[click a question below to find an answer, then click the heading of an answer to return here]

What can I do on the first page of the database?

  • You can go back to PBW home, to look up a reference, or maybe take another look at this FAQ
  • You can go to a chronological table, look up a year, choose a narrative unit and explore it
  • You can explore the lead seal records, choosing one of two lists, collections or bibliography
  • You can search all of PBW (but better use a Firefox or Google Chrome browser)


  • You can look up all the people in the database whose first names start with the same letter (say, “T”)
  • By clicking “Choose factoid” you can restrict your work to one factoid (i.e. one kind of information), say “second names”. In this and the next two cases you must then remember to click “Confirm choice”
  • You can similarly restrict your work to one or more source texts, say “Anna Komnene”
  • You can restrict work to a factoid and one or more source text

What is a factoid, and what kinds are used here?

  • Factoids are small packets of information, a word or a sentence, sometimes with a word or two in the original language, and a source-reference. This is the way that PBW stores its information; when you ask a question, the program collects factoids to reply
  • You can see the kinds of factoids used by pressing the “choose factoid type” button; you can then choose to work only with one factoid type (don’t forget “Confirm choice”)
  • You might try out some examples so as to understand what the labels mean

  • When you find somebody’s person card, it will show statistics of that person’s factoids, so that you can decide what sort of information to look for. E.g.:

How do you find what PBW has to say about the following: Eumathios Philokales, one of Alexios I’s generals? Andronikos, often blamed for the disaster of Mantzikert in 1071? Theodoros, hypatos of the philosophers around 1200?

  • Eumathios Philokales is an easy case, as both his names are quite rare. Click on “E” on the first page and then “Eumathios”, and you will find there are only a handful of them, so you can easily choose the right person card (Eumathios 102). Or you could click on “Choose factoid”, “Second name”, “Confirm choice”, “P”, then “Philokales”, approaching the same person card by another way
  • For Andronikos, you will obviously depend on a connection to Mantzikert. However there are no persons called “Andronikos” marked under the factoid location “Mantzikert”. There is another way, as you have the date 1071. Go to the chronological table for that year and find “Mantzikert” or “Andronikos”. Since chronological tables tend to be very full pages, perhaps you might use your browser’s search mechanism. Open the narrative unit (as shown in the next question) and you will discover that you need the person Andronikos 61
  • Theodoros is too common a name to be used as the only way to find the third person. But you can choose the Dignities/offices factoid to find “hypatos of the philosophers”, and that reduces the possibilities to a manageable number, ending in Theodoros 109 (Theodoros Smyrnaios)
  • Methods, like these, to reduce the number of persons to search by inspection are essential when using PBW. Once you have found a difficult person, keep a record of the database code (name + number) or the Permalink URL, so that you do not have to search again

What is the meaning of the symbols at the beginning and end of this line?

  • The cross-in-square symbol at the beginning of the line is more widely used than the page symbol at the end. It opens the next layer of a hierarchy. In the first line of the example shown, it has just opened a list of narrative factoids to show the second line. However the special case in the second line, preceded by the cross-in-square and followed by the page, indicates that this is the title of a narrative unit, opening other possibilities (for the three blobs that follow the cross-in-square, see the question on dates below)
  • Narrative unit titles appear under “Narrative” on person cards, and in lists which appear if you choose narrative factoids or chronological tables. Clicking on the two symbols expands the unit in different ways. The cross-in-square just produces a summary of the unit, while the page opens up the full unit, showing the summary, then all the persons attached to the unit and their narrative factoids. One of the most useful ways of navigating round the site is to employ such lists to investigate other actors in the same narrative by clicking on hypertext links. In the chronological tables, you may look at other events happening at the same time and transfer attention to a person active in one of them
  • PBW expects to be used by people with different needs. Few will be satisfied by the brief information and factoid list of the person card, and most will explore one or more factoid types — for example, to find out about persons’ families or phrases used to describe them. Those who want narrative will find a list of dates and events, which they can expand into longer summaries by clicking on the cross-in-square for headings in which they are interested. If they want the full scholarly apparatus available on a narrative unit, they will click the page. We hope this is a dynamic way of allowing users to navigate round the dossier of material on a person without overwhelming them with unwanted detail
  • Permalinks are references to pages of a database designed to allow permanent linking to other digital projects. Anybody who intends to look regularly at a specific group of persons in PBW can remember that they are named Alexios 103, Andronikos 109 etc., and find them by these names and numbers. However it may be quicker in the long run to copy the permanent URLs of their person pages (or their seal pages) and store them in the favourites list or bookmarks of their browser
  • Permalinks are also a courtesy extended to other on-line publications and sites of many kinds (seal auction houses, genealogical explorations, university course outlines, Wikipedia etc.) which may want to reference the material provided here

Why are some Greek texts, especially from monasteries, full of linguistic mistakes?

