use to include generic, collective, and abstract nouns, essentially becoming noun markers.

This usage is not confined to just a few languages, either, but is found in most, if not all,

Romance languages, major and minor. This thesis examines the question of how this

came to be, whether through diffusion from one language to all others, or through

independent parallel development. I first trace the history of definite articles in three

language from the late thirteenth century, and one in each language from around the year

1500. The results show definite articles were used more frequently than expected in the

earlier Spanish work, perhaps hinting at diffusion from Spain. Nevertheless, placing these

results in historical context, I argue that this use arose through independent parallel

development through the process that gave birth to definite articles in the first place -

insight she has given me over the many months I have been working on this thesis, as

well as how patient she has been throughout this process. I also must thank Dr. Danny

Law of the University of Texas-Austin whose comments at the Linguistics/TESOL

Symposium in February 2018 gave men the inspiration for this thesis.

LIST OF TABLES ................................................................................................................... iv

CHAPTER

CHAPTER Page

REFERENCES ...................................................................................................................... 58

Table Page

bit from their parent language Latin. One of these features is the article, both definite and

indefinite. Some, such as French and Italian, have also developed a partitive article, while

in others, such as Spanish and Portuguese, the lack of an article often serves the same

function as the partitive. While the use of articles differs somewhat from language to

Table 1: Comparison of same sentence in ten different Romance languages, Rudder (2012, p. 206) italics

are mine

Definite articles are normally used to specify a particular person or thing, but in the

sentences above there is nothing particular about the nouns at all, they are used in a

general sense. In every single one of the above languages, definite articles are used with

both nouns, even though they are used in a generic or abstract sense. Neither “language”

nor “technique” refers to a specific language or technique, but rather language and

technique in general.

There are three ways to explain this similarity in usage. It could have been

Another way is through diffusion by language contact. The use with generic nouns could

have developed in one language, and then spread to others. It is known that Romance

languages borrowed vocabulary from each other, and so may have borrowed syntactic

usages as well. The third way is through independent parallel development. The same

usage could have developed in each language independently, though how that may have

and time. The classic model of diffusion is the wave model. In this model, an innovation

starts at some place, and spreads outward to neighboring communities. It is often

compared to the spread of ripples in water (Wolfram & Schilling-Estes 2003). Modern

research has shown that this model is not the best, however, as time and distance are not

A newer model, based on the processes observed in the natural sciences, is the

gravity model, or hierarchical model. In this model, an innovation spreads from a center

of heavy population density to other high density centers, and from these centers to

regions of increasingly less density, until rural areas, last of all, adopt it (Wolfram 2003).

Several studies have shown this model to be a better fit to observed changes in time. For

example, Trudgill (2014) discusses the diffusion of th-fronting, or the change of the

interdental fricatives represented by -th- in English to the labiodental fricatives /f/ and /v/,

the Southwest and the Northeast coast. By 1980, it was found as far north as Glasgow,

though it was still absent in Liverpool and Wales (Trudgill 2014). This follows the

hierarchical model, as the innovation first spread to nearby larger cities, and then to

smaller cities and larger cities further away.

A weakness of the hierarchical model is that it only includes population and

Estes 2003).

The one thing required in any diffusion model is, of course, time. It takes time for

an innovation to spread throughout a population and in physical space. Many modern

studies use populations of different ages as a proxy for time, as younger speakers are

2.1 Demonstratives and Definite Articles

What exactly are definite articles, and how do they differ from demonstratives?

Definite articles and demonstratives are both part of a class of words known as

determiners, so called because they make the noun phrase they modify more specific.

Other classes of words can be classified as determiners, but this can change from

(Lyons 1977). This is known as deixis, which also includes temporal reference (Lyons

1995). English demonstratives are this and that, and the corresponding plural forms these

and those. This indicates something near the speaker, while that indicates something

further away (Lyons 1977). Other languages have different systems, including Latin,

which I will discuss in more detail in 2.2 below. The actual use of demonstratives in

category of definiteness (Lyons 1995). In general, definite articles refer to a noun

previously mentioned, or specified in some other way. The idea behind a definite article

is that both speaker and listener understand which noun specifically is being discussed,

without specifying it in space (Penny 1991). Both the speaker and listener understanding

detail in 5.2. As with demonstratives, the actual use of definite articles in individual

languages can differ considerably.

