Bynames are forerunners to the surname or family name, but they are not hereditary. They are sometimes called ‘cognomina’ or ‘second qualifying names’. Just as the Middle Ages give us the first true indexes, so they give us qualifiers and, as discussed below, glosses. They are extremely common for the later Middle Ages, meaning after the year 1100. They tended to change with each generation and were not necessarily applied to other family members (Postles, 1997: 28).

Although there is variation across Europe, bynames are a durable part of medieval naming culture, as the geographer and data scientist Jakub Marian (n.d.) has shown in a map of the most common surnames in Europe. According to Marian’s research, names deriving from ‘smith’ predominate in England, from ‘red’ in Italy, from ‘miller’ in Ukraine, from ‘priest’ in Romania, and from ‘of bad hearing’ in Finland. These surnames originate in bynames. Though this article focuses on Christian naming practices, bynames were used by Jewish and Muslim communities as well, making them a truly pan-medieval cultural phenomenon.

Bynames first start appearing in the written record around 1050, and the number of people with bynames steadily increases into the early modern period (c.1500), when hereditary surnames begin to predominate (c.1600) (Clark, 1987: 44; Skinner, 1999: 24–5). Medievalists have a couple of theories for why bynames emerge in the eleventh century. Firstly, we cannot discount the fact that the number of written records increases during this same period, though we should be careful about making sweeping statements when dealing with a spotty historical record. Were there more records, or just more that survived? Historians know that the use of written records was increasing, not just doing a better job of surviving down to the twenty-first century, especially for court proceedings, taxation, legislation, inheritance, property transactions, and merchant accounts. As scribes faced the task of keeping records of who owned what and who killed whom, there arose a practical need to differentiate the many Johns who paraded through the courtroom, and so we get bynames like those in Table 1.

The scribes scribbling down these names were often, by the thirteenth century, over-educated, under-employed university graduates who, as members of the clergy, were supposed to be celibate. How do we know these scribes were not exercising a bit of pique or having a go when they named John here Latethewaterga (Let-the-water-go, or wet-himself)?¹ We know that this is not a bit of scribal humor in a few ways, four of which are worth highlighting. Firstly, it would be quite the feat for scribes all across Europe, at around the same time, and across cultural and religious divisions, to be engaging in nearly identical behaviour. Secondly, and relatedly, while the occupational and geographic identifiers might be accessible to the scribe and present themselves as useful to the cause of identifying someone, bynames like Tiny Prick are unlikely to be accessible to the scribe, unless, like poor William who titles this article, John lost his pants while speaking before the court. Thirdly, this was a time of demographic increase. Some historians posit that along with the growth in population and bureaucracy came a new kind of public consciousness and public activity, in which the name served as a social marker. Names were thus not simply a means of identification. They carried important indicators of how a person fitted into their community. Names could imply assets or position; thus the byname could be self-generated by the individual or the family, rather than an identifier imposed from without (Skinner, 1999: 23). Fourthly, we know that another cultural phenomenon was moving through Christian Europe at this time, one that most onomastists point to when explaining the rise of bynames: the name stock was shrinking.

If we look at chronicles and tales from the early Middle Ages (that is, approximately 500 to 1000), we find names like those in Table 2. It is likely that everyone reading this article has met someone or heard of someone named John, William, Maria, or Margaret. At the same time, it is pretty certain that none have met someone named Childebert. There is a reason for this change in name familiarity: name stock contraction. And there is a reason for the contraction, too, though to be honest historians have not entirely explained it yet, as the contraction coincides with increased pluralism and multiple social, linguistic, and religious influences (Bartlett, 1993: 274). Whatever the reason, we know that over the course of the later Middle Ages, the range of forenames (or given names) narrowed (a phenomenon sometimes called homonymity) (Postles, 1991: 30, 32–3; Shagrir, 2007: 50).

While male name stock narrowed earlier, the phenomenon can be seen across both genders. By the early fourteenth century, 40 per cent of the male population of London was named either John or William (Postles, 1991: 46). There was even a large party thrown one evening for over 200 guests, all of whom were named William. In Italy, name contraction led to 38 per cent of women being named Maria (Skinner, 1999: 30). With such a limited pool of names to choose from, it is therefore unsurprising that people turned to bynames to differentiate – something people still do, especially in school, where multiple Michaels and Katherines can cause confusion.

