The evolution of the written accentuation system in Spanish since 1726

Douglass (1964, 1988) examines the Greek and Roman precursors and early uses of the written accent in Spanish, tracing its development to 1726, when the first Diccionario de Autoridades from the Real Academia Española (RAE) appeared. This system continues to evolve even today, but there is little published research on its development over the last almost 300 years. Along the way there have been inconsistencies, missed generalizations, and overlapping, unmotivated or ad hoc rules that have undergone significant revisions in the Diccionarios, Ortografías, Prontuarios, and Gramáticas by the RAE between 1726 and 2010. The ambiguous and tentative first rules of written accentuation in Spanish have, over the years, been modified and clarified, evolving into a streamlined and elegant system that constitutes a useful tool for the pronunciation and spelling of unfamiliar words. Early accentuation rules differed widely from the current ones in questions relating to prosody, the accentuation of vowels appearing singly as separate words, the use of the diacritical accent, diphthongization and hiatus, the relation between written accentuation and grammatical categories, and the consolidation of rules for the accentuation of words ending in -n, -s, or a vowel, all of which are explored in the current study.


Introduction
Any instructor faced with the question of why and how Spanish came to use a system of written accentuation would be hard pressed to find many answers in the existing literature.The purpose of this study is to draw attention to the only research we have found regarding the origins of the written accent in Spanish (Douglass 1988), and to provide an understanding of how written accentuation has evolved through modernday Spanish.This study grew out of a question asked in a History of the Spanish Language class that, at the time, went unanswered.The class had been comparing a passage from Genesis from different periods (the fourth and eighth centuries and Modern Spanish) (Pharies 2007:10).Among other changes, reflected in the texts was the appearance of written accentuation after the fourth century.Thus came the fateful question: "Profe, where did written accentuation in Spanish come from?" Douglass (1988) answers the question in part, tracing the development of written accentuation up to the Discurso proemial of the first Diccionario de Autoridades (1726), but to the best of our knowledge, there is no research that traces the evolution of the rules from 1726 until the present.

Accentuation in Classical Latin
Classical Latin did not require written accentuation because stress was invariably determined by syllable weight.A light syllable ended in a short vowel, and a heavy syllable in either in a long vowel or a vowel followed by a consonant.Words with two syllables were always stressed on the penult (e.g.facit [fá-kit], from facere 'to make').In words of more than two syllables, if the penult was heavy it was stressed (e.g.similāre [si-mi-lá:-re] 'similar'), and if it was light, the antepenultimate syllable was stressed (e.g.bibere [bí-be-re] 'to drink').
In the evolution from Latin to Spanish the tonic accent has remained highly stable, in most cases falling on the same syllable of a word in Spanish as it did on the same word in Latin (Penny 2002:41).Nonetheless, phonological changes caused a need for written accentuation in Spanish.These included loss of Latin long vowels, apocope of word-final [-e] (e.g.aequāle>igual 'equal'), and the creation of the palatal glide from atonic /e/ or /i/ in hiatus (e.g.superbia [su-pér-bi-a]>soberbia [so-ér-i a] 'pride') (Lapesa 1986, Lloyd 1987, Penny 2002, Pharies 2007).

Written accentuation in Modern Spanish
An examination of stress assignment in Spanish is beyond the scope of this paper, but a number of approaches have been discussed elsewhere (e.g.Harris 1983, 1995, Lipsky 1997, Roca 1988, 1990a, 1990b, 1992, 2006).Rules for stress and written accentuation in Spanish are found in many phonetics and introductory linguistics text books (e.g.Dalbor 1997, Teschner 2000, Guitart 2004, Hualde et al. 2010).Words ending in a vowel (-V), -n, or -s (e.g.casa [ká-sa] 'house ', comen [kó-men] 'eat (1 st p. pl.pres.indic.)', comes [kó-mes] 'eat (2 nd p. sg.pres.indic.)'are stressed on the penult (llanas).Words ending in other consonants (e.g.hablar [a-lár] 'to talk') are stressed on the final syllable (agudas).Words that conform to these patterns require no accent mark.Written accents: 1) indicate the stressed syllable in words that do not follow the above rules; 2) reflect hiatus between vowels that would otherwise form a diphthong (e.g.hacia [á-si a] 'towards' vs. hacía [a-sí-a] 'used to do'), and; 3) distinguish between homophones with different meanings and grammatical functions (e.g.se '3 rd p. pronoun'/sé 'I know').Words with stress on the third or fourth syllable from the right edge of the word (esdrújulas, sobresdrújulas) always have a written accent (e.g.pájaro [pá-xa-ro] 'bird', dígaselo [dí-a-se-lo] 'tell it to him/her/them').The rules are based on letters, not sounds.For example, vejez [be-xés] 'old age', ends with the phoneme /s/ in most varieties of Spanish, but is spelled with z, and so is stressed on the final syllable.Likewise, Uruguay ends in the sound /i / but the letter -y, and thus, is by rule stressed on the final syllable.This first appeared in the Gramática (GRAE) of 1883: "La y final, aunque suena como vocal, se considera como consonante para los efectos de la acentuación" 'Final -y, although it sounds like a vowel, is considered a consonant in terms of written accentuation' (RAE 1883:365).