How does a demonstrative become a definite article, then? This occurs through

grammaticalization, and has been observed in many languages (Faingold 1993, Smith

2011). Among these languages are English and other Germanic languages, as well as the

Romance languages (Lyons 1977). As mentioned above, they are both determiners, and

so belong to the same larger class of words. They both specify nouns in some way, and

as definite (Lyons 1977). If a demonstrative expands its use to include more than just

deictic reference, it would take on the role of a definite article. If it were then to

eventually lose its deictic reference altogether, it would then become a definite article. As

it expands to other uses, it would naturally be used more and more frequently, while the

strictly deictic use could be reinforced, either with another word, or in some other way.

however, have a very rich system of determiners, sometimes all called demonstratives. In

addition to the true demonstratives, proximal, medial, and distal, there were also

demonstratives used for emphasis, identity, and anaphoric reference (Penny 1991). The

masculine nominative form of these demonstratives are listed in the following table:

Table 2: Latin demonstratives (Penny 1991)

the conversation or text (Penny 1991). These latter were most often used as third person

pronouns in Classical Latin, as third person pronouns often have the same type of

reference. The identity demonstratives, formed by adding -dem to the anaphoric

demonstratives, were also frequently used as pronouns. Other demonstratives could be

used as third person pronouns to differentiate when necessary (Wheelock 1992).

This.dat.s counsel.dat palm branch.acc give.1s.pr

I give a palm branch to this counsel

2) Nōn est locus istīs hominibus in hāc terrā

Valor.acc truly that.gen.s man.gen.s love.1pl.pr

Truly we love the valor of that man

4) Ipse signum suum et litterās suās

recognōvit

Himself seal.acc.s his own.acc.neu.s and letter.acc.pl his own.acc.f.s

recognize.3s.past

All.nom.pl same.neu.s feel.3pl.pr

All men feel the same

6) Id sōlum est cārum mihi

3s.neu.nom alone.neu.acc be.3s.pr dear.neu.acc 1s.dat

Several changes occurred in the Late Latin period. These changes took time,

naturally. Sylvia's Peregrinatio ad Loca Santa, written between 380 and 540, shows that

demonstratives were still largely used as such, though the use of certain demonstratives

had started to increase. There is still a wide variety of demonstratives used, as well. As

so to distinguish between these two uses, those demonstratives used in that capacity are

in italics, as well as the corresponding word in English. The following examples are all

from this text.

aquam

In 3.f.s.ab then day.s and in 3.f.s.ab hour.s that divert.3pl.past Persian.nom.pl

water.acc

The water which the Persians diverted

9) hii fontes quos uide

This.m.pl.nom fount.pl.nom which see.2s.pr

The fountains which you see

10) Sedens i eadem spelunca, quae in ipsa ecclesia est

In the above passages, there are five different demonstratives used. Faingold calls ea (see

1 above) and a similar form eo innovations (Faingold 1993). They are in fact ablative

forms of is (Wheelock 1992). Their use in this text may very well be an innovation, of

course. Please note that in some cases the case of the noun was difficult to determine, and

Forms of hic are used with some frequency in Sylvia's text, though they are the

least used of the demonstratives in this text (Faingold 1993), and later fall out of use

entirely. Their use as proximal demonstratives was replaced by iste. Those Romance

languages that kept the three-way distinction, such as Spanish, generally used ipse for the

medial demonstrative (Penny 1991).

Not only was hic lost, but anaphoric is was replaced by ille (Penny 1991). As (11)

and (12) below show, by the seventh and eight centuries, ille had expanded its use to the

each language individually.

Latin demonstratives. As the three languages to be studied all derived their definite article

from ille, and as iste has the same declension, the following table shows the nominative

and accusative forms of ille:

Table 3: Declension of ille in nominative and accusative cases (Wheelock 1992)

Ipse only differs in the neuter singular, which is ipsum (Wheelock 1992).

