All complex societies (with a division of labor determined not solely on the basis of sex, with a hierarchical sociopolitical organization, with an economy capable of producing a surplus) leave a potential space for those people referred to, among other terms, as "groups that don't want in." Such a definition is to be understood in both a sociological and an epistemo-logical sense: these groups "don't want in" (1) as far as the hierarchical organization of the society in which they live is concerned, and (2) as far as traditional anthropological categories are concerned. Regarding both these characteristics, one could say that they have been considered by Europeans as "good to think about" symbolically, "good to prohibit" Politically, but "indigestible to study" anthropologically. Social anthropology has discovered them only in the last few decades, rejecting the results of the two main theoretical approaches with which they were previously studied: the sociopsychology of disadjustment and positivist, racist criminology. Beyond this rejection, however, no unanimous consensus exists as to how to categorize the "groups that don't want in"; certain scholars consider it erroneous to attempt to create a single defined category. Among the various terms proposed, "peripatetics" has had the greatest theoretical elaboration and today enjoys the greatest consensus. The three main characteristics of the peripatetic groups are: spatial mobility; subsistence based on the sale of goods and/or services outside the group; and endogamy. Since these three features may vary greatly from one group to another, some groups occupy marginal positions that are difficult to define in terms of such a theoretical elaboration. We could thus assert that the main characteristic of the "groups that don't want in" is their extraordinary structural flexibility. The "groups that don't want in" are those who can be categorized as "peripatetics" at certain historical-geographical junctures, but not at others.
Identification. Such groups are currently referred to in Europe as Gypsies and Travellers. The difference between the two categories would appear to reside in their "origin": the former are thought to come from India, the latter to be native Europeans. Since the "origin" is not always verifiable and since several present-day groups may be the result of a fusion between groups of autochthonous origin and groups of an extra-European origin, certain scholars have merged the two terms into "Traveller-Gypsies." In order to maintain the traditional distinction between Gypsies and Travellers, we can subdivide the former into two large sets based on self-denominations: (1) the "Rom" set includes all those groups whose autonym is Rom or one of its phonetic variants (Rom, Róma, Roma, Romje, etc.); (2) the "rom" set includes all those groups that, though having other autonyms, use or formerly used the term "rom," or its variants, with the meaning of "men" or "husband" (Kale, Manuš, Romaničel, Sinti). In some cases, among these last groups, "rom" can also mean "man of our group," thus becoming concurrent with the normally used autonym.
Location. With the possible exception of Iceland and Malta, all European countries are host to a permanent Presence of peripatetic groups. Although the three sets we have categorized—"Rom," "rom," and "Travellers"—are represented today in communities throughout the European continent, we can indicate, nevertheless, approximate areas of major concentration, historically speaking. An imaginary line (Rome-Vienna-Prague-Helsinki) divides Gypsy Europe into two parts: the western half is noted for its preponderance of "rom" groups, while in the eastern half there is a large majority of "Rom" groups. This line is only an indication of concentration tendencies—after the great migrations of the nineteenth and twentieth centuries some "rom" communities (especially Sinti) are found in the east and, more importantly, many "Rom" groups have moved to the west. The "Traveller" groups, though in general widely dispersed, also seem to be concentrated in specific regions, which are either marginal or enclaves of the "rom" zone. On the Celtic fringe (Ireland and Scotland), in Scandinavia (but not in Finland) and in the northern Alps (especially the Swiss part, inhabited mainly by Jenischen), these Traveller groups appear to be in the majority. From here, along a corridor running up through Alsace-Lorraine and the Rhine valley (where the Jenischen are outnumbered by "rom" groups, though their number is by no means negligible), we reach the Netherlands, where the local Travellers (Woonwagenbewoners) appear to outnumber the "rom" and "Rom" groups.
Demography. Many estimates have been made as to the numbers of Gypsies and Travellers present in Europe. Here we cite only three: Puxon (1973) counts exactly 4,745,475; Vossen (1983) gives a minimum number of 1,988,000 and a maximum of 5,621,000; Liégeois (1986) calculates a minimum of 3,421,750 and a maximum of 4,935,000. More consistent estimates, however, can be obtained using the same authors' data for areas of larger concentration. Their presence seems concentrated in the Danubian-Carpathian region (the Czech and Slovak Federative Republic; Hungary; the former Yugoslavia; Romania; and Bulgaria) with percentages Between 59.1 percent (Puxon) and 64.6 percent (Vossen) of the total European peripatetic population. The southwestern region (Spain and France) is also important with estimates between 15.2 percent (Liégeois) and 18.7 percent (Puxon), whereas percentage estimates for the nations of the former Soviet Union prove to be of little significance, given the lack of more precise data on concentration within this vast territory—between 6 percent (Liégeois) and 10 percent (Puxon). In the rest of Europe there results a more dispersed presence amounting to a percentage somewhere between 12.2 percent (Puxon) and 14.3 percent (Liégeois) of the total Gypsy and Traveller population.