It is generally agreed that there are no present-day regional differences in written accentuation rules.The Prólogo of the Ortografía of 1999 affirms that due to the collaboration between the Real Academia and 21 sister academies in the Americas and the Philippines, standard Spanish orthography is "verdaderamente panhispánica" 'truly pan-Hispanic' (RAE 1999:V).Martínez (2010) points out that the general acceptance of a standardized orthography is a concept that goes back to Nebrija, who writes of: "el común consentimiento de los que tienen poder para hazer uso" 'common agreement among those with the power to make use of it.'(Nebrija 1492:38).Martínez traces the history of the standardization of Spanish orthography, and argues that while it is more difficult to standardize syntactical and lexical variation across dialects: ...es en la ortografía donde la convención inherente a la estandarización puede llegar a conseguir un grado mayor de estabilidad y aceptación, independientemente de los problemas relacionados con su aprendizaje."(Martínez 2010:151) ...it is in the orthography that the convention inherent in standardization reaches its highest level of stability and acceptance, independently of any learning problems associated with it.
With regard to learning problems, much research shows that native Spanish speakers from primary school through university, not only foreign language learners, routinely fail to apply the rules of Spanish orthography.For example, Medina writes that even at the university level: ...muchos de los alumnos de la asignatura de Lengua Española de primero de Filología [son] capaces de escribir en un examen varios folios sobre el acento en español sin colocar una sola tilde.(Medina 1994:73) ...many Spanish language students in the first year of Philology are capable of writing several pages on an exam about the accent in Spanish without ever using a single accent mark.

Early accentuation rules
According to Douglass (1988), the Discurso proemial was a tentative first step in establishing the rules of written accentuation.In subsequent years inconsistencies, missed generalizations and ad hoc rules were modified and clarified.Several illustrative examples follow.In all citations, spelling and accentuation are reproduced as they appear in the originals.All translations are our own.The Discurso proemial of 1726 states: En el uso de los acentos tambien se ha padecido grande equivocación, causada de la ignorancia, ò poca advertencia de su uso.En la Léngua Latina son tres, grave, agúdo, y circunflexo.En la Léngua Castellana el circunflexo, que se forma assi ˆ, no tiene uso alguno, y si tal vez se halla usado por algun Autór, es sin necessidád, porque no sabémos yá el tono que los Romanos usaban y explicaban con este accento.(RAE 1726:LXIV) In the use of accents there has been much error, caused by ignorance or little knowledge of their use.In Latin there are three, the grave, the acute, and the circumflex.In Castilian the circumflex, formed ˆ, has no use, and if it is used by any author it is without need, because we no longer know the tone that the Romans used and indicated with this accent.
Use of the circumflex accent is explicitly rejected, but reappears in the Ortografías of 1763 through 1792 to indicate when ch is pronounced /k/ (e.g.chîromancia 'chiromancy'), and x, /ks/ (e.g.exâcto 'exact').However, if another accenting rule already assigned an acute accent, the circumflex rule did not apply: Pero se advierte que hay algunos nombres en que la Ch se pronuncia como K y la X como CS, donde no es practicable el uso del Circunflexo, porque la vocal en que correspondia ponerle, debe acentúarse segun las reglas dadas,v.g.Chímica, Chímico, hexámetro, exámen, &c.Y la razon de preferir el acento consiste en que el circunflexo, ó capucha es una Nota nuevamente usada para el oficio que se la aplica, y no tan necesaria: porque en estas voces, como facultativas y comunes á otras lenguas, es mas conocida esta pronunciacion de la Ch y de la X que no el acento que las mismas voces tienen en castellano.(RAE 1763: 123) But be aware that there are some nouns in which Ch is pronounced as K, and X as CS, where use of the circumflex is not practicable because the vowel on which it would appear should be otherwise accented according to the given rules, v.gr.Chímica 'chemistry ', Chímico 'chemical', hexámetro 'hexameter', exámen 'examination', etc.And the reason that the accent is preferred is that the circumflex, or the hood, is a mark newly used for the purpose it has been assigned, and is not so necessary, because in these words, which are optional and common to other languages, this pronunciation of Ch and X is better known than which stress these words have in Castilian.