An inscription said to be from the seventh century shows the further development of

Latin demonstratives. The following is taken from this inscription. Words in parentheses

are reconstructed, as the edges of the rock have broken away

(cian)t illas cupas collige calas

and self.m.pl servant.m.pl in oath ask must.2s.pr and 2s.dat fraud no do.3pl

that.f.pl cask.pl collect stored

and that you should take an oath of the servants so that they do not deceive

you. Collect the barrels which are stored.

demonstratives are used more frequently than in more literary texts, and that they seem to

be used with “the anaphoric or the defining function of the Romance definite article”

(Pountain 2000, 17).

In the following passage from the eighth century, only forms of ille are used, and

its use has expanded beyond that found in (11). Its use is so frequent as to be a functional

and that.f.pl cape.pl and that.m.pl cloth.pl and that.f.s footwear of that.m.pl

tollhouses above named

and the capes and the pieces of cloth, and the footwear from the toll-houses mentioned

above

While (11) and (12) show that the use of ille has definitely expanded from (7-10), it is not

used with every noun, as shown in (13) below

13) illa media pars cleri qui seniores fuerint annis singulis accipiant cappas novas

that.f.s middle part clergy who elders be3.pl.sub year.pl single receive.3pl

cape.pl new.pl

Let the middle part of the clergy who are the elders receive the new cloaks

Most modern Romance languages derive their definite articles from ille, as the above

passages show. But not all of them did. As (11) shows, both ipse and ille were used, and a

few Romance languages derived their definite articles from ipse. These forms were found

in early Southern Italian dialects, Sardinian, and medieval Catalan (Rudder 2012), though

now they are only found in Sardinian and the Balearic and Costa Brava dialects of

genitive, dative, and ablative (Wheelock 1992). As Classical Latin developed into later

Vulgar Latin, certain phonetic changes caused a loss of distinction in the case system.

Specifically, the loss of final –m and the loss of distinction between long and short

vowels often caused what had been three distinct forms of a word to merge into one

use prepositions to indicate grammatical function (Price 1971). This removed the need

for the case system, while the phonetic changes removed the case distinctions. Not all

Romance languages lost case entirely. Old French had two cases, nominative and

oblique, but this distinction was only found in masculine nouns (Price 1971). The only

modern Romance language to retain case is Romanian, which has two productive cases,

nominative/accusative and dative/genitive, and a third case, the vocative, only used with

some nouns (Mallinson 1990).

covers from the middle of the ninth century to the end of the eleventh. There are few texts

from this time period. Later Old French covers the twelfth and thirteenth centuries, as

French started to become a literary language, while Middle French covers the fourteenth,

fifteenth, and sixteenth. These limits are somewhat arbitrary, of course, and certainly do

by Late Latin, the demonstrative ille was already losing its value as a demonstrative. To

reinforce its demonstrative use in Northern Gaul, it was common to use another word

with both ille and iste. This was in fact the pattern throughout the Latin-speaking world,

though each emerging language chose its own deictic particle. In Northern Gaul, this

particle was ecce, meaning lo! or behold! (Price 1971). In Old French, this had given rise

Table 4: Demonstratives in Old French (Price 1971)

4.1 Spanish Demonstratives

Spanish retained the three-way deictic contrast of Latin, with proximal, medial,

and distal demonstratives. As stated earlier, the old Latin proximal demonstrative hic was

lost, and replaced with the old medial demonstrative iste. In Old Castilian, and other

Iberian dialects, the former emphatic demonstrative ipse then filled the role of the medial

used in other areas, such as ecce, atque, and eccu. In the medieval period, accu can be

used with the other demonstratives, but is always used with ille (Penny 1991). The forms

found in early Castilian are given in the following table:

the masculine singular. By the end of the thirteenth century, the forms with -e become

dominant, and the forms without -e soon disappear entirely (Penny 1991).

