Retracing the historical evolution of the Portuguese address pronoun você using synchronic variationist data

: This chapter describes the historical evolution of Portuguese address forms based on regional documents from the Inquérito Linguístico Boléo (ILB) recorded from 1942 to 1974 (University of Coimbra). It focuses on regional and social variation of the vossemecê-você family of address forms. The analysis leads to a (hypothetical) diachronic pedigree of these pronouns as well as to a survey of regional speakers’ evaluations of the forms. It concludes that in Portugal there are numerous many-layered islands of address-norm systems that may be understood as a complex system of socioglosses .


Introduction
This chapter analyzes the diachrony of the vossemecê-você 'Your Graceformal you' family of address forms in order to provide a systematic account of regional and social variation in Portuguese. The analysis is based on unpublished dialect data from Portugal from 1942 to 1974 -the Inquérito Linguístico Boléo (ILB). Basing the analysis on ILB documents showing synchronic variation, this study aims to make a diachronic reconstruction of elements illustrating the development of the series of variants associated with the pronouns vossemecê and its reduced variant, você, used in the present-day standard variety of Portuguese.
Section 2 discusses some theoretical and terminological issues, followed by the presentation of the the ILB corpus in Section 3. Section 4 examines a representative set of data, and Section 5 outlines the morphological diachrony from Latin VOSTRA MERCED to Portuguese você based on the analysis of the ILB corpus data. Section 6 takes the District of Viseu as a representative example of address form choices, and Section 7 outlines a geographical survey of você evaluations across Portugal. Finally, Section 8 summarises the results and findings of the study.
Given the complexity of address in Portuguese, the glosses and translations of the different terms used for address can only be tentative. The letters T and V in cases such as you T and you V refer to the informal (close) or formal (distant, polite) relationship expressed by the form of address. If the personal relationship is expressed with the verb only, notations such as come V may be used. Intermediate terms may appear as, e.g. you VT . In general, the terms are only glossed on their first occurrence.

Theoretical considerations
In undertaking an analysis of Portuguese address forms, there are a number of theoretical questions that need to be addressed.

Basic address conventions
Addressing one another is a universal behavior that is realized interpersonally in a given situation. We make an individual evaluation of the persons in front of us and our relationship with them. These evaluations may be described as positive or negative according to theories of "face" (e.g. Goffman 1959;Brown & Levinson 1987). Either a speaker intuitively classifies an addressee and treats her/ him accordingly, or strategically chooses an address term in order to voluntarily define or negociate the interpersonal relation. Communities united by language and/or culture may appear to be more or less dynamic in their addressing behavior. The Portuguese address system seems to be a first-rate example of a less static system if we compare it, for example, to the traditional German one, where a change from Sie to Du among adults is usually controlled by an explicit ceremonial agreement (see Glück & Koch 1998;Clyne, Norrby & Warren 2009). In Portugal, you may increase or reduce perceived interpersonal distance by using você instead of tu or o senhor, etc., or simply by switching from third person to second person, and vice versa (see e.g. de Oliveira 2009 and Hummel, in this volume). The most important forms of address in use are: tu 'informal you T ', você 'respectful but not distant you T/V ', vo(sse)mecê 'forerunner of você used in dialects for roughly the same purpose as você', o senhor/a senhora (very formal you v , e.g. you, Sir), amigo/a 'friend', camarada 'comrade, mate', excelência 'Your Excellency', third person form of the verb.

How not to be confused by a third person designating second persons
In Portuguese, the information conveyed by the grammatical person, gender or number (second or third person/feminine or masculine/singular or plural) does not always match the features of the referential person. Thus, a morphological third person form may refer to the direct interlocutor, which is in fact a referential second person (e.g., O senhor 'you V '), a morphological feminine may be used to address a man (e.g., Vossa Senhoria, a former address of high status persons), and a morphological plural can be used to denote a single person (e.g., archaic or regional vós 'sg. you V '). This study therefore establishes a terminological difference between the person(s) addressed and the grammatical forms used to refer to the person. The distinction between 1P (first person singular) to 6P (third person plural) as "morphological persons" (grammatical form) and P_1 (EGO), P_2 (VIS-À-VIS), P_3 (TRACTATUM) (person addressed by reference), etc., as semantic partners in communication, as well as combinations of these, provides unambiguous terms. The terminological set proposed here is based on conceptual deliberations on onomasiology1 by Klaus Heger, who referred to the ideas of Karl Bühler in particular (see Heger 1965 andBühler 1967). Table 1 applies this terminology to Portuguese. 1 The term onomasiology refers to a methodology that looks for linguistic items covering a given conceptual or functional domain, e.g. the lexical fields of 'anger', 'address', and so on.