This inconsistency was repeated in actual practice.The Diccionario of 1729 lists chiromancia with no circumflex, those of 1780, 1783, and 1791 list chîromancia, and in 1792 chiromancia is listed again without a circumflex.From 1803 onward it appears with its modern spelling, quiromancia.The GRAE of 1851 decidedly put the circumflex to rest: Otro acento que se llama circunflejo, figurado así (ˆ), se usaba ántes tambien sobre algunas vocales [...].Por manera que no habiendo ya motivo para incurrir en equivocacion [...] es inútil servirnos del acento circunflejo, y así quedará solo el que se llama agudo para notar la letra sobre qué se haya de apoyar al pronunciar las palabras que lo exigen en castellano... (RAE 1851:303) Another accent, the circumflex, formed (ˆ), was used before on certain vowels [...].Since there is no longer any motive for error [...] it is useless to use the circumflex, so that only the accent called acute will be used to indicate which letter should be most strongly pronounced, in words that require it in Castilian.
There were also missed generalizations leading to abundant overlapping rules of various types, beginning, but not ending, with the Discurso proemial of 1726: ...no hai necesidád de tal accento, respecto de que por su naturaleza es larga la pronunciación, exceptuandose de esta regla las voces en que las consonantes son muda y líquida, en cuyo caso muchas tienen breve la vocál antecedente, especialmente en la prosa: como álgebra, árbitro, Cáthedra, Fúnebre, Lúgubre, Quádruplo.Lo mismo se debe observar en las voces llamadas esdrúxulos: como águila, Música, Máximo, Philósopho, Theólogo, Escolástico, Doctísimo, en las quales la sylaba penúltima es breve: y para que se reconozca que la pronunciación no se detiene sobre ella sino sobre la antecedente, se pone el accento agúdo sobre esta.(RAE 1726:LXIV) ...there is no need for such an accent, considering that the pronunciation in question is naturally long, except in words with a mute and liquid consonant, in which case many have a preceding short vowel, especially in prose: like álgebra, árbitro 'arbiter', Cáthedra 'professorship', Fúnebre 'funereal', Lúgubre 'lugubrious', Quádruplo 'quadruple'.The same should occur in words called esdrújulos 'proparoxytone'; like águila 'eagle', Música 'music', Máximo 'maximum', Philósopho 'philosopher', Theólogo 'theologian', Escolástico 'scholastic', Doctísimo 'extremely well-versed', in which the penultimate syllable is short: and to indicate that the pronunciation does not linger on the penult, but on the preceding syllable, the acute accent is placed on the preceding syllable.' There are two rules, the second to account specifically for esdrújulas (accented on the third syllable from the right edge of the word).However, all the examples used to illustrate the first rule are also esdrújulas.
Finally, there are unmotivated, ad hoc rules like those for patronymics, last names given to children derived from the name of the father, for example, Fernández, from Fernando.In the somewhat circular rule from the Ortografía of 1741, no written accent is required where it otherwise would be, because the pronunciation of the surnames is so commonly known: Los Apellidos Patronímicos acabados en az, y en ez, como Diaz, Perez, Marquez, y otros conocidos, nunca se hallan de acentuacion aguda, acaso por haberse formado de los plurales de los nombres proprios, y no haber mudado su acentuacion, aunque hayan mudado en Z la S de su terminacion, y siendo tan conocido, comun, y freqüente el modo de acentuarse, no necesitan de nota, que los distinga de las otras voces de su misma terminacion.(RAE 1741:253-254) Patronymic surnames ending in az and ez, like Diaz, Perez, Marquez, and other well-known ones, never have an acute accent, perhaps because they were formed from the plural of the proper nouns, and because their accentuation did not change, even though their last letter changed from S to Z, and because their accentuation is so well-known, common, and frequent, they do not require a mark to distinguish them from other words with the same ending.
