A History of the Spanish Language
through Texts
‘A meticulous and enlightening examination of a broad selection of texts, which
are representative of Spanish during the last millennium and across the world
. . . elegantly and succinctly presented. An indispensable tool for all those working
or interested in the history of the Spanish language.’
Ralph Penny, Queen Mary and Westfield College, University of London
A History of the Spanish Language through Texts examines the evolution of the Spanish
language from the Middle Ages to the present day.
Including chapters on Latin American Spanish, US Spanish, Judeo-Spanish
and Creoles, the book looks at the spread of Castilian as well as at linguistically
interesting non-standard developments. Pountain explores a wide range of texts,
from poetry, through newspaper articles and political documents, to a Buñuel
film script and a love letter.
A History of the Spanish Language through Texts presents the formal history of the
language and its texts in a fresh and original way. The book has user-friendly
textbook features such as a series of keypoints and a careful indexing and cross-
referencing system. It can be used as a freestanding history of the language
independently of the illustrative texts themselves.
Christopher J. Pountain is a University Lecturer in Romance Philology at the
University of Cambridge and a Fellow of Queens’ College. He has over twenty
years’ experience of teaching Spanish and Romance linguistics. His publications
include Using Spanish (CUP 1992) and Modern Spanish Grammar (Routledge 1997).
1111
2
3
4
5111
6
7

8
9
40111
1
2
3
44
45111
1111
2
3
4
5
6
7
8
9
1011
1
2
3111
4
5
6
7
8
9
20111
1
2

1111
2
3
4
5
6
7
8
9
1011
1
2
3111
4
5
6
7
8
9
20111
1
2
3
4
5
6
7
8
9
30111

c
i
s
G
r
o
u
p
•
R
O
U
T
L
E
D
G
E
First published 2001 by Routledge
11 New Fetter Lane, London EC4P 4EE
Simultaneously published in the USA and Canada
by Routledge
29 West 35th Street, New York, NY 10001
Routledge is an imprint ofthe Taylor and Francis Group
© 2001 Christopher J. Pountain
All rights reserved. No part of this book may be reprinted or reproduced
or utilised in any form or by any electronic, mechanical, or other means,
now known or hereafter invented, including photocopying and recording,
or in any information storage or retrieval system, without permission in
writing from the publishers.

3111
4
5
6
7
8
9
20111
1
2
3
4
5
6
7
8
9
30111
1
2
3
4
5
6
7
8
9
40111
1
2

5
6
7
8
9
30111
1
2
3
4
5
6
7
8
9
40111
1
2
3
4
45
46
47111
vi Contents
Contents
List of illustrations xi
List of maps xii
Transliteration and other notational conventions xiii
List of abbreviations xvi
Acknowledgements xvii

9
1011
1
2
3111
4
5
6
7
8
9
20111
1
2
3
4
5
6
7
8
9
30111
1
2
3
4
5
6
7
8

century) 104
14 Aljamiado aromatherapy. An aljamiado document
from Ocaña (late fourteenth–early fifteenth century) 109
15 Mad with love. Fernando de Rojas, La Celestina
(1499) 115
VIII The Golden Age: linguistic self-awareness 122
16 The first grammar of Castilian. Antonio de Nebrija,
Gramática de la lengua castellana (1492) 122
17 The ‘best’ Spanish? Juan de Valdés, Diálogo
de la lengua (1535) 128
18 The etiquette of address. Gonzalo de Correas, Arte
de la lengua española castellana (1625) 133
XI The Golden Age 141
19 A model for Castilian prose. Juan de Boscán,
El cortesano (1534) 142
20 Standing on ceremony. Lope de Rueda, Eufemia
(mid-sixteenth century) 146
21 Santa Teresa, Letter to Padre García de Toledo
(1562) 152
1111
2
3
4
5
6
7
8
9
1011
1

2
3
4
45
46
47111
viii Contents
22 Streetwise in Seville. Francisco de Quevedo,
El Buscón (1626) 159
X The Enlightenment 167
23 A policy for linguistic standardisation, from the
Diccionario de autoridades (1726) 167
XI Modern Peninsular Spanish 174
24 Renting a flat in nineteenth-century Madrid. Ramón
de Mesonero Romanos, Escenas matritenses (1837) 174
25 A busy housewife. The spoken Spanish of Madrid
(1970) 178
26 King Hassan of Morocco arrives in Spain. A
newspaper article (1989) 183
27 An Andalusian maid bemoans her lot. Carlos Arniches, Gazpacho
andaluz (1902) 187
XII Latin America 191
28 A love letter from Mexico (1689) 193
29 The gaucho conscript. José Hernández,
Martín Fierro (1872, Argentina) 199
30 Caring for the wounded. Marta Brunet,
Montaña adentro (1923, Chile) 203
31 On the streets of Mexico City. Luis Buñuel,
Los olvidados (1951, Mexico) 209
XIII US Spanish 214