"Formas de tratamento" and/or "Fórmulas de tratamento"?
In Portuguese, the distinction between the expressions formas de tratamento 'address forms' and fórmulas de tratamento 'address formulas' is an interesting one. On the one hand, there are formal differences between so-called pronominal forms such as tu/você/vossemecê, and on the other hand, the so-called nominal forms such as o senhor/a senhora or o Manuel/a Maria. At the same time, it is clear that the syntactic positions of these elements show at least two patterns: they are to be found in either the subject or object position within a given sentence.

Elaborating onomasiological tools for specific address studies
For the analysis of the ILB corpus with respect to historical traces of Portuguese address forms, it is necessary to choose the best adapted methodological tools and terminology. The term for address forms in Portuguese is formas de tratamento, literally 'forms of treatment'. Fundamentally, speakers experience verbal address as an immediate and more or less direct treatment by their interlocutor. An addressee can take this treatment positively or negatively, so that -by the speaking self (EGO) -it has to be seen as a more or less immediate access to the other person (VIS-À-VIS). In order to prevent the respective inventory of onomasiological concepts from being too near to tautology, valid conceptual instruments are required that represent a special meta-language. The interlingual definitions of these conceptual terms should be able to establish functional tertia comparationis (that is, a common point of reference that allows comparison) as linguistic tools for onomasiological and/or contrastive investigation in the domain of address (and others).

The concept of RESPECT as an onomasiological tool
Linguistic analysis in the domain of address has led many investigators to discover the relevance of concepts such as "respect", "reverence", "veneration", "courtesy", "politeness", and others (see Hummel, in this volume). These mostly Latin-based philosophical terms have been accepted as being rather universalat least for an enlarged European context (cf. the keyword respect in the Stanford Encyclopedia of Philosophy). They seem to have proved helpful for the description of the necessary conditions regarding address. Trying to establish a useful intercultural tertium comparationis, they enable us to evaluate (and to accept or not) the way in which we are addressed. In particular, the widely accepted (because of being less polysemic?) term of respect seems to offer itself as first choice for onomasiological research in the field of address forms and conventions, as we shall see below.4 Without claiming to have a perfect conceptual hypothesis for the wider field of address, I propose a suitable approximation for the initial analysis of the relevant ILB data. The aim is to discover specific language development implications in the domain of the descendants of late medieval vossa mercê. As a first step, it will be useful to distinguish the categories of in-group and out-group addressees (or: insiders vs. outsiders) as well as those of older and/or higher status addressees. The two types can be compared in order to show possible affinities. Table 2 presents some typical examples (all based on occurrences in the ILB database) used in north-eastern parts of Portugal, but also elsewhere. In order to reduce complexity, I propose a simplification with regard to some basic categories. It might be useful to have special distinctions at least for younger people, equals, and older people outside one's family, as well as a complementary categorization of higher status and lower status addressees. Although this presentation of types of address remains limited, it shows how it is possible to gradually refine onomasiological tools needed for specific purposes.