In a rule from the Prontuario of 1854, surnames that are adjectives or the same as common nouns have no written accent mark, even if they end in the same letter as other surnames which do: Se dará tambien acento á los apellidos terminados en consonante, cuya última sílaba fuere larga, como Ardanáz, Muñíz; pero no lo llevarán los apellidos que fuesen al mismo tiempo sustantivos comunes como Calderon, Doncel, Pastor ó adjetivos como Leal, Capaz, ó nombres geográficos como Teruel, Motril, ó verbos como Escalar, Amar.(RAE 1854:33-34) An accent will also be given to surnames whose last syllable is long and that end in a consonant, like Ardanáz, Muñíz, but surnames that are also common nouns like Calderon 'paragraph mark', Doncel 'young nobleman', Pastor 'shepherd', or adjectives like Leal 'loyal', Capaz 'capable', or geographical names like Teruel, Motril, or verbs like Escalar 'to climb', Amar 'to love', will not have one.
Except for Calderon (today Calderón), all the words fall within the unmarked pattern of final stress on non-verbs ending in a consonant, so their lack of written accentuation can be explained without reference to other factors.

Written accentuation since 1726
Current accentuation rules differ widely from the earlier rules.Early in its evolution, the acute accent was almost always described as indicating length rather than stress.The grave accent was maintained on à,è,ò, and ù when used as separate words, and was later replaced by the acute accent, which has since ceased to be used in these cases.There was, until quite recently, confusion regarding the accentuation of monosyllables.There was no clear rule for indicating hiatus, and no clear understanding of diphthongization.Accents were required for certain grammatical categories and verb forms.And finally, unlike in the current rules, the plural markers -n and -s were grouped with all other consonants.Based on these differences this study sought to answer the following questions: 1. Until when was accent equated with syllable length rather than stress?2. Until when was the written accent maintained on vowels used as separate words?3. When did the diacritical accent on monosyllables become clearly defined? 4. When were the rules for indicating hiatus clearly defined? 5.When were certain verb tenses no longer defined as requiring written accents?6.When were -n, -s and -V grouped together in a single rule?

Length versus stress
The Diccionario of 1726 defined the function of the accent in terms of syllable length: En nuestra Léngua los accentos no sirven para explicar el tono, sino para significar que la sylaba que se accentúa es larga.(RAE 1726:LXIV) In our language the accent is not used to indicate stress, but rather that the accented syllable is long.
According to Douglass (1988) larga meant 'stressed' rather than 'long'.This was: "...another case of confusion caused by developments from Latin to Spanish, since a vowel in Latin was 'long by nature' if it was double in pronunciation (long) and 'long by position' if it was followed by two or more consonants" (Douglass 1988:931).
The Ortographia of 1741 is closer to the modern Ortografías than to those of its time, defining accent in terms of greater prominence, regardless of syllable length: Nosotros, como solo acentuamos la pronunciacion en una sylaba de la palabra, haciendo en ella fuerza con animar algo el pecho, y deteniendo con alguna pausa la voz, solo necesitamos (sin distinción de sylabas breves, ó largas) de una nota de acento, con la qual, puesta sobre la vocal, damos a entender, y conocemos que en ella hemos de acentuar... (RAE 1741:245) Since we only stress the pronunciation of a single syllable in a word, with greater force coming from the chest, and pausing the voice, (with no distinction between short and long syllables) we only need a single accent mark placed on the vowel with which we indicate and know that it is to be accented...In later Ortografías (1763-1826), for reasons that are unclear, 'stressed' was again equated with 'long': Acento en su sentido propio es el tono con que se pronuncia una diccion, ya subiendo ó ya baxando la voz; pero en la ortografía española vulgarmente entendemos por acento aquella nota ó señal con que se denotan las sílabas largas, porque las breves no se acentúan en castellano: y también llamamos acento la misma pronunciacion larga de alguna sílaba.(RAE 1763: 108) Accent as such is the stress with which a word is pronounced, raising or lowering the voice, but in Spanish spelling we commonly understand accent as the mark or sign with which long syllables are indicated, because the short ones are not accented in Castilian: and we also refer to the long pronunciation of a syllable as accent.
The Prontuario of 1854 states that the written accent indicates which syllable is pronounced with greater prominence: Se llama acento una rayita oblicua [...] que se pone, en los casos que se dirán, sobre la vocal de la sílaba en que carga la fuerza, segun la verdadera pronunciacion de la palabras.(RAE 1854:29-30) Accent is what we call a slanting line [...] which is placed, in the cases that will be enumerated, on the vowel of the syllable pronounced with the greatest force, according to the true pronunciation of the word.
Finally in the GRAE of 1883, accent was equated solely with stress, and length disappeared from the discussion: Se llama acento ortográfico, ó simplemente acento, una rayita oblicua (´) que [...] se pone, en los casos que se dirán, sobre la vocal de la sílaba donde carga la fuerza en la pronunciación del vocablo.(RAE 1883:364) A slanting line (´) placed, in the cases that will be enumerated, over the vowel of the syllable pronounced with the greatest force, is called orthographic accent, or simply accent.
