As the quantity and variety of potential sources is enormous, there have been and continue to be various attempts to impose a sense of order on this mass of information by compiling source typologies. Some of these approaches will be introduced briefly in this section. In doing so, a distinction will be drawn between typologies which base themselves on the material properties of sources and those which relate to source content.

Given the extraordinary variety of written sources, it is neither possible nor worthwhile to list or classify all source typologies here; nevertheless, Hans-Werner Goetz’s attempt to provide an overview of medieval text sources is worth noting (Hans-Werner Goetz: Proseminar Geschichte : Mittelalter, Stuttgart 2014, pp. 109-217). Knowledge of the key types of written sources is useful for source interpretation, a process which involves relating a source’s original aim and purpose – its “bias” – to your own research question.

Material Properties

The historian Paul Kirn (1890-1965) classified sources according to their material characteristics, defining them as texts, objects or facts. For example, a Roman-era battle may have yielded reports about the battle (texts), archaeological artefacts left behind on the battlefield such as coins, weapons and skeletons (objects), and the legacy of a Roman imperial border running to the south of the battle site for centuries thereafter (historical facts). Most abstract is the notion of historical facts, which can also include linguistic features, names (place names, in particular, often testify to the dominant local language at the time of a settlement’s founding), institutions or customs.

Source typologies which draw from the field of media studies reflect the growing significance of audio-visual media by distinguishing between texts, images, audio documents and material objects. Digital sources pose new challenges both for history as a discipline and for archives as institutions due to their sheer mass, the difficulties of ensuring that they remain permanently archived, and their instability (they can be altered at any time in ways which are barely detectable).

Substantive Criteria

Substantive criteria for source typologies are derived from the distinction between vestiges and traditions which was drawn by the German historian Johann Gustav Droysen (1808-1884), a difference later expanded upon by Ernst Bernheim (1850-1942). This distinction assumes that there are a) records of the past which were only created for contemporary purposes and which happened to survive into the present merely by chance (vestiges); and b) records which were consciously produced for the sake of posterity with the aim of conveying a particular image of the contemporary era to subsequent generations (traditions). Sources which can be classified as vestiges, however, still do not provide unfiltered access to the past, because each new source would have been created with particular intentions in mind and/or for a specific purpose. As such, these objectives must still be considered when interpreting vestiges. However, because it is often difficult to draw a strict dividing line between vestiges and traditions (are personal diaries or newspaper articles, for example, written for the sake of posterity or with only the present in mind?), this distinction has come to be regarded as outdated. It does remind us, nevertheless, that for every source which we analyse, we must always carefully consider what its creator sought to convey to future generations and which information is revealed rather more inadvertently. More recent theories also reflect this distinction based on purpose of production, distinguishing, for example, between monuments or messages (i.e. traditions) and documents or traces (i.e. vestiges).

The idea that sources can be categorised according to how they relate to reality was first posited by Gerhard Theuerkauf (1933-2014). He proposed three categories of sources: factual sources (i.e. those which describe what happened), fictional sources (what could have happened), and normative sources (what should have happened). According to Theuerkauf, however, these categories can overlap within the same source.

Source Groups

Besides attempts to compile source typologies, scholars have also sought to divide sources into groups on the basis of other common characteristics identifiable during source analysis. These efforts, which divide sources into genres, do not reflect an attempt to compile a more comprehensive classification system for sources.

Notable here, for example, is the genre of self-testimony (also known as ego documents), i.e. sources which are primarily written to provide information about their author. These sources have received increasing attention from historians in recent years.

Serial sources are related sources which are systematically produced over an extended period of time, for example by bureaucrats or bookkeepers. The individual documents within a source series often follow a similar format.

Source Editions

In archival science, a source edition is an original source or series of related original sources which has or have been published alongside accompanying critical commentary. These comments draw readers’ attention to errors in the original or to inconsistencies between different versions of a source. They also provide background information and identify aspects of a text which may not be apparent to readers. Source editions can cover a specific collection of sources, the collected works and writings of a particular individual, or assorted sources relating to a specific topic (known as a sourcebook).

Source editions, especially sourcebooks, are by their very nature incomplete; they can never incorporate more than a fraction of the available source material. Nevertheless, source editions do come with a number of benefits for researchers. An introduction can shed light on the context in which the original sources were created, while the marginal comments provide explanations of key terms, informative commentary, and references to further interesting sources.