The Inquérito Linguístico Boléo (ILB)
The ILB Archives of the Faculdade de Letras at the University of Coimbra constitute a rich collection of regional language and ethnographic data from all over Portugal (Boléo 1942. They comprise about 3,100 questionnaires (plus complementary documents) completed between 1942 and 1974 by Manuel de Paiva Boléo, Professor of Romance philology and Portuguese linguistics at Coimbra (see Hammermüller 1995). The documents contain two types of spoken language data obtained from local native speakers. The first set of data was produced in 1942. Boléo posted out a huge number of questionnaires (in total around 15,000 copies were printed) to primary school teachers and local village priests, with the majority going to the north and the center of Portugal (Boléo 1942). This first printed version (with 550 items) of the questionnaire had 1,829 completed returns (Boléo 1974: 322).
A second set of data was collected by Boléo's students after a theoretical and practical introduction to dialectological methods in the second year's part of their Introdução aos estudos de filologia/linguística (2 a parte) 'Introduction to the study of philology/linguistics (part II)' (Boléo 1974: 322-323). From 1942 onwards,5 Boléo's students were required to undertake fieldwork, using an expanded 1942 questionnaire with up to 757 items (see last edition in 1978 of the 3rd version by Fátima Matias, University of Aveiro).
The results gathered via questionnaire constitute the Inquiry (Inquérito). From 1945 to 1974 (the year Boléo retired from teaching), his students were also asked to write up a Relatório, which described their local experiences and documented ethnographic and language facts they found worthwhile recording. Most of the original copies are still conserved in Coimbra, with a total of approximately 3,100 Inquéritos and 1,800 Relatórios.
The ILB database can be seen as nothing less than a linguistic (and ethnographic) bonanza for diachronic reconstruction based on synchronic variation. The data enable us to analyze oral forms of address conserved by a rural milieu (at least to a very large extent) which was -up to the second half of the 20th century -still widely marked by illiteracy. This derives from the fact that in Portugal the general school system (promoted during teachers' education in Escolas de Magistério in many district capitals) had only slowly been developed from the first half of the 20th century onwards, as illustrated by the picturesque description of a regional school by Gibbons (1984). The data thus reproduce the limited literacy of rural, lower status speakers (and as a product of internal migration, also of urban speakers).

Address forms (e.g. você) and their use in ILB questionnaires
Exploring the ILB data from 1942 onwards, Boléo and collaborators produced more than 160,000 excerpted words on paper sheets and a huge number of published and unpublished memoirs and theses written by their students (cf. Boléo 1971bBoléo , 1976Boléo , 1979. However, these efforts never led to the realization of Boléo's original idea of producing a Dicionário do Português Regional (first in 1959), proposed also as DFPM (Dicionário dos falares portugueses modernos) (cf. Boléo 1959Boléo , 1971aBoléo , 1976Hammermüller 1995). The following sections present parts of the ILB corpus that shed some light on the geolinguistic remnants of address forms used up to the latter part of the 20th century. Table 3 provides an abstract map of continental Portugal including the districts used to classify the data (see Hammermüller 2011: § 5.1). The numbers go from north (1) to south (8), including the two archipelagos in the Atlantic Ocean (9), and the names refer to districts (see map of Portugal in Lara, this volume).

The questions
This section presents an analysis of ILB summaries of address usage in the third edition modified by Boléo's former student Fátima Matias.6 Table 4 sets out the questions in the ILB survey dealing with forms of address (see Matias 1978: ILB- § 371 + § 371b (pages 75/75 A)).7 What do children use to refer to these people if the latter are absent: mother is sick or my mother is sick?
What is usually chosen in conversations with higher status people, senhor your father, senhora your mother? [5] Do you usually address your older brother or sister with mano, mana (short forms of irmão 'brother' and irmã 'sister')'