The term acento de intensidad 'stress accent' first appeared in the Esbozo of 1973 and is still used today, along with acento prosódico 'prosodic accent'.The current rule is in spirit very similar to the one from 1883: Para señalar la sílaba tónica de una palabra, el español emplea en ciertos casos el acento gráfico llamado también tilde (´), signo colocado sobre la vocal de la sílaba tónica de la palabra según reglas bien establecidas.(RAE 1999:24) To indicate the tonic syllable in a word, in certain cases Spanish uses a graphic or written accent (´) over the vowel of the tonic syllable, according to well-established rules.

Vowels as separate words
The Diccionario of 1726 presented a rule for vowels used as separate words: ...el grave, que es el que baxa obliquamente de la izquierda à la derecha en esta forma `, unicamente puede tener uso sobre las quatro vocáles à è ò ù , quando cada una es como voz separada de otras, y hace cabál sentído por sí sola, dexando de ser mera vocál.(RAE 1726:LXIV) ...the grave accent, which slants downward from left to right is this way (ˋ), can only be used on the four vowels à è ò ù, when each one is used as a word separate from the others, has meaning in and of itself, and ceases to be simply a vowel.This use of the grave accent is illustrated above (de la izquierda à la derecha 'from left to right'), and in frases such as "en palabras de tres ò mas vocáles" 'in words with three or more vowels' (LXIV).
The Ortographia of 1741 eliminated the grave accent in favor of the acute, albeit with some inconsistency in the application of the rule (ò á Pedro): La misma nota, ó virgulilla se ha de poner sobre qualquiera de las vocales, a,e,o,u, quando hablasse por si sola, v.g.á Juan, ò á Pedro, ú á otro: no porque demuestre su acentuacion; sino porque no se pronuncie unida á la voz antecedente, ó subsiguiente.(RAE 1741:258) The same mark or dash should appear over the vowels, a,e,o,u when they stand alone, v.gr.á Juan 'to Juan', ò á Pedro 'or to Pedro', ú á otro 'or to another': not because it indicates their accentuation, but because they are pronounced separately from the preceding or following word.
In the Prontuario of 1854 the grammatical function of vowels used as separate words comes into play: Cuando a es preposicion, y la e, la o y la u son conjunciones, llevan siempre acento, v.gr.Ama á tu prójimo; diez ú once; ayer ú hoy.(RAE 1854:30) When a is a preposition, and e, o, and u are conjunctions, they always have an accent, v.gr.Ama á tu prójimo 'love your neighbor'; diez ú once 'ten or eleven'; ayer ú hoy 'yesterday or today'.
The GRAE of 1883 conceded that these vowels were accented out of custom rather than for prosodic reasons: La preposición á y las conjunciones é, ó, ú, se acentúan ortográficamente por costumbre y no por ninguna razón prosódica.(RAE 1883:366) The preposition á and the conjunctions é, ó, ú have a written accent mark out of custom and not for any prosodic reason.
The 1883 rule remained intact through 1906.The GRAE of 1917 argued that it was unfounded, with one exception: La preposición a y las conjunciones e, o, u, no llevando acento prosódico... tampoco deben llevarlo escrito.No obstante, lo llevará escrito la conjunción o cuando, por hallarse inmediata a cifras, pudiera confundirse con el cero; así, 3 ó 4 nunca podrá tomarse por 304.(RAE 1917:490) Since the preposition a and the conjunctions e,o,u do not have a prosodic accent...they should not have a written one either.However, the conjunction o will have a written accent when it immediately precedes a number, because it might be confused with the zero, so that 3 ó 4 can never be confused with 304.

The diacritical accent
The term tilde diacrítica appeared only recently, in the Ortografía of 1999: La tilde diacrítica es aquella que permite distinguir, por lo general, palabras pertenecientes a diferentes categorías gramaticales, que tienen, sin embargo, idéntica forma.(RAE 1999:27) The diacritic accent is the one that, in general, allows us to distinguish between words belonging to different grammatical categories, but that are identical in form.
The current concept of the diacritical accent evolved slowly.It is not mentioned in the Diccionario of 1726.The Ortografía of 1741 grouped de and el with all other monosyllables: En los monosylabos, ó voces de una sylaba no se debe poner nota de acento: porque nunca puede variarse la pronunciacion, v.g.dan, de, el, que, y otras semejantes.(RAE 1741:248) On monosyllables, or words with a single syllable, there should be no written accent mark: because their pronunciation can never vary, v.gr.dan 'they give', de 'of', el 'the', que 'that', and other similar ones.