Returning now to our list of Johns from Table 1, we can sort these various bynames into four main categories (Shagrir, 2007: 55–7), as shown in Table 3. All four types can be found in just about every cultural and linguistic community in medieval Europe, though the distribution and grammatical constructions vary (and, of course, layering Latin or Greek on top of the vernaculars complicates the picture, but we are not venturing into those waters here). Moreover, some people had multiple bynames from multiple categories. For example, consider Jean Lefebre dit Mareshal (John Blacksmith called Marshal). ‘Lefebre’ was probably an occupational byname. Jean may also have served as the town marshal, though the use of ‘dit’ may indicate that ‘Mareshal’ was a nickname. Perhaps Jean tended to police his buddies and be a bit of a killjoy.

Speaking more generally, let us turn to a breakdown of bynames from the Latin Kingdom of Jerusalem (Shagrir, 2007: 49):

Over in Kievan Rus′, patronyms, a subcategory of anthroponyms, were favoured, while across Europe, women as a whole were unlikely to have recorded nicknames (Skinner, 1999: 23).

With all of this information about bynames and the complications they present, how should an indexer tackle all this, especially when few readers are likely to learn all the various European languages (including the medieval ones) or specialize in onomastics?

My recommendation as a medievalist, an indexer, and a researcher is to keep it simple. Bynames are not surnames, though ultimately that is what they will become. My own last name, Komornicka, derives from the occupational status byname Komor (the -nicka/-nicki ending is adjectival, meaning it’s a descriptor, a byname), a komor being a small chamber in a country house where dry goods were stored. Likely one of my ancestors was a cellarer and at some point the name stuck, even for those no longer performing that job. Or maybe it’s an ethnic toponym referring to the Komornica culture of the Mesolithic period. I honestly don’t know, though the former is more likely.

Though bynames will become surnames, for the Middle Ages and much of the early modern period it is often impossible to know if one is dealing with a true surname or not. Presentism can be its own form of bias. If a piece of text speaks of Richard Cooper and his wife Wilhelmina, one should avoid the temptation to index them as

It would be more accurate to treat Wilhelmina’s relationship with Richard as a gloss:

Also, it is best not to invert unless both indexer and author are certain that the name in question is a surname. How will one know? Noeline Bridge and Kate Mertes recommend looking up the name in a reputable source, like the Library of Congress or a national biographical dictionary (Bridge, 2012: 5; Mertes, 2012: 56). While these can be of some limited use for very big-name members of the nobility and clergy, your average medieval commoner (and even lower members of the nobility and gentry) will not be in there. So what can you do? If Richard’s name was ‘Richard the Cooper’ would that indicate a byname, while ‘Richard Cooper’ is a surname? Unfortunately, lack of an article or preposition is not determinative. Better to keep it simple. Don’t invert.

A second piece of advice is to be wary of false cognates. Someone named Lepere or Leper is more likely to be a dancer (leaper) than someone with leprosy. A ‘mower’ is not someone out cutting the grass but one who makes faces with their mouth – a jester or mocker. A ‘trechetour’ is not treacherous or a traitor; they are a juggler, adept at sleight of hand. Meanwhile, a ‘wommanespleiere’ is not a lady’s man or womanizer but a female entertainer (Dobrovolska, 2017: 19).

We risk running into trouble if we let our modern assumptions cloud the index. Having a byname did not make someone more important than someone without one. Cher needs no surname for us to identify her. When I was growing up, everyone wanted to ‘be like Mike’,⁴ and no one had to ask, Mike who? The same held true in the Middle Ages. Another assumption that is easy to make as indexers, especially when trying to gloss multiple people with the same or similar names, is to create family relationships that are not there. The godfather John naming the baby John after himself, not after the father John, is an example of how easy it is to become confused. We also risk misrepresenting the nature of the medieval family. Our culture tends to make sharp distinctions between inside and outside the family, distinctions that largely did not apply in the Middle Ages. The term amicus, Latin for ‘friend’, was as readily applied to a kinsman as to a companion. A ‘sib’ or sibling might be your brother, your cousin, or your friend. And a ‘gossib’ was your best friend, the person you chose as godparent to your child, the one who named your child, not a frivolous tale-teller (Niles, 1982: 103–5).

In the end, the message is a cautionary one, but one that has an easy solution. Keep it simple. Don’t invert. Don’t assume you have a surname when working with European names prior to 1600, and don’t assume family members shared names, through either heredity or marriage. Just as the index was a medieval creation to help find one’s way through the text, the gloss was a medieval invention to make sense of the complicated and the obscure. Let the gloss be your friend in indexing medieval names, even if that results in some long constructions. People are complicated. They have many facets. They fall in wells, marry, and travel to neighbouring villages. All of that is part of them, part of their names, and can be part of your index too.

This article is based on a presentation at the June 2023 conference of the Indexing Society of Canada/Société canadienne d’indexation (ISC/SCI).