In the Medieval period, the long and short forms were both used, as in the

following passages from the Poema de Mio Cid, from the early thirteenth century

22) Con aquestas mys duenas de quien so yo seruida

With these my ladies of who am I served

Together with these my ladies by whom I am served

23) Que aun con mis manos case estas mis fijas

That still with my hands marry.1s.subj these my daughters

used as needed to obtain the right number of syllables per line. Both lines do have

fourteen syllables, and this is done through the choice of the longer or shorter form.

The longer forms were used as late as the mid-sixteenth century, as shown in the

following passage from the play Eufemia, by Lope de Rueda

24) ¿Qué parte eres tú para pedirme aquesso, cortabolsas?

In Modern Spanish, the longer forms of the proximal and medial demonstratives are no

longer used, and the medial forms only have one -s- due to orthographic changes (Penny

1991).

4.2 Spanish Definite Articles

seems to be some debate as to the origin of the singular forms. The plural forms

developed from the accusative, and some claim the singular forms did as well, while

others claim they developed from the nominative (Penny 1991, Pharies 2007). Penny

(1991) admits that the expected masculine singular form derived from the accusative, elo,

never occurs. The following is from a land transaction involving the Abbot of San Millan

in the twelfth century

From early on, the masculine singular form is el, the form that would be expected from

nominative ille. The feminine singular would be the same whether it derived from

nominative illa or accusative illam.

The -ll- of ille reduced to -l-, though el was often written ell before a word

early on in forms with another vowel (Penny 1991). This led to the feminine definite

article having two very different forms, la and el (Plank 1984, Penny 1991, Pharies

2007). The latter of course corresponds to the masculine article, but historically is an

allomorph of the feminine article (Pharies 2007). It was originally used before all words

beginning with a vowel, as seen in (26) and (27) below, but after the sixteenth century

had become restricted to words beginning with a stressed /a/ (Penny 1991, Pharies 2007).

In Modern Spanish, feminine el is used only before nouns, and not adjectives, beginning

(He) took his foot from the stirrup

(27) que es del otra part

that is of-the other side

that is on the other side

The soul have.1s.pr lost, the body despised

I have lost my soul and despised my body

Castilian dialects (Pountain 2000). The use of la with all feminine nouns may be one

difference. The later restrictions on the use of el with feminine nouns may be due to

dialect mixing and leveling as Castilian spread further south during the Reconquista.

As stated above, el is no longer used in front of feminine adjectives, and only with

feminine nouns beginning with stressed /a/. This is seen in the following examples:

29) la alta voz y la áspera manera de las muchachas de la alta sierra

the high voice and the coarse manner of the girls from the high

the water

water

Spanish also developed a neuter article lo, used only in the singular, which

developed from the Latin neuter form illud (Pharies 2007). In Spanish, it is used to make

nouns, nor was it used with geographic place names, like rivers. By the sixteenth century,

definite articles were used much as they are today (Penny 1991).

4.3 Definite Articles with Prepositions and Possessives

Just as in French, certain prepositions contract with certain definite articles. The

Modern Spanish, and was generally true of Old Castilian, though some dialects, such as

that of La Rioja, seem to have allowed other contractions, as in (31) below

(31) firmemientre lo creo, enna su mercet fío

firmly it believe.1s.pr., in-the his/her mercy trust.1s.pr

Firmly I believe, in her mercy I trust

survived into Modern Spanish – the use of definite articles with possessives, and this was

not limited to La Rioja, as (32) below demonstrates

(32) Ya doña Ximena la mi mugier tan complida

Oh lady Ximena the my wife so excellent

“Oh Doña Jimena, my most excellent wife”

definite articles with possessives in poetry was dependent on metrical concerns (Pountain

2000).