The modern diacritical accent began to take shape in the Ortografía of 1763: Ningun monosílabo, ó voz apelativa ó propia de una sílaba, ya acabe en vocal, ya en consonante, se acentuará [...].Exceptúanse aquellos monosílabos que teniendo mas de una significacion, se pronuncian con mayor pausa en una que en otra [...] como él quando fuere pronombre, á diferencia de quando fuere artículo; [...] (RAE 1736:110-111) No monosyllable, or appellative or individual word of one syllable, whether it ends in a vowel or a consonant, will have a written accent [...] except those monosyllables which have more than one meaning, and where one is pronounced with greater pause than the other, for example él 'he' when it is a pronoun as opposed to an article; [...] The list also includes mí 'to or for me'/mi 'my', dé 'give (imperative or subjunctive)'/de 'of', sé "I know"/se 'third person pronoun' and sí 'yes'/si 'if'.The Prontuario of 1854 (32) added tú 'you (personal pronoun)' and más 'more', and included the adverbs sólo 'only' and aún 'still', interrogatives, exclamatives, and demonstratives.It also distinguished between the prepositions para 'for', entre 'between', and sobre 'above', and the verbs pára 'stop', éntre 'enter' and sóbre 'to be left over/plentiful', not mentioned before or since: Para, entre y sobre, preposiciones, no llevarán acento; pára, éntre y sóbre, verbos, pueden necesitarlo.Se escribirá pues: Cochero, pára el coche; para coches no es bueno este piso.Si hemos de elegir entre Juan y Pedro cuál ha de entrar en casa, éntre Juan.Sóbre el pan aunque falte el vino; esto importa sobre todo.(RAE 1854:35) The prepositions para, entre, and sobre will not have a written accent; the verbs pára, éntre, and sóbre may need one.So the following will be written: Cochero, pára el coche; para coches no es bueno este piso.'Driver, stop the car; this surface is not good for cars'.Si hemos de elegir entre Juan y Pedro cuál ha de entrar en casa, éntre Juan.'If we have to chose between Juan and Pedro who enters the house, let Juan enter'.Sóbre el pan aunque falte el vino; esto importa sobre todo.'Let the bread be plentiful even though there is not enough wine; this matters above all'.
The GRAE of 1917 established a connection between the grammatical function of monosyllables and prosody: ...se escribe el acento cuando existen dos monosílabos iguales en su forma, pero con distinta función gramatical en una de las cuales lleva acento prosódico y en otra es átona.(RAE 1917:490) the accent is written when there are two monsyllables with the same form but different grammatical function, one of which has a tonic accent and the other of which is atonic.
The Esbozo of 1973 reiterated the connection between forms with a written accent and prosody.Demonstrative pronouns (e.g.éste 'this one') and adjectives (este libro 'this book') are both considered prosódicamente acentuados 'prosodically stressed', but not so pairs like cómo 'how'/como 'as, like' or él/el, among others: En los casos restantes de dos acentuaciones, la diferencia se establece entre dos voces homófonas prosódicamente acentuada la una e inacentuada la otra.(RAE 1973:140) In the remaining cases of two accentuations, the difference is established between two homophones, one prosodically stressed and the other unstressed.
There is no mention of prosody in relation to this type of accent in 1999 and 2010.Now the diacritical accent is the only one that is truly only a written accent.Its very name indicates that it is not meant to reflect any difference in pronunciation.
Notably, there is no distinction between words with hiatus between a high and a nonhigh vowel (via, tia, lie, rio, mio, pua, rue, duo), and between two mid or low vowels (nao, cae, brea, sea, lee, reo, veo, loa, loe, roo).Under current rules, two adjacent mid or low vowels (e.g.re-o 'defendant/accused') each form a separate syllable.Words like rí-o 'river' require a written accent indicating hiatus, since an atonic high vowel adjacent to a mid or low vowel (e.g./i o/) forms a diphthong.Another unmotivated rule distinguished between simple and compound forms: Adviértese que hay algunas voces acabadas en dos vocales, que aunque en sus simples no necesitan de acento, deben llevarle los compuestos, v.g.lia y via que se componen solo de dos sílabas, no se acentúan [...]; y sin embargo sus compuestos deslía, desvía [...] se deberán acentúar en la penúltima, conforme á esta quarta Regla: y así otras voces semejantes.(RAE 1763:119) Notice that there are words ending in two vowels that, in their simple forms do not require a written accent, but that should have one in their compound forms, v.gr.lia 'bind (3 rd p. sg.pres.indic.)'and via 'road/way', which only have two syllables, do not have a written accent [...]; however, the compound forms deslía 'unbind (3 rd p. sg.pres.indic.)'and desvía 'detour (3 rd p. sg.pres.indic.)'[...] should have one on the penult according to this, the fourth rule; and likewise in similar words.