Questionnaire responses
The ILB summaries refer to several locations within Portuguese districts ordered from north to south. The answers come from different years, and illustrate a variety of address form uses that -at first glance -do not show much regularity concerning geolinguistic, sociolinguistic and transgenerational developments. Rather, it can be observed that the individual locations (mostly consisting of villages) have (or at least had) a social life of their own, using their own system of address accordingly. When looking at the regional distribution of different pronounciations, e.g. the bilabial onset [v-, b-], it must be borne in mind that many of the fieldworkers writing down the file cards, and particularly those inquiries completed by correspondence from 1942, may have preferred to follow the standard spelling, e.g. (standard) initial <v-> orthography.
'The striking fact in Afife is the almost familiar treatment of all people. The oldest address the youngest with tu 'you T ', and people of the same age also use tu. Young people address old people with 'uncle', with the obvious exception of higher status people. When the children address their parents, godfather or grandfather, they say: <you my father, you my godfather, you my grandfa-ther>*. When they address them they use <bòsê>* (variant of você 'you VT ')' 371b: < bòsê>* (não sendo ofensivo) e <bòsəməsê>*.
'When they address their parents or older relatives, they only say: <father or you father, have you done this?> They address their older siblings with their first name, without marking any difference regarding the younger ones' 371b: Dizem quando se dirigem a estranhos: <u siñuar>*. A uma mulher mais idosa podem dizer: <sãntiña>*.
'you T -for people of the same age and social condition. <bòsê, bòsəməsê> (= você, vossemecê) -for richer and older people. Senhora for higher status people'      '<vòsê>* (= você) is the most usual form of address. Children address their parents as <pajziñu, mãjziña>* (= paizinho 'father + diminutive', mãezinha 'mother + diminutive'). The custom of addressing their parents with tu starts to appear in lower status families, not without shocking the older people. Mano (short form of irmão 'brother') starts to be used amongst brothers-inlaw -a possible influence from the Algarve. Stepchildren address their stepfather or stepmother with tio 'uncle' and tia 'aunt'. Compadre is used between parents and godfathers (it also denotes this relation), as well as between participants of local religious feasts who donate money for this purpose' 'Almost all children address their parents with paizinho, mãezinha. They habitually use você'    Table 20 provides a synthesis of the most common forms of address used in different districts of Portugal. Italics are used for the most commonly used forms.     serving as a label for common judgements classifying the use of você as "less educated/offensive". Table 22 seeks to show as clearly as possible the varieties of Portuguese address forms in their conceptual positions, and especially the address form você evaluated according to the context of its use. In parallel with its etymological relatives vossemecê and vomecê, there mostly remains a positive, respectful evaluation when você is situated in a respect marking context like ó pai, você/vo(sse)mecê. However, você also acquires a special value of solidarity (in competition with the use of tu?) among (mostly younger) equals that prompted older generations and/ or many higher status individuals to label this use as estrebaria.10

From Late Latin VOSTRA MERCED to Pt. você -a pronominal pedigree
The geolinguistic distribution of the você forms (and their variants) in Table 20 will now be reconsidered in a pedigree, trying to establish the possible diachronic sequence of as much of the vossa mercê descendants as possible. For this purpose, Figure 1 presents historic types of VOSTRA MERCED (via vossa mercê) descendants belonging to the vossemecê-você family. The elements have been extracted from the available ILB corpora, and we can take them as a basis for a hypothetical pedigree in terms of diachronic reconstruction from the synchronic variationist data in the ILB. Some of the more or less generalized phonetic spellings of our examples have been normalized by bringing them close to standard (or at least near-standard) Portuguese written conventions (e.g. vocemecê > vossemecê). In order to graphically visualize this pedigree structure, uninterrupted lines are used to show the supposed mainstream, and dotted lines lead to some possible sideline evolutions. All forms in italics in Figure 1 stand for ILB-documented types; "ò" represents an open vowel and "ê" or "ô" a closed one; possible missing links are marked by "*[…]", and underlining is used to reinforce the distinctness of the forms in italics.   Table 23 groups the address forms documented in the ILB categorizing their respective use in the District of Viseu (Beira Alta). This District may be seen as representative of the surrounding areas. The Table presents relevant onomasiological categories that mainly separate in-group from out-group addressees (i.e. village insiders vs. outsiders). The documented types of address forms are mostly given by abbreviations which keep the regional marks of [v-: b-] as bilabial or labiodental initial consonants in order to shed light on the diatopic distribution of this phenomenon. Among the subjects asked to fill in the questionnaires, many (especially those completing questionnaires by correspondence from 1942) will have preferred to give the standard orthographic version instead of the genuine local phonetic variation. This means that the b-forms appearing in the answers to the questionnaires can be taken as valid evidence of this local phenomenon, while the v-spelling may hide b-pronunciation. The annotations used in the Table 23 are as follows:

