The Universal Lexicon is not the Universal Lexicon, that is the central finding of the following paper that I presented at ISECS2019 in Edinburgh. By analysing metadata of this 68-volume lexicon and taking a stratified random sample of 680 biographical articles, I could show how the lexicon changed considerably during the 23 years of its production period. When using the Universal-Lexicon as a historical source, we cannot simply compare articles from “A” and “Z”. Instead, we must be careful to consider the different contents and contexts of its early, middle and late phase.
Sometimes it is the smallest detail that matters most and that leads us to lasting insights. On the page above nothing seems extraordinary. It is just one of 134.000 pages in some German encyclopedia. Yet when we look at it more closely, we find that there is something just a little off. Somehow the printer skipped 96 pages. It’s just an error in the page count, one amongst many, as I soon found. Yet oddly enough these errors occurred, and occurred only, within the first 20 of its 68 volumes.
As we see in the chart above, soon after, these errors diminished. This led me to consider for the first time what different phases this lexicon might have gone through and how the 20 years of its production considerably impacted its content.
When working on this lexicon, I found over and over again that the same force, time, left its marks everywhere. During its production period of 20 years it did not just move straight forward from the letter A to Z but instead its nature changed so profoundly that volume 1 and volume 64 have little in common. “The Universal-Lexicon is not the Universal-Lexicon” that is my core insight that I will share with you today.
I. The History of the Universal-Lexicon
Johann Heinrich Zedlers “Grosses vollständiges Universal Lexicon aller Wissenschaften und Künste” is in many ways an extraordinary source. Not only is it the largest encyclopedic lexicon of its time but it is also a source consulted by many historians today. Amongst its 289.000 lemmata there seems to be an answer on almost anything. Yet even it is often consulted, it is hardly ever read. Its sheer size renders it almost impossible to study it in its entirety. An estimate from Dorn, Oetjens and Schneider in 2008 assumed that no more than 1.000 of its 289.000 articles have indeed been studied in detail.
From the public dispute that immediately followed on the publication of its first volume in 1731 we know today that the Universal-Lexicon did not have a smooth start. When Johann Heinrich Zedler, a born shoemaker, proclaimed that he would create a universal lexicon that would render 28 different specialized encyclopedias obsolete, the resistance by his competitors followed on the spot. Thomas Fritsch’s heirs and Johann Gottlieb Gledtisch, major publishers in Leipzig, quickly put up resistance against it in what is sometimes called the “Publisher’s fight”.
Publishing lexicons themselves, they knew that a lexicon was not written out of one person’s mind but instead was a compilation of many bits and pieces of other existing literature found on the current book market. They defended their territory with patents, by appealing to the book commission in Leipzig and by starting a public defamation campaign against Zedler and his “Charlatanry of book trade”. His first volume was confiscated and he had to pay a fine for violating Fritsch’s patent. Only after changing the place of printing several times, gaining supporting patents by the king of Prussia and the Emperor and almost bankruptcy in 1737 the Universal Lexicon entered a more stable phase. In 1738 the printing was relocated to Leipzig, the patent of Fritsch had expired and after this tumultuous and well-studied phase followed many years of silent and steady publication.
With the Professor Carl Günther Ludovici as the new editor in chief innovations were introduced: First, living people were included in the lexicon, a difficult and therefore rare endeavour. Second, external contributors were encouraged to send in articles about themselves, their families and places thus opening up the authorship to a larger public, an unusual step at that time.
From 1739 on, with Ludovici as the new editor and the merchant Johann Heinrich Wolff as a new financier, who saved the lexicon from bankruptcy, the production continued smoothly until the last volume “Z” was completed in 1750. Four more volumes followed until Ludovici abandoned the project of supplementing the lexicon further. The lexicon had reached alphabetical completion and public interest on the supplements was low.
This story is well known to anyone who is familiar with the history of the Universal Lexicon. Yet as exciting and as consistent as it may seem, it brushes over a 23 year period and reduces it just to a few events. More importantly, it has not been studied how this chain of events influenced the contents of the lexicon. What impact had the “publishers fight”, the editorship of Ludovici, the financially profitable phase and the working on the supplements on the actual contents? To gain new insights into these questions, I turned towards the most overlooked source for the history of the Universal Lexicon, the lexicon itself.
II. Analysing the Meta Data
Most changes that have been brought to light about the Universal-Lexicon, like the inclusion of living people or the participation of external contributors, were built on prefaces and the advertisements of the publisher. Yet, I argue, looking at article lengths, reference frequencies, errors in page count and lastly by taking random samples of its content, we can take a different perspective. A perspective that enables us to supplement (if not even challenge) the statements of the lexicon’s creators.
To achieve this, I used an online version of the lexicon, Zedler-Lexikon.de, that was created at the university of Leipzig in a DFG funded project from 2004 to 2007. This website allows scholars to easily search the list of 289.000 lemmata and links them right to the scanned pages. They also began a rough and incomplete classification of assigning the articles to different topics.
Building on this valuable pioneer work, I gathered a list of all lemmata and references from the website . This list includes the title, volume, page and length of each article. I used this simple table, built by sending requests via the search function, to look for patterns in order to answer my questions.
Here, we see a statement by Johann Heinrich Zedler from the preface of volume 19. He promises his readers that the articles will become shorter and more precise in the future. However, when we look at the average length of the articles, we see that he is far from delivering on this promise. Quite the contrary is true.