7
8
9
20111
1
2
3
4
5
6
7
8
9
30111
1
2
3
4
5
6
7
8
9
40111
1
2
3
4
45
46

4
5
6
7
8
9
20111
1
2
3
4
5
6
7
8
9
30111
1
2
3
4
5
6
7
8
9
40111
1
2
3

9
20111
1
2
3
4
5
6
7
8
9
30111
1
2
3
4
5
6
7
8
9
40111
1
2
3
4
45
46
47111
Contents xi

4
5
6
7
8
9
30111
1
2
3
4
5
6
7
8
9
40111
1
2
3
4
45
46
47111
xii Contents
Transliteration and other
notational conventions
Arabic transcription
The systems of equivalences used by the periodical Al-Andalus, widely followed in
the Spanish-speaking world, are used in this book. Differing usage by authors

d]
d
da¯l [ð]
r ra¯’ [
r]
z za¯y [
z]
s sı¯n [
s]
sˇ sˇin [
ʃ]
s¸ s¸a¯d [
s] (‘emphatic’, with centre
of tongue lowered)
d¸ d¸a¯d [
d] (‘emphatic’, with centre
of tongue lowered)
t¸ t¸a¯’ [
t] (‘emphatic’, with centre
of tongue lowered)
z¸ z¸a¯’ [
ð](‘emphatic’, with centre
of tongue lowered)
‘ ‘ayn [
ʕ
] (voiced pharyngeal)
1111
2
3
4

5
6
7
8
9
40111
1
2
3
4
45
46
47111
Contents xiii
Arabic Transliteration Name of letter Approximate phonetic value
letter in Modern Arabic
g (Cor. g˙) gayn [
γ]
f fa¯’ [
f]
q qa¯f [
q] (voiceless uvular)
k ka¯f [
k]
l la¯m [
l]
m m
ı¯m [m]
n nu¯n [
n]

u]
a¯[
a]
ı¯ [
i]
u¯[
u]
mark of sˇadda
consonant
doubling
Phonetic symbols
The symbols used in this book are generally those of the International Phonetic
Alphabet and so are not described further here. The signs [
j] and [w] have been
used to indicate both onglides and offglides, e.g. [
je] and [ej], [we] and [aw].
A distinction is made between phonetic and phonemic transcription, the former
being indicated, as is usual, by square brackets [ ], and the latter by obliques / /.
Phonetic transcription is used when the point under discussion is primarily a
matter of pronunciation, and phonemic transcription when systematic distinc-
tions are implied. There is no extensive discussion of the phonemic status of
particular sounds unless this is crucial to the matter in hand; it has been found
more convenient to treat /
w/ and /j/ as phonemic throughout (but see Keypoint:
vowels
, p. 296), and a distinction is made between /r/ and /ɾ/ only in intervo-
calic occurrence.
1111
2
3

4
5
6
7
8
9
40111
1
2
3
4
45
46
47111
xiv Transliteration and other notational conventions
Latin
Vowel length is indicated where significant, in line with practice in modern Latin
dictionaries (see
Keypoint: vowels, p. 296). Citation forms are given in square
brackets (see p. 12).
Other symbols
See p. 12 for an explanation of the use of the symbols <, >,
р, у, ←, →, ?
and √.
|| cognate with
° is used to indicate a construction in Latin or Old Castilian that is hypothe-
sised and not directly attested.
* indicates an unacceptable construction.
1111
2

3
4
5
6
7
8
9
40111
1
2
3
4
45
46
47111
Transliteration and other notational conventions xv
Abbreviations
Ar. Arabic
Cat. Catalan
Du. Dutch
Eng. English
f. folio
Fr. French
Germ. Germanic
Gr. Greek
It. Italian
l., ll. line, lines
Lat. Latin
MS manuscript
MSp. Modern Spanish

1
2
3
4
5
6
7
8
9
30111
1
2
3
4
5
6
7
8
9
40111
1
2
3
4
45
46
47111
xvi Contents
Acknowledgements
The author and publishers would like to thank the following copyright holders