Because a carefully compiled source edition represents a major undertaking, most source editions are produced as part of larger projects which can run for decades or even centuries. Most nation states in the eighteenth and nineteenth centuries promoted their own source edition projects. Alongside those in Switzerland and Germany, these included the “Muratori” project in Italy, France’s “Recueil des Historiens de Gaule et de la France” and the “Rolls Series” in England. Source editions vary in terms of their structure, their scholarly standards and the type of sources which they select. Because key features of source editions have evolved over time, the same large source edition project may have produced publications of varying academic quality. For a lack of alternatives besides much more demanding archival research, historians often continue to work with older source editions which do not conform to present-day academic standards. As a general rule when consulting a source edition, consider its aims and corresponding selection of sources.

In recent years, source editions have also been released in a variety of new digital forms. To be classified as source editions, these digital publications must fulfil the same criteria as source editions which are published in print: above all, they must offer a precise reproduction of an original text with accompanying critical commentary. Ideally, digital source editions should also conform to new digital standards, for example by adhering to the Text Encoding Initiative. Genuinely digital publications enjoy many advantages over printed source editions (see also the section on working with digital sources). Most significantly, digital texts can be subjected to a full-text electronic search. Furthermore, headwords can easily be assigned, while other information (such as names of persons and places) can be inserted into the commentary to promote more precise search results and to establish links with related passages and texts. The digitised annual accounts of the city of Basel (1535-1610) provide a good example of the opportunities and benefits which a digital source edition can offer researchers. Users can search for different financial records using various freely combinable search criteria. Within very little time it is thus possible, for example, to trace the development of one specific tax receipt across the whole reporting period. This example illustrates that the availability of source editions in digital form can significantly reduce the workload of researchers and stimulate new avenues of research. On the one hand, this is a very positive development (which you should certainly take advantage of during your studies), but it also carries with it the danger that other equally interesting source collections which have not been published in (digital) source editions may be overlooked and thus ultimately end up ignored by research.

The two linked examples from source editions – the first a reproduction from a medieval manuscript and the second a copy of a diplomatic note from the records of the Diplomatic Documents of Switzerland (Dodis) project – illustrate how source editions are structured as well as how they should be read.

Calendars

A calendar (also known as a regesta) is a special form of source edition which does not reproduce original texts in full, but briefly summarises their most important points. Typically used to provide a synopsis of legal documents, calendars usually offer a paraphrased version, but occasionally incorporate selected quotations from the original. Calendars also contain information on where and when a source was created as well as details on relevant persons (especially in the case of individuals mentioned in legal documents), the call numbers with which the original sources can be located in an archive, and references to related source editions. Older source edition projects in particular were often published in a hybrid form, combining elements from both conventional source editions and calendars (e.g. the “Eidgenössische Abschiede” or the “Basler Urkundenbücher”).

Digital Copies

Not to be confused with source editions, digital copies are scans or photographs of sources which have been made available online. While these may also spare you the need to visit an archive, they lack the transcriptions and critical commentary of a quality source edition. Image sources are also made available in digital form in some databases. The quality of the so-called metadata (i.e. information about a source, such as the date of its creation or the name of its author, which is not necessarily provided in the source itself) can vary significantly from digital copy to digital copy. Those with more complete metadata come closer to fulfilling the standards of a source edition.

Digitisation projects can generally only capture a fraction of the available sources in archival collections. As such, they digitise a selection of sources chosen according to various criteria, e.g. their content, how well they are preserved, legal considerations and/or available funds. Many archival catalogues, including that of the Basel-Stadt State Archive, also offer access to digitised records in their collections. Catalogues of digitised sources are listed under “Source Databases” in the section on research resources.

Digital copies of printed sources are sometimes also available in a hybrid form as documents which are fully searchable, albeit with some degree of error, using Optical Character Recognition (OCR) software. However, these texts are often not checked for errors, nor are they accompanied by any critical commentary. Although the original source is available as a text (and not just as a mere scan thereof), such texts do not meet the standards of a source edition.

Source Databases

There have long been efforts among historians to compile comprehensive registers of archival holdings. The old auxiliary science of source studies has been rendered obsolete in recent times by the proliferation of source databases, which has made searching for sources significantly easier. The value of a database depends on the accuracy and detail of its headwords and source descriptions; a database does not search through the actual content of its sources, but through the meta-details recorded for each source (i.e. any information on content, author, date of creation, archival location and/or source editions in which it has been published). More and more databases now provide direct access to digitised copies (i.e. scans or photographs) of sources. More specialised databases allow advanced searches by source type (e.g. images, manuscripts or early printed works).

Depending on the aims of your research, it can also be worthwhile to search for smaller, more specialised databases. A good example thereof is the self-testimony database developed at the History Department (see here). Source databases are not to be confused with digital source editions.

In the Literature, Bibliographic Aids and Research Resources section of this toolkit, source editions and source databases are listed separately.