In volume one, created in 1731, an article was about 1/10th of a page long. By 1750 the average length had increased by the factor of ten. Where in the beginning an article only rarely exceeded one page, later, articles became enormous in size (up to 174 pages!). At the moment when Zedler made his promise, their strong increase in length was just about to begin. This increase in length also sheds a new light at the methodological difficulties when comparing articles from A like Aachen (4 pages) to W like Wien, Vienna, (134 pages).A possible reason for this change of mind becomes clear when we look at the amount of volumes produced per letter of the alphabet (discounting the supplements here). While letter A, B and C were covered in 2 volumes each, the letter S and W received 9 volumes each and “Z” received four more volumes again. Keeping in mind that the size of each volume remained fairly stable at about 1000 scanned pages we see how the lexicon is largely skewed towards the ending of the alphabet. As it seems, the publishers were well aware that their subscribers (having already bought dozens of volumes) would under no circumstance abandon the project now. This allowed them to make more money on books that were guaranteed to sell.
Frequency of References
However, apparat from this empty promise, we can also see the earnest endeavour to make the lexicon more accessible for its readers. In the same preface Zedler promises that the reference articles will become more frequent in the future and indeed their number saw a steady increase. Starting at only 2.000 in 1731, references reached their height in 1746 with up to 8.000 references per volume. Thereafter their number declined again as the alphabet became more and more stretched out over the volumes. This also supports the claim made by the editor Ludovici in the preface to the first supplement in 1751 that writing these “Siehe-Artikel” (reference articles) took up almost a fourth of his time as an editor.
Therefore, we can see that looking at some simple metadata can provide us with new insights on the lexicon and allows us to validate the claims made by its creators. Combining these distributions with the known history of the encyclopedia, we can distinguish different stages with blurred yet still important boundaries. The early phase of 1731-1738 is marked by instability due to the legal and financial struggles. The late work from 1744-1750 sees a strong increase in article length and volumes per letter. The supplements from 1751 to 1754 however, contrast the rest of the meta data so strongly, they have to be considered separately.
III. Changing Contents
So far we have only looked at the changing metadata but not said a single word about change on the content level. So how can this lexicon be analysed meaningfully when the changes that it incorporates are so drastic? We know that no single article, not even an entire volume, can stand pars-pro-toto for the entire lexicon. Almost everything that is said about “the” Universal Lexicon only applies to a specific sub-section of it. When studying it, we have to consider this change and adjust our samples accordingly.
For my dissertation I drew a first random sample of 680 biographical articles, 10 articles per volume. By stratifying the random sample like this, I could make sure that each volume was represented evenly despite the strong fluctuation in the number of articles per volumee. By employing this sampling method I later was able to compare the different phases of the lexicon to each other.
My question here was a simple one: Who is represented in the Universal Lexicon? Men or women? Noblemen, clergy or scholars? Antique personas or recently living people? In this last section we will not be able to look at all results in detail, instead I just will demonstrate how change over time also influenced the content.
Biographical Articles by Epoch
I assigned all of the 680 persons to either antique, medieval or early modern times. In total early modern people made up 62% of the sample collected here. However, as we can see this was far from set out in the beginning. For example, in volume one only 1 out of 10 people lived in the early modern period.
We can see this change in detail on the chart above. Each bar represents one volume and each block one of the ten sample articles that I drew randomly. In blue we see antique, in red medieval and in grey early modern people. We can spot a clear tendency here: Interest towards antique personas declined drastically and contemporaries of Zedler and Ludovici began to take up more and more space. Although there is no drastic cut with the editorship of Ludovici, the trend is clear: Interest shifts towards the present.
Biographical Articles by Region
We can see a similar picture when we look at where the different people come from. Following the same logic we see in the chart above in red German speaking territories, in grey other European territories and in blue extra-European territories. We observe a clear shift towards including more German people. Whereas they were hardly covered in the first ten volumes they became increasingly frequent from volume 50 onwards.
Over all I could observe that during the production time of the lexicon the interest was shifting from antique to contemporary, from sacral to secular, from European to German, from regional to local, from noble to scholarly. These shifts are so drastic that the early and late volumes bear little resemblance. Volume one was a lexicon with short and quick definitions for antique personas, whereas volume 64 provided lengthy articles on local scholars side to side with their extensive bibliographies. This shift in content and form likely reflects several factors: The changing intent and influence of the creators, the interest of the audience but also larger shifts of the book market on which it fed.
The Universal-Lexicon is not the Universal-Lexicon, that is the central finding of this paper. Over the 20 years of its production this enormous lexicon did not just experience shifts in article size, frequency of reference and production quality, but also major changes in content occurred. Therefore, when employing this lexicon as a source, it is of crucial importance that it is not treated as one holistic and consistent book but rather as a collection of volumes created by different people, on different sources within different contexts. To take a single section of the Universal-Lexicon pars-pro-toto or to gather articles on one topic without considering from which volumes they come, could be a crucial mistake. Instead, each article has to be placed within its specific content. A consistent opinion of “the Zedler” must be discarded for a multitude of voices coming to the fore within it.
I hope this can also be an inspiration for the wider study of lexica of the 18th century. Here, looking at metadata allowed an additional perspective on the history of the lexicon. Of course, in many instances the Universal Lexicon is extraordinary and exceptional even for its time. Nevertheless, for the study of multi-volume works or lexica that faced many editions and revisions over a longer period of time, the perspective and methods used here can be worthwhile exploring. After all, lexica might today seem like static book-archives, storing the general knowledge of their time. Yet as we have seen today, they can be strongly dynamic sources that are vastly influenced by a multitude of factors. They often provide not the general or consistent opinion that we might be looking for, but instead many contradictions and disagreements that make up the world inside and outside of books.
Section I. and II. of this paper were published in a peer reviewed article. If you have trouble accessing it, just contact me! The article is: Andreas Müller (2019): Vom Konversationslexikon zur Enzyklopädie? Das Zedlersche Universal-Lexicon im Wandel der Zeit. In: Das 18. Jahrhundert 43/1.