An example of address form choices documented for the District of Viseu
xyz* (etc.) "*" marking a possible use of the same item in a different context vc/bc você/bocê vm/bm vossemecê/bossemecê vmc/bmc vomecê/bomecê estreb(aria) comments like <você é estrebaria; quando eu nascí já você palha roía> offs offensive -old used to/expected by less old/younger people +old used to/expected by older people old-used by less old people old+ used by older people o_pai context o pai/a mãe quer, etc. óPai_vc context ó pai/mãe, você quer, etc. >stran(gers) to strangers >urban to urban people yng/young younger people (when necessary, marked as < = by, > = to, <-> = between Table 23 only contains data from questionnaires completed later than 1942 and their respective evaluation, which contain responses about address use in outgroup relations. The original § 371 of the questionnaire was aimed explicitly at  in-group relations, and very few questionnaires had any responses about outgroup uses of address variants. However, Boléo added questions on address use in out-group relations in § 371b (see Table 3) in the first reprint of his second version of the questionnaire edited in 1962 (Boléo 1976: 25, fn. 2). Table 24 outlines the most typical evaluations found within each district. In this table, a short characterization follows each district's name in "{…}". Negative evaluation of VOCÊ is marked by underlining (VOCÊ-neg.), and positive evaluation in bold (VOCÊ-pos.). These evaluations are (partially) marked by "(neg)" or "(pos)", but mostly using the following gradations: "++, +, +/-, -, --" (where "-" corre-  sponds to algo provocante 'middly offensive' and "--" to ofensivo 'offensive'). We will use part of the original ILB differentiation (cf. ILB § 371 and 371b), but mark by "_a" comments belonging to § 371 (targeting insiders' use) and by "_b" those belonging to § 371b (added to the questionnaire only after 1942 and targeting address use with outsiders). As an illustration, "Bragança-0480a/64_a" refers to family/village in-group, and "Bragança-0480a/64_b" to strangers/village out-group.   It is challenging to give a concise survey of evaluations of the use of você when trying to base statements on the large amount of address evaluation data in the ILB documents. Thus we cannot design a simple map of a neat north-south and/ or east-west diatopic differentiation. Besides the phonetic (more or less northsouth orientated) variation of initial bilabial vs. labiodental b-/v-(making bocê contrast with você), we can, however, identify a number of essential phenomena.