In addition to the short possessive forms, mi, tu, and su, and their corresponding

plural forms, Spanish also developed longer forms, mío, tuyo, and suyo, along with

following the noun. The first and second person plural possessives were the same in

either position. The forms su and suyo covered third person singular and plural (Penny

1991). When used alone, they typically require the definite article. By the late Old

Castilian period, the use of the definite article with the short forms was lost, so that by the

beginning of the sixteenth century the construction was very rare (Penny 1991). The

longer forms could be used with a definite article or other determiner as in (33) below,

though the construction is very rare in Modern Spanish

considered separate languages today, such as Neapolitan, Sicilian, or Milanese

(Grandgent 1927, Rudder 2012). Modern Standard Italian developed from the literary

Italian of the Renaissance, which was largely based on the Tuscan dialect, with some

influence from other dialects (Migliorini 1960/1966). In the following sections, therefore,

Southern and central Italian dialects continued the three-way distinction of Latin

demonstratives, while northern dialects followed the French pattern of reducing the

system to two. The particle ecco, derived itself from Latin ecce, meaning “behold”, was

used to reinforce the demonstratives, and was usually used with all demonstratives.

Following the pattern in Iberia, the more southerly dialects replaced Latin hic with iste,

Table 8: Demonstratives in Tuscan dialect, and Standard Italian (Grandgent 1927, Maiden 1995)

The medial demonstrative was derived by placing the second person pronoun te in front

of iste. The particle ecco was placed in front of that, giving cotesto or codesto, as seen in

the above table. In the modern language, the medial demonstrative has fallen out of use,

and is considered archaic (Maiden 1995). A form of the proximal demonstrative that was

not reinforced used to exist (sto, sta, etc.), and is still used in set phrases such as stanotte

and stasera, meaning tonight and this evening (Grandgent 1927). See 5.2 for the

differences in the distal masculine forms. Also note that the singular forms derive from

seen in (34) and (35) below, both from a thirteenth century poem.

(34) Di questo mar

Of this.m.s sea

“of this sea”

5.2 Italian Definite Articles

According to Grandgent (1927), the modern definite articles in Italian developed

from early ello, ella, elli, and elle, attested mostly after certain prepositions. As the

a vowel, they could all reduce to l'. In the central dialects, including Tuscan, a second set

of masculine articles developed from ello and elli. Some used el for the singular and ei

for the plural; in Tuscan, it was il and i. As the distal demonstratives were reinforced

forms of the definite article, multiple forms also developed in the masculine singular and

plural. The forms of the definite article in Tuscan were thus

altogether by the nineteenth century (Migliorini 1960/1966, Maiden 1995). The form gli

developed from li in front of a vowel, and is pronounced /ʎ/. It is now the plural form of

lo, and can be used in front of a consonant, where it is pronounced /ʎi/ (Maiden 1995).

Originally, il and i were only used after a word ending in a vowel, and before a word

beginning with a consonant. Even then, they were not used before words beginning with

sC (Grandgent 1927, Maiden 1995). They were not syllabic (Grandgent 1927), and even

in the modern language il is frequently pronounced 'l following a vowel (Maiden 1995).

Undoubtedly one form would be chosen over the other for that purpose in poetry, as with

Spanish demonstratives.

As early as the late thirteenth century, the use of il and i spread. They came to be

two hundred years later than the earlier works, and were therefore likely show some

change in the use of definite articles. The specific works were chosen due to ease of

access. They were either in the ASU Library, or available from an electronic source.

As stated in 1.3, the purpose of this study was to determine if the current use of

the definite article in modern Romance languages might be due to diffusion. If the

percentages for the total number of nouns with a definite article are similar in each time

period, it is safe to say it is due to independent development through grammaticalization,

article or other determiner. Certain other words, especially quantifiers, normally preclude

the use of articles as well, and so these were not counted. This includes nouns modified

by numbers. Occasionally, a noun modified by a quantifier was used with a definite

article, especially in Bembo's work. These were counted, otherwise, they were not.

Sometimes, definite articles were used in phrases that were more pronominal than

these cases, I did not count the words used with a definite article.

I photocopied pages from the texts or printed them from an electronic source, and

then marked the nouns used with a definite article in green, and those used without any

article or other word that would have blocked the use of an article in pink. I then counted

work. I then divided the number of nouns with a definite article by the total number of

nouns, both those with a definite article and those that were bare, to find the percentage

for each work.

