pointed out that the “search for a tight definition or classification of species or subspecies of form is unhelpful. There is no one essential category of ‘castle.’ It was a fluid idea in the medieval mind.”75 This fluidity and uncertainty is not always something we’re comfortable with when we encounter it in medieval history, yet it is present throughout the sources: “it is not only medieval notions of space that are characterized by imprecision and approximation. As we shall see, time was measured in an even more capricious way. In general, with regard to quantitative terms—measures of weight, capacity, numbers of people, dates, etc.—arbitrariness and imprecision were the norm.”76 When exploring chestnut production in Italy, Paolo Squatriti noted that records of the practice show “the inappropriateness of applying to it rigid classical agronomical and legal categories like ‘ager,’ ‘saltus,’ and ‘silva,’ or even the more modern historiographical distinction between ‘incolto’ and ‘coltivato.’”77
For a few of the forest words, flexibility of use dates to the classical period. Silva was a classical term that carried over to the Middle Ages, and could be applied broadly to many types of natural and managed woodlands. It was the “most generally-used word for woods,”78 and therefore more specialized vocabulary also developed during the classical and medieval periods that incorporated and adapted the word silva. These more exact woodland terms included: silva caedua, coppice woods; silva glandifera, oak-acorn forest; and silva pascua, woodland pasture. During the classical period, silva also had mixed cultural meanings. Sometimes it represented chaos, or disordered matter, and at other times it was associated with pastoralism, highlighting one of the chief features of woodland and forest vocabulary: a literary, legal, and cultural flexibility of meaning, use, and application.79
Despite its flexibility, silva was indisputably associated with features of the natural world. This is not the case with saltus, a classical term associated with both geographical and legal ideas. Heinrich Rubner defined the Roman saltus as land that (though often wooded) was characterized by its legal status as a “res nullius” or no-man’s land on which civilians could hunt freely.80 Josef Semmler echoed Rubner’s claim for a jurisdictional definition of saltus: “The late antique saltus excluded no one: everyone was allowed to hunt large and small game, and to catch birds and fish.”81 And Chris Wickham defined saltus as a land that “could be hunted over by anyone.”82 Such definitions evoke Georges Duby’s optimistic statement that “the forest in the early Middle Ages had been a bottomless reservoir open to all, in which every man could plunge according to his needs.”83
Yet saltus was also associated with woods, and with the word silva. By the early Middle Ages, Isidore of Seville (whose works were found in Stavelot’s library) defined saltus as “vast, wooded (silvestria) areas, where trees leap up to the heights.”84 Despite Isidore’s eloquent topographic definition, saltus has continued to be studied primarily as it relates to legal rights and limits during the classical and Late Antique periods, and how those might have been transferred over to the Middle Ages.
Fluidity of vocabulary can be seen when the many different descriptions of the Ardennes are compared. The Ardennes were not consistently defined or labeled in Stavelot’s charters (a body of sources where some precision of terms might be expected). This is not due simply to changing chancery practice: Sigibert’s chancery issued three charters to the monasteries, and in these the Ardennes are labeled as forestis, saltus, and silva.85 Richard Keyser has pointed out a similar “triple synonymy” for the Orthe, which appears “as alternatively the silva, nemus, or foresta of Orthe.”86 The earliest of Sigibert’s Ardennes charters, from 644, makes a direct claim to royal ownership through the phrase “terra nostra silva Ardenense.”87 In 648 the Ardennes are both a “foreste nostra nuncupate Arduinna,” and a saltus. Finally, a charter from 652/3 labels the Ardennes as neither silva nor forestis, instead using the phrase the “great solitude of the Ardennes” (vasta heremi Ardenensis).88 In the twenty subsequent references to the Ardennes that are found in Stavelot-Malmedy’s pre-1158 charters, they are labeled as a silva, a saltus, a fundus, a pagus, a comitatus, and a fiscus. In the broader record, René Noël noted that before the year 1000, the Ardennes was described as a “forêt” (he does not distinguish between silva and forestis—perhaps like some medieval authors) either twenty-one or twenty-two times, and as a saltus five times.89
Most famous of these is Sigibert’s 648 charter, which famously notes that Stavelot-Malmedy was established in foreste nostra nuncupante Arduinna, or “in our forest that is called the Ardennes.” The charter also describes the Ardennes as a deserted place “in which a host of wild animals sprouts forth,” which has been connected to the royal interest in hunting.90 But wild animals were not only of royal interest—they were just as much a part of the monastic imagining of the forest. This reminder of multiple authorship of charter evidence further warns against an attempt to view charter vocabulary only as applied, state-sponsored language. Language not only could shift over time and in different contexts, but could also be used and appropriated by different groups, sometimes in spite of the intentions of the original author or speaker.
The most problematic of medieval forest and woodland terms is forestis.91 The importance that has been placed on attempting to define forestis is based in part on the fact that the term appears to have been created during the early Middle Ages. Because it occurs generally (but not exclusively) in legal sources, it is often interpreted as a medieval legal invention. The earliest provable use of the term is from Sigibert’s 648 charter. Sönke Lorenz recently argued for a rehabilitation of two earlier charters (once suspected to be forgeries) that could contain earlier uses of the word forestis.92 The related word forestarius also made its first recorded appearance in a charter issued to Stavelot. In 670, Childeric II gave a part of the royal forestis to the monks, and the forestarii were forbidden to violate monastic rights within that zone.93
Forestis circulated quickly through the Merovingian chanceries, and during the first centuries of its use, the word appears mainly, though not exclusively, in royal charters and diplomas. As Lorenz points out, at least one Merovingian vita contains the word forestis.94 Yet in some ways this is still a “charter” usage, since the vita records a donation to the monastery by Dagobert’s widow. The author likely had access to a royal charter, the language of which he adopted in his own work (just as the Carolingian author of the vita prima transferred the word forestis from Sigibert’s charter into his work). Such transmissions of language and vocabulary between source genres highlights the fact that even though specific words might have had a clear meaning to one group or in a certain context, their use and appropriation by others during the Middle Ages could blur legal or technical meanings.
General consensus among modern scholars is that forestis “is not a natural fact” and is instead a term of ownership (particularly royal) connected to hunting and game rights.95 F. Vera adds that “there is some agreement that the concept of ‘forestis’ applies to the wilderness in general and to trees, forest, shrubs, wild animals, water and fish in particular,” and that all of this belonged to the king.96 Heinrich Fichtenau concluded that this was “an institution, stretched not only over woodlands, but also over wasteland and rivers, whose hunting zones and fisheries were opened up only to the lord for pasturage, etc.”97 This emphasis on hunting rights has also led to an attempt to disassociate the forestis from the trees; Richard Hayman even wrote that the term “defined a place of deer rather than a place of trees.”98 Chris Wickham offered a definition of forestis that best expresses this consensus: “‘forest’ is not a woodland, or not necessarily.” “The history of the term,” he wrote, linking forest with royal rather than monastic identity, “is, in fact, nothing other than the history of the development of exclusive hunting reserves