Informants' evaluations according to district
The more or less typical examples given seem to confirm the not uncommon tendency of younger people (sometimes already in 1942!) to use explicit você more generally than their older neighbors. The latter seem to have still been more orientated by presupposed respectful contexts such as Ó pai, você já ouviu …? 'You Father, have you VT already seen…?' and their equivalents. This positive evaluation by younger people seems to be the case more in the western and southern parts of Portugal.
More exact statements about the present-day distribution of this phenomenon (its increase seems to be evident based on the author's own informal observations over the years) will certainly demand new and very detailed sociolinguistic research. The characteristic geolinguistic differences regarding salient você values results from the fact that, very often, in-group você usage (with family and neighborhood insiders) is evaluated more positively than out-group usage (with village outsiders and strangers).
Summing up and taking in account the ILB data reviewed here, I propose the following hypotheses: 1. The list of Portuguese address forms continues to comprise a large inventory ranging from an intimate tu to a very formal Vossa Excelência, together with different verb forms (second and third person to combine with singular and plural); 2. Not all forms are omnipresent, for all speakers or in all situations; 3. The forms may be used to express more or less polite address behavior; 4. Linguistic observation and fieldwork to describe the whole set of actively used Portuguese address forms will have to deal with a range of challenges. For example, when trying to interpret the actual meaning of a seemingly unambiguous statement like Trato-o/a por você 'I address him/her with você', two readings are possible: (a) Trato-o/a por você can actually refer to an explicit use of você as address pronoun; (b) In many regions, perhaps more in Northern regions neighboring the town of Porto, speakers saying trato-o/a por você may not imply the use of você itself at all.
In reading (a), we have to consider at least two possible explanations. Firstly, in a context where você takes over from its evolutionary predecessor vossemecê (or vomecê as its more colloquial and/or intimate realization), it will undoubtedly be accepted by everybody as respectful when it is presented, for example, as Ó pai, você quer …? This respect marking context may moreover have been pronounced only once during an ongoing dialogue (or even not at all, but be presupposed as conventional background) to support the positive acceptation of você. Secondly, apparently, the use of the address pronoun você (as had already been the case with vossemecê and vomecê, both of which did not lose -with very few exceptions -their positive values as a respectful address) was freed from an obligatory respect marking context (accompanied by a vocative like ó Fulano, … or equivalents).
In some milieus, as represented, for example, by fishermen or muleteers,12 besides the continuing intimate (or family bound) tu of solidarity, the use of você (besides or instead of tu?) seems to have become an expanding marker of solidarity not only for lower status workers. The same lower status speakers, however, may have been disgusted when higher status representatives (excepting perhaps their direct superiors) addressed them by an explicit você. This specific solidarity-based value may find its echo in notorious address blaming formulas like você é estrebaria, still to be heard particularly in the northern parts of Portugal. As original sources for those statements we may imagine, among others, primary school teachers and clergymen ("their masters' voice") who were influenced by their own experience and pedagogical studies in urban centers (mostly orientated towards the standards of Lisbon/Coimbra/Porto). So they may have been propagating their respective cultural and language norms.
In reading (b), one has to bear in mind that nominal forms found their place in actantial positions, as in O pai quer …/O senhor Fulano quer …/o João/a Maria quer …, etc., with their deictic function paralleling other paradigmatic elements that we commonly call pronouns. These nominal forms can most easily replace a você meant with disgust. This tendency seems to apply rather for certain higher middle-class individuals -who partly had also been used to addressing their siblings by mano 'brother' and mana 'sister'-and all those pupils influenced by their teachers' judgements. This use has presumably originally been established in and around the town of Porto.

Concluding remarks
Taking in account the rich ILB data and our attempts to discover diachrony within (a relative) synchrony with regard to special forms and modes of Portuguese address conventions, what sort of methodological approach can be proposed? We should perhaps not restrict our analysis to geolinguistic, sociolinguistic and transgenerational variation (see Endruschat & Schmidt-Radefeldt 2015: 214-240), and search instead for another type of explanation. This makes it difficult to find a way out of this complex and dynamic network -not least for the inhabitants of Portugal themselves: it is remarkable how easily Portuguese people speak about their address uncertainties or their experiences in general. Looking at heuristic categories such as isoglosses, traditionally delimiting language islands -based on phonetic, lexical, or other criteria -promises even better access to that multiplicity of rapidly expanding address subsystems that existed all over 20th century Portugal.
There are cultural observers who view as a special Portuguese phenomenon the impressive socio-cultural coexistence of different ages or historical periods up to the present day. The German journalist, author and poet Hans Magnus Enzensberger, alluding to the notorious synchronic coexistence of historical ves-tiges in present-day Portugal, proposed Isochrones to describe possible features of a topology of time. These should describe islands of time paralleling geographical topology (Enzensberger 1986: 49-50).
In accordance with this characterization, I would like to assert for 20th century Portugal (and probably not only as a phenomenon limited to Portugal) the existence of numerous many-layered islands of address-norm systems. These could be understood as being delimited by socioglosses defining address domains which are overlapping and exchanging with others, as individual speakers may, at least partly, participate in different address systems. My terminological proposal of socioglosses owes something to a similar effort in the domain of modality constructions by Franco Benucci, speaking of faisceaux de chronoglosses 'bundles of diachronic isoglosses' (Benucci 1988: 6).
These conventional address islands -delimited and perhaps connected by the respective socioglosses -are certainly subject to continuous conflicts with each other and with educational address standards suggested by teachers or other authorities. These authorities will have received their orientation from layers of society of so-called higher status. In any case, further, more detailed studies will be required to more fully analyze the address complexes (or islands) documented by ILB data.