Accent associated with verb form
We know today that verbs do not have a written accent because of tense or mood, but because they break the Spanish stress rules.For example, comió ends in -o, and so should by rule be accented on the penult.Vino is also preterit, but follows the stress rules and does not have a written accent.However, the Diccionario of 1726 called for accents to distinguish between verb forms, and nouns and verbs: Fuera de estos fines tiene otro mui conveniente y oportúno, que es de evitar la equivocación en la pronunciación de algunos tiempos de los verbos, señalando quando es presente, ò preterito, quando es indicativo, ò subjuntivo, &e.como Enseño, Enseñó, Amára, Amará, Deseáre, The rules varied in each publication.The Ortographia of 1741 eliminated the above rule, instead requiring an accent on third person plural future verbs ending in -n, and making an exception to the general stress rules for plural verbs in -mos or -is: Las acabadas en qualquiera de las otras consonantes, se han observado generalmente de pronunciacion aguda, y por ser esta su mas comun, y freqüente acentuacion, solo necesitan de acento sobre la vocal de la penúltima sylaba, quando no fueren agudas, como en charácter, fácil, cánon; pero como en las terceras personas del plural de los verbos, que acaban en n, solo sean de acento agudo las de los futuros, á estas se les debe poner nota de acento en la última vocal, y no á las demás, como en amarán, enseñarán.(RAE 1741:252-253) Y se advierte que las voces que tienen nota de acento, la conservan [...] en los verbos, quando para formar el plural se les añade el mos, ó is, ó se les pospone unido alguno de los pronombres como en amarémos, amaréis, enseñóme, daréte, [...]; porque como estas adiciones no inmutan su pronunciacion, tampoco deben hacer novedad en las notas de sus acentos.(RAE 1741:254-255) Words ending in any of the other consonants are generally oxytone, and since this is their most common and frequent accentuation, they only need a written accent on the vowel of the penultimate syllable when they are not oxytone, as in charácter 'character', fácil 'easy', cánon 'canon'; but since in third person plural verbs that end in n only future tense verbs are oxytone, only these and no others will have a written accent on the final vowel, as in amarán 'love/ enseñarán 'teach' (3 rd p. pl.fut.indic.).
And be advised that words that have a written accent will maintain it [...] on the verb forms in the plural that are formed with -mos or -is, or to which is added one of the pronouns, as in amarémos 'love (1 st p. pl.fut.indic.),amaréis 'love (2 nd p. pl.fut.indic.)',enseñóme 'show to me (3 rd p. sg.pret.indic.),daréte 'give to you (1 st p. sg.fut.indic.)'[...]; because since these additions do not alter their pronunciation, they should have no effect on their written accent mark.
The rules were changed in 1763, and called for accents on first and third person sg.
The rule refers to verbs in the subjunctive, but the examples given are in the conditional.Similar rules persisted until the GRAE of 1883, in which there are no longer separate rules for verb forms.

Words ending in -n, -s and -V
The two current rules that account for almost all written accentuation not covered by the rules of hiatus and the diacritical accent group -n, -s and -V together: Las palabras agudas llevan tilde en la sílaba tónica cuando terminan en vocal, -n o -s.[...].Las palabras llanas llevan acento gráfico en la sílaba tónica cuando terminan en consonante que no sea -n o -s.(RAE 1999: 24-25) Oxytone words have a written accent on the stressed syllable when they end in a vowel, -n or -s.[...] Paroxytone words have a written accent on the stressed syllable when they end in any consonant other than -n or -s.' The only exceptions are words ending in -s preceeded by another consonant (e.g.robots, cómics), but these are all linguistic borrowings.A third rule falls out directly from the first two: Las palabras esdrújulas y sobresdrújulas siempre llevan tilde en la sílaba tónica.(RAE 1999:25) Words with stress on the antepenultimate and pre-antepenultimate syllable always have a written accent on the stressed syllable.