It is important to note that the number of bare nouns does include indefinite

nouns, or nouns that today would use a partitive article, not a definite article. Normally in

such studies, the number of actual constructions involving a certain usage is compared to

the total number of constructions where said usage would be possible. In this study, the

decided it was better to use the same method of comparison across all three languages

than to use a method which could only be imperfectly applied to some.

This is also the reason why the study focuses on all nouns, rather than the

collective and abstract nouns discussed in the introduction. Determining whether a noun

is collective or not, or even abstract or not, involves judgments I did not feel I was

nouns with a definite article increases significantly, it is safe to assume that more types of

nouns are taking definite articles, even if indefinite nouns are included in the total.

I did not take the partitive article into consideration, either, as only the later

French text showed any use at all of a partitive article, and even then only sporadically.

1971), I counted the few uses in that text as definite articles.

I did not count proper nouns, such as names of people or places. While names of

places may be accompanied by a definite article, especially in the modern language, it is

not a hard and fast rule in all three, and I decided it would be best to not include them.

The various words for “God” were also not included where they obviously referred to the

Christian God as an individual, rather than a concept.

I also counted uses of nouns with a preposition and nouns modified by an

some overlap between the two totals.

Not counted in any of the above categories are nouns used with possessives. I

counted nouns used with possessives and definite article, and those used without. If some

other article or determiner was used, the noun was not counted. As with the other nouns,

the total was written on each page and then added up to obtain the total for each work. I

and I wanted to compare them specifically cross-linguistically and over time.

The part of the later Spanish work that I analyzed contained a long list of nouns,

which I removed from the final count. This list alone, which was less than a single page,

contained over one hundred nouns, or more than a tenth of the total. Nearly all of them

whole. The original numbers as well as the revised numbers are given below. The later

French work also contained a shorter list of nouns, which I also removed from the final

count, for similar reasons. Again, both the original and revised numbers are given below.

6.3 Data

Following are several tables with the counts for each work.

7.1 Totals

While the early French and Italian totals are close, the early Spanish totals are

noticeably higher, at 46 percent, compared to 35 and 31 percent for French and Italian

respectively. It is in fact the same as the later total for this language. Other than this being

indicative of Castilian at the time, why might this be? As noted in section 6.1, the Poema

poems used more abstract nouns than the Spanish poem did, as they dealt with more

abstract subjects. The Poema did use collective nouns, as did the French Roman de la

have had an effect on the total percentage of nouns with a definite article, as at this time

period collective and abstract nouns rarely had articles.

but might also explain why there are more definite articles. Just as longer and shorter

versions of the demonstratives could be used for metrical reasons, so could definite

articles with possessives, as noted by Pountain (2000). And if definite articles were used

with possessives for metrical reasons, they could easily have been used without a

hard to determine, as it is also possible the other languages could have done the same

thing.

Another possible effect could be personal style. Each writer uses the language a

little differently, especially in regards to optional features. As the use of the definite

article was optional with certain types of nouns, such as collective nouns or abstract

nouns, it is possible that one writer might simply use the definite article more often than

another. Register could also play a role, as a higher register uses more formal language,

different subjects. This would also account for different types of nouns being used in

different works.

As for the later texts, the French Épitres de l'Amant Vert shows a significantly

lower total, at 40 percent, compared to 46 and 49 percent for Spanish and Italian

respectively. While this represents an increase from the Roman, it is nothing comparable

Italian Asolani is a prose work. Poetry tends to be a little looser in its syntactic structure

than prose, especially when the metric structure is strict. It is not impossible that poetry

might also be looser in its use of definite articles in Romance languages for the same

reasons. Again, looking at different texts by different authors in different types of

regarding the use of definite articles.

In all three later texts, and occasionally in the earlier ones, I noticed several cases

where the first noun of a series would have a definite article, but not the following nouns,

though they appeared to be definite as well. Huchon (2005) notes that this was common

in French as late as the sixteenth century, even where the gender of the nouns was

different, and was apparently also true of Spanish and Italian as well.