1111
2
3
4
5
6
7
8
9
1011
1
2
3111
4
5
6
7
8
9
20111
1
2
3
4
5
6
7
8
9
30111

student Yasmin Lilley, whose consuming interest in caló (Chapter XV) awakened
mine, to Larry and Simone Navon, and to Avi Shivtiel of Cambridge University
Library, who patiently helped me with the Hebrew (Chapter IV).
Finally, a special word of thanks to my long-suffering family, who have been
subject to even greater abandonment than usual during the writing of this book,
for which the dedication attempts to make amends.
Christopher J. Pountain
Cambridge, January 2000
1111
2
3
4
5
6
7
8
9
1011
1
2
3111
4
5
6
7
8
9
20111
1
2

word examples for the former and paradigms, or sets of morphological forms,
for the latter. This method has followed from the belief that changes in these
areas essentially conform to regular patterns which can be abstracted from such
data. Vocabulary has rarely been approached in a similarly systematic way (indeed,
many linguists would deny that it can be), and has thus been the almost exclu-
sive preserve of compendious etymological dictionaries which have generally dealt
with the semantic histories of words on an individual basis. Syntax was for a long
time a relatively neglected area of historical linguistics, finding a natural home
in neither of these formats. Textual references to individual examples are of course
frequently given, but continuous texts are cited less often, and usually in the form
of an appendix.
A History of the Spanish Language through Texts reverses these prior-
ities. Making the study of individual texts a starting point for the history of the
language does not lend itself to a comprehensive and systematic account of phono-
logical and morphological change; it is like turning jigsaw pieces out of a box
rather than seeing the whole picture at once. It cannot be guaranteed, even with
careful choice of texts, that all phonetic changes will be illustrated, and quite
unrealistic to assume that even a representative selection of morphological forms
will emerge. (At the same time, jigsaws are easiest to solve by looking at the
picture on the box, and accordingly, I have described some of the more impor-
tant formal features of the history of Spanish in a series of some forty Keypoints,
which are listed on p. 262–97, and to which many of the points made in connec-
tion with individual texts are cross-referenced.)
What is the justification for such an approach? First, Romance linguistics occu-
pies a unique status within historical linguistics precisely because of the wealth
of its textual records, both in the parent language (Latin) and in the many
Romance varieties observable from the Middle Ages down to the present day
(see Malkiel 1974). The interpretation of texts as a source of data is therefore a
skill which no Romance linguist can possibly ignore, the more so, because in fact
it sometimes turns out that the jigsaw piece does not exactly correspond to the

9
30111
1
2
3
4
5
6
7
8
9
40111
1
2
3
4
45
46
47111
idealised. Second, the study of texts embeds language in its cultural, social and
historical matrix, a dimension which, while a hallmark of the work of the great
Spanish philologist Ramón Menéndez Pidal (1869–1968), and recently reinstated
by the insights of modern sociolinguistics, has been, and continues to be, very
seriously neglected by some structuralist approaches. Third, the study of contin-
uous texts allows us to pursue lexical histories in a more interesting way, since
we can see vocabulary used in context, and to give a higher priority than is usual
to syntax, since continuous texts are the only satisfactory source of syntactic data.
Fourth, we will also be able to investigate questions of register and style, which
have often been scarcely mentioned in formal histories, though such variation is
increasingly recognised as being of crucial importance to an understanding of

respect, but also as a basis for comparison with the texts which come from earlier
periods, for which on the whole we have much less overt evidence of such variation.
For the sixteenth, seventeenth and eighteenth centuries I have included samples of
such important ‘secondary’ documents as grammars and other writings on language
1111
2
3
4
5
6
7
8
9
1011
1
2
3111
4
5
6
7
8
9
20111
1
2
3
4
5
6

It can of course be plausibly objected that a history of a language which is based
on literary texts is in an important sense elitist, since it is unlikely to take into
account the spoken language or other written registers of the language such as
legal or technical usage. This is an important objection which I fully accept, and
which it is of crucial importance to bear in mind at every point in the exploita-
tion of literary material. However, there are also advantages in using at least
some literary material. Literary authors can be highly sensitive, consciously or
subconsciously, to different linguistic registers, even if their representations of these
registers are sometimes rather conventionalised (see especially Texts 13, 20, 22,
30, 31 and 37). Quite apart from their intrinsic merits, literary texts have for
better or worse been widely used and discussed as source material in the philo-
logical tradition, and students of the history of the language may therefore
reasonably be expected to be in a position to engage in that discussion. Lastly,
the language of literature has often served as an important model for the stan-
dard language (see especially Texts 19 and 23), and has hence been an important
factor in its development.
Further reading
Mondéjar (1980).
The ‘Spanish language’
One of the things that will hopefully become apparent in this book is that it is
extremely difficult to delineate in a linguistically rigorous way any notion of the
‘Spanish language’. In the present day, the notion of the ‘Spanish language’ is
often used, with some justification, to refer to the standardised language that has
official status in a number of countries, including Spain, and under this view
‘Spanish’ would be equatable with the codification of vocabulary and grammar
periodically made by the Real Academia Española. But such a view is in prac-
tice impossibly restricting. Official codifications of this kind always lag behind
the reality of new words and turns of phrase which are constantly found in the
use of native speakers. Furthermore, even within educated registers of usage with-
in the ‘Spanish-speaking world’, there is much variation (one thinks immediately