  • There are different conventions in Byzantine Studies (as elsewhere) for editing texts classed as “literature” (including, e.g., histories and letters), and others called “documents”. The spelling of literary texts is regularised, that of documents is generally not. A large proportion of Byzantine documents are preserved in monasteries
  • Thus there is serious inconsistency between these two major categories of the sources used for PBW. But it is not appropriate for that reason to impose regular spelling on documents in PBW. The only solution is to leave the inconsistency
  • The legends on seals are edited in several different ways — different conventions for the same material. Special fonts may be employed to represent letters used on the seals, the text may be set in the lines in which it is presented on the lead, abbreviations may be expanded or not, and the spelling regularised or not. This diversity is confusing, especially as PBW’s intended audience is made up of historians rather than sigillographers. PBW has decided to regularise everything into a line of standard Greek

There are a lot of dates in PBW. Are they all certain?

  • • No. The PBW database program has been written to demand a date for everything, whether or not it is datable. And if we only included dates which were certain, most of the material would form a kind of undatable soup, kept somewhere in reserve, ready to pour out when dates appeared. Such material can be included in personal biographies (“For the next few years he occupied various offices, including …”), but it is hard to display it in any meaningful way when seeking to give a context to biographies in a more general historical structure

  • • PBW believes that by constructing a detailed dating framework and putting in the undatable as well as the datable, historians are forced to confront the problems of the sources in a dynamic way. “Are these events connected?” and “Which came first?” are not high-order historical questions, but they do prepare the way effectively for more useful enquiries. Many such preliminary questions have been posed and solved here for the first time by the need to include dates

  • • The key issue is that users should not be fooled into believing that dates are certain when they are not. PBW includes symbols to indicate the degree of certainty of dates, as seen in the truncated example on the left, dating events in the life of the patriarch Michael Keroularios from his appointment to the crisis of 1054. The symbols have the following meanings:
  • ••• 3 dark blobs, 0 light: a certain date at a level within the year. This might be a full date, an ecclesiastical date (Easter Sunday), a month or at least a clear indication of the season
  • •• 2 dark blobs, 1 light: this event is more likely than not to have occurred within the stated year. To put it another way, it will have occurred within a three-year period, and most probably in the middle year of the three, the one which is marked
  • • 1 dark blob, 2 light: this event probably took place within a five-year period including the year given and two years before and after
  • 0 dark blobs, 3 light: this dating is little more than guesswork. Often it is a median date between the earliest and latest likely year. Many letters of Michael Psellos dated within his public life, for example, are assigned the date 1060, the median date between the beginning and end of his career

Why does the searching mechanism not work?

  • The searching page (access via “Search PBW” on the first page) allows three invaluable searches which cannot be done in any other way: searching for words in all fields in factoids apart from those containing original languages or listing the holy figures usually found on seals, and finding words used in headings on the person cards. The other searches can be duplicated by choosing a particular factoid. Only if you want to combine two factoid searches is “Search PBW” essential, e.g. to find those with the ethnic label “Turk” with connections to the location “Nicaea”
  • This part of the site is rather fragile, as it is particularly susceptible to changes made in browser technology. At present (July, 2011) the Firefox and Chrome browsers give the best results, while Safari does not seem to work at all. This situation has changed dramatically in the past and will doubtless change again. If your current browser does not operate the searching, it is worth trying another, as they may all be downloaded free.

Does PBW give useful statistical information about Byzantine society?

  • Unfortunately not. PBW just records what the sources say, and they rarely state the obvious
  • Most of the officials in the palace and patriarchate, for example, were Greek-speaking, Orthodox Christians who called themselves “Romans” (Byzantines). But there is rarely any need to say so. Thus Greek-speakers, Christians and “Romans” tend to appear in PBW round the borders of Byzantium or outside them, where their presence is interesting or surprising
  • There may be limited statistical information on dignities and offices: for example, if an office considered rare appears on many seals. But the source of the information must always be tested: the date of an official may have been suggested because prosopography shows that the office was specially prominent at one period. There is always a danger of circular reasoning.

Is PBW complete?

  • No. It can never be complete, because several hundred new seals are unearthed each year, new texts are edited, old texts are re-edited, and whole episodes are re-examined and re-interpreted
  • Even with regard to information already published, PBW has read only the texts mentioned in the list of sources. This is becoming more complete than before for Greek sources, but the entering of non-Greek sources is still at an early stage
  • PBW must always be examined for what it contains. It should never be assumed that what it does not contain does not exist