Changing the Focus: an Empirical Study of Focalizing Ser ('to Be') in Dominican Spanish *

This article examines the perception and production of the Focalizing Ser (FS) structure by Dominican speakers living in three urban communities: Santiago de los Caballeros, Sosúa, and Santo Domingo de Guzmán. FS has received increasing attention over the past years and several studies have investigated it from various linguistic perspectives in Spanish varieties where it is typically found (e.g. Venezuelan Spanish and Colombian Spanish). Although previously reported in Dominican Spanish (Toribio 2002, Alba 2004), FS has still not been extensively analyzed in this variety. The results of three different tests (two acceptability judgment tests and one semi-production test) indicate that FS is certainly entrenched in Dominican speakers' linguistic inventory and that certain focus types (e.g. FS-focused subject and object DPs, FS-focused CPs and IPs, FS-focused complex VPs, and FS-focused prepositional phrases) are more favorably perceived and more frequently produced across all FS dialects.

FS has been generally identified as a dialectally -and even sociallymarked phenomenon.In Caracas, for example, FS has been described as a result

Methodology
The data presented in this study was collected in the summer of 2013 in three Dominican cities: Santiago de los Caballeros, Sosúa, and Santo Domingo de Guzmán.The city of Santiago is located in the north-central region of the country, in the Cibao Valley.As the capital of the largest province in the region, Santiago is home to almost 600,000 inhabitants. 6The city of Sosúa is also located in the Cibao Valley, but it belongs to the Province of Puerto Plata.Resting on the 3 Toribio (2002Toribio ( , 1992)), for example, argues for a unified syntactic analysis of FS and PC by proposing that the only difference between these two structures is that the relative clause is elided in the former.

4
A detailed description of this syntactic configuration can be found in Méndez Vallejo (2014Vallejo ( , 2012Vallejo ( , 2009)).Providing a more in-depth syntactic explanation escapes the scope of this paper.

5
A community of linguists (M.Carmen Parafita Cuoto, Manuel Delicado-Cantero, Liz Castro, among others) is currently advancing data collection in several FScommunities, hoping to elucidate the uses of FS within and across dialects.country's northern coastline, this city, of almost 30,000 inhabitants, 7 is a popular touristic place and has been increasingly populated with European and North American migrants.Finally, the city of Santo Domingo is the capital and the center of the national government of the Dominican Republic.Located in the southern coastline, the Santo Domingo Province (including the National District), inhabits almost 3,000.000residents in its metropolitan area. 8

Participants
As shown in Table 1, a total of 123 informants participated in this study: 83 of them from Santiago, 31 from Santo Domingo, and 9 from Sosúa.All of these participants were born and raised in their respective cities.

Table 1: Total number of participants in this study
Most of the participants from Santiago (73/83) belonged to the university community at Pontificia Universidad Católica Madre y Maestra (PUCMM).The majority of them were undergraduate students, but some of them worked as staff or faculty members.The other participants from Santiago (10/83) were residents of the city and worked in local businesses or offices.
Similarly, the participants from Santo Domingo were university students or residents of the city who worked as professionals in their field.In fact, most of these informants (28/31) were undergraduate students at Universidad Acción Pro-Educación y Cultura (APEC).
Finally, although the participants from Sosúa were also part of the working class in their community, none of them had continued their education after finishing high school.Most of them (7/9) had only finished primary school and, at the time of the data collection, they had temporary employments (mostly as motocab drivers).

Data collection procedures
The data analyzed in the present study originates from three different sources: an acceptability judgment test in audio format, an acceptability judgment test in written format, and a semi-production test.As illustrated in Table 2, participants from Santiago completed all tests (the acceptability judgment tests and the semiproduction test 9 ), participants from Santo Domingo only completed the Changing the focus: An empirical study of Isogloss 2015, 1/1 Focalizing ser ('to be') in Dominican Spanish 71 acceptability judgment tests (both in audio and written format), and the participants from Sosúa only completed the written acceptability judgment test.

Acceptability judgment test in audio format
This acceptability judgment test was created using an identification experiment in a computer program (Praat 5.0.36 10 ).The test was designed so that participants could hear a pre-recorded mini-dialogue and make a selection on a computer screen.After a brief tutorial, participants were instructed to rate the last sentence they heard based on an acceptability judgment scale from 1 (the least natural/familiar) to 5 (the most natural/familiar).After participants made their selection, the program automatically saved their answers and played the next mini-dialogue.A total of 93 sentences were included in the test, but only 61 of them were cases of FS-focus.A wide variety of cases of FS-focus was examined in this test: 12 sentences with FS-focused subject and object DPs, 3 with FS-focused PPs, 7 with FS-focused AdvPs, 6 with FS focus in impersonal and passive voice constructions, 9 with FS-focused complex DPs and VPs, 5 with FS focus in negative and indefinite constructions, 6 with FS focus in questions and constructions with wh-islands, 3 with sentence-final FS focus, and 10 with FS focus in other types of constructions (various tenses, clitic climbing, and individual/stage predicates). 11 The sentences that we tested in this acceptability judgment task were also used in a previous study (Méndez Vallejo 2009).In order to compare the results obtained in the Dominican Republic with those obtained previously in Colombia, we decided to maintain the same sentences and the way in which they were grouped. 12The only three sentences that differ between these two corpora are the ones that include sentence-final FS focus (e.g.No, llovió en la sierra fue 'No, it was in the mountains where it rained').We included these three sentences to explore whether or not Dominican speakers found sentence-final FS focus acceptable.

Acceptability judgment test in written format
While the test in audio format presented participants with a pre-recorded minidialogue, the written test included both the mini-dialogue and the acceptability judgment scale in a written questionnaire.These participants also completed a brief tutorial and were instructed to make their selections based on how familiar 10 Retrieved from: http://www.praat.org/(Boersma. Paul and Weenink, David 2008) 11 See Appendix A for a complete list of sentences tested.
12 See Méndez Vallejo (2009) for more information regarding the methodological procedures and the theoretical reasons to select specific sentences for testing.
(common) or unfamiliar (uncommon) the last sentences in the mini-dialogues seemed to them.
A total of 35 sentences were included in this test, but only 25 of them were cases of FS-focus.The written test comprised different sentences from those evaluated in the auditory test, and different cases of FS-focus were incorporated: 5 cases of FS-focused IPs and CPs, 6 cases of FS-focused subject DPs (4 of which with varying agreement patterns), 4 cases of FS focus with emphatic adverbs, 8 cases of FS-focused object DPs