6
7
8
9
30111
1
2
3
4
5
6
7
8
9
40111
1
2
3
4
45
46
47111
Preliminaries 3
result in mutual incomprehensibility among speakers. However, there would still
remain awkward questions of identity with such phenomena as chicano (Texts 32a
and 32b), Judeo-Spanish (Texts 33 and 34), caló (Texts 35 and 36) and especially
creoles (Texts 39 and 40). The rationale for giving all of these attention in this
book is that from a historical point of view all these are developments in various
ways of ‘Spanish’. For the medieval period, the label ‘Spanish’ is in fact totally
inappropriate (see Names below), and in addition to texts originating in Castile,

different contexts of use (‘communicative competence’), and in particular educated
speakers will also have judgements about which of a number of coexisting vari-
ants is ‘correct’, based on puristic teaching, aspiration to or affirmation of a
particular social group, or personal prejudice: evidence of such judgements is to
to be found in Texts 17 and 18.
A serious objection to the exploitation of texts as a source of data is posed by
some sociolinguists (e.g. Labov 1994; Milroy 1992: 5), who claim that certain
aspects of change can only be studied empirically in the spoken language of the
1111
2
3
4
5
6
7
8
9
1011
1
2
3111
4
5
6
7
8
9
20111
1
2

is true that writing is secondary to speech, the written language is an important
manifestation of language in any literate community, and that its study should
therefore not be marginalised.
Names
Turning now to the historical perspective, we come up against a number of termi-
nological problems. Even today, ‘Spanish’ continues to be known both as español
and castellano, the latter term reflecting the fact that in origin ‘Spanish’ was the
Romance variety of the area of Old Castile. These two terms currently have a
number of connotations: español is sometimes resisted by those who wish to deny
the association of the language with España, while castellano sometimes carries the
notion of ‘standard’ Spanish; but mostly they may be regarded as interchange-
ably denoting ‘Spanish’. The term español only really gained currency once the
political notion of España came into existence with the union of the Castilian and
Aragonese crowns in 1474 and the conquest of Granada in 1492, but español is
clearly nothing more than castellano by another name (see Text 17), and so this
double nomenclature is linguistically unimportant: in this book I shall likewise
use both the terms ‘Castilian’ and ‘Spanish’, ‘Castilian’ being primarily reserved
for the pre-1492 language and for contrast with other Romance varieties of the
Iberian Peninsula, and ‘Spanish’ being used for the modern language and for
contrast with other national Romance standards.
When did ‘Spanish’ begin?
Much more problematic is when ‘Spanish’/’Castilian’ began, the question begged
by any history of the language. Any precise date or event is arbitrary (see 2.0),
and it is tempting to regard the question as unimportant, since there is a continuum
between ‘spoken Latin’ and ‘spoken Romance’ (see Chapter II, introduction). But
there is an answer of sorts. Just as we may say that español begins when speakers
become conscious of España and the fact that español is its official language, so
the same point may be made about romance which is recognised as being different
from Latin and castellano which is recognised as being different from, say, aragonés
or bable. ‘Castilian’ may therefore be said to start at the point when there is a

4
5
6
7
8
9
30111
1
2
3
4
5
6
7
8
9
40111
1
2
3
4
45
46
47111
Preliminaries 5
periodisation can be shown to be unsatisfactory, and yet attempting to write a his-
tory of the language without recourse to some general notions like ‘Old Spanish’,
‘Golden-Age Spanish’, etc., would be cumbersome even if it were feasible. The
ideal solution of being able to characterise a particular phenomenon as being typ-
ical of, say, ‘the speech of upper-class Toledans between 1320 and 1480’ rather

3
4
5
6
7
8
9
30111
1
2
3
4
5
6
7
8
9
40111
1
2
3
4
45
46
47111
6 Preliminaries