The generalization regarding the accentuation of multisyllabic words ending in any vowel was not captured until 1883.The Ortographia of 1741 distinguished between words ending in -a,-e,-o, and those ending in -i and -u: La voces acabadas en qualquiera de las vocales a, e, o tienen su mas freqüente, y comun acentuacion en la penúltima sylaba, y solo necesitan de nota de acento en la última vocal quando fueren agudas, como en allá, amará, cupé, amaré, amó, enseñó.A el contrario las acabadas en i latina, ó en u, tienen su acentuacion mas comun, y freqüente en la última vocal: por esso solo necesitan de nota de acento en la vocal de la penúltima sylaba, quando no fueren de terminacion aguda, como en cási, tríbu.(RAE 1741:251) Words that end in the vowels a, e, o are most frequently and commonly stressed on the penultimate syllable, and only require a written accent mark on the final vowels when they are oxytonic, as in allá 'there', amará 'love (3 rd p. sg.fut.indic)', cupé 'coupe', amaré 'love (1 st p. sg.fut.indic)', amó 'love (3 rd p. sg.pret.indic)', enseñó 'teach/show (3 rd p. sg.pret.indic)'.On the other hand, words that end in i or u are most frequently and commonly stressed on the final syllable, and so only require a written accent mark on the vowel of the penultimate syllable when they are not oxytonic, as in cási 'almost', tríbu 'tribe'.
The Ortografía of 1763 eliminated this distinction, but proposed ad hoc rules like the one below, in which words with three or more syllables ending in any two vowels have a written accent on whichever syllable is stressed: Las voces que fueren de tres ó mas sílabas y acabaren en dos vocales se acentuarán por su mucha variedad en qualquiera vocal, dondé cargare la pronunciacion.(RAE 1763:116) Due to their great variety, words with three or more syllables that end in two vowels will have a written accent on whichever vowel is stressed.

Other orthographic developments
As a point of interest, the inverted interrogative (¿) first appeared in the Ortografía of 1754: ...hay periodos largos en los quales no basta la Nota de Interrogante que se pone á lo último, para que se lean [las oraciones] con su perfecto sentido.[...] La dificultad ha consistido en la elección de Nota; pues emplear en esto las que sirven para los Acentos y otros usos, daría motivo á equivocaciones, y el inventar nueva Nota sería reparable, y quizas no bien admitido.Por esto, despues de un largo examen ha parecido a la Academia se puede usar la misma Nota de Interrogación, poniendola inversa antes de la palabra en que tiene principio el tono interrogante... (RAE 1754:126-127) ...there are long extensions of text in which the question mark placed at the end is not sufficient for [the sentences] to be read with their intended meaning.[...] The real problem was one of choosing the notation to be used; employing the notation used for accents and other purposes would lead to mistakes, and inventing a new notation might lead to modifications or be ill received.For this reason, after a great deal of study it occurred to the Academy that the same question mark can be used, inverting it in front of the word where the interrogative tone begins.

Conclusion
Compared to earlier versions, the current accentuation rules represent a streamlined and elegant system that constitutes a useful tool for the pronunciation and spelling of unfamiliar words.As seen above, in Modern Spanish written accents indicate the stressed syllable in words that do not follow the Spanish stress rules, distinguish between homophones with the same form but different grammatical function, and indicate hiatus between two contiguous high vowels or a contiguous high vowel and a mid or low vowel.Many of the current rules took shape in the GRAE of 1883, but significant modification and clarification have continued until very recently.To summarize, accent was equated with syllable length until the GRAE of 1883, and the term acento de intensidad 'stress accent' did not appear until the Esbozo of 1973.The written accent was maintained on vowels used as separate words until 1906.The GRAE of 1917 changed the rule, maintaining only the accentuation of ó between numbers, and the Ortografía of 2010 eliminated this exception.The diacritical accent was recognized as a purely written accent very recently, in the Ortografía of 1999, which first used the term tilde diacrítica.The GRAE of 1883, established the current distinction between diphthongs and hiatus, as well as the current accentuation rules.Clarifications continued to be made through the Ortografía of 1999, which defined a triphthong as a strong vowel between two weak vowels.Certain verb tenses were defined as requiring written accents until the GRAE of 1883, in which such rules disappeared.Finally, words in -ión were not uniformly written with an accent mark until the GRAE of 1883, and -n, -s and -V were not grouped together in a single rule until the Esbozo of 1973.Students very often question why the written accent came to be in Spanish.Instructors often do not know where to turn to find the answers.We expect an understanding of the origins and evolution of the written accent in Spanish, and of the current rules for written accentuation, to help clear up at least some of the questions about its inner workings.