7.2 Prepositional Phrases

prepositional phrase with a definite article than the overall totals, at 32 percent for the

earlier Roman and 34 percent for the later Épitres, compared to 35 percent and 40 percent

for the totals respectively. This indicates that nouns in a prepositional phrase are

somewhat less likely to have a definite article than nouns generally.

The same cannot be said for Spanish and Italian, however. The earlier texts show

Celestina shows a rate of 54 percent, noticeably higher than the 46 percent overall total,

and the Asolani shows a rate of 51 percent, compared to an overall total of 49 percent.

Please note that the original count of the Celestina shows a rate of 45 percent compared

to an overall of 41 percent, a difference more in line with the Asolani, though the

phrases are no less likely to have a definite article than nouns generally. This could have

to do with the fact that Old French used the preposition de as a partitive marker (Price

1971), unlike Spanish and Italian. The true partitive article is still rare in the later Épitres,

which could explain its low rate as well.

7.3 Modified Nouns

Modifying a noun can serve a similar function to determiners. “Little green

apples” may not be less generic than “apples”, but the phrase is certainly more specific,

In the earlier texts, it does appear to make a noticeable difference. In the Roman,

the percentage is 46, compared to 35 percent overall, in the Poema, it is 53 percent

compared to 46 percent, and in the Fiore, it is 40 percent compared to 31 percent. It

certainly appears that modified nouns are more likely to have a definite article than nouns

overall, even in Spanish, where the overall percentage is already fairly high.

earlier texts, though the numbers are higher. In the Épitres, modified nouns with a

determiner account for 48 percent of the total, while the overall total is 40 percent. In the

Asolani, the rate is 52 percent compared to the 49 percent overall total, showing that in

both languages modified nouns were more likely to have a definite article, especially

Bembo, P. (1808). “Gli Asolani” [PDF file]. Minetti, P. (Ed.) Opere del Cardinale Pietro

(Original work published 1505) Retrieved from:

https://babel.hathitrust.org/cgi/pt?id=nyp.33433022653798;view=1up;seq=15

De Lorris, G. & J. De Meun. (1965). Le Roman de la Rose. Vol. 3. New York: Johnson

Reprint Corporation.

De Rojas, F. (1995). La Celestina. M.E. Lacarra (Ed.) Madison: Hispanic Seminary of

Poema de Fernán González. (n.d.) Biblioteca Saavedra Fajardo de Pensamiento Político

Hispanico. Retrieved from:

http://www.saavedrafajardo.org/Archivos/LIBROS/Libro0206.pdf

Faingold, E.D. (1993). “The development of the definite article from Latin to Spanish

and Portuguese.”

Grandgent, C.H. (1927). From Latin to Italian. Cambridge, MA: Harvard University

Press.

Halm, T. (2018). “From possessive suffix to affective demonstrative suffix in Hungarian:

A grammaticalization analysis.” Morphology 28:359-396. Retrieved from:

https://link-springer-com.ezproxy1.lib.asu.edu/article/10.1007/s11525-018-9332-4

Heap, D. (2006). “Dialect Chains.” in K. Brown (Ed.) Encyclopedia of Language and

Linguistics 2

language-and- linguistics#book-info

Heine, B. (2003). Grammaticalization. in B. D. Joseph and R. D. Janda (Eds.) The

Publishing.

Huchon, M. (2002). Histoire de la langue française. Paris: Librairie Générale Française.

Lacarra, M.E. (1995). “Introducción.” in M.E. Lacarra (Ed.) La Celestina. Madison:

Hispanic Seminary of Medieval Studies.

Latini, B. (1951). “Il Tesoretto.” in G. Petronio (Ed.) Poemetti del 200. Torino: Tipografia

Temporelli. (Original work written 1260's)

Llorach, E.A. (1982). “Prólogo.” in E.A. Llorach (Ed.) Poema de Fernán González.

Madrid: Unigraf.

Mallinson, G. (1990). “Rumanian.” in B. Comrie (Ed.) The World's Major Languages.

New York: Oxford University Press.

Migliorini, B. (1966). The Italian Language. (T.G. Griffith, Trans.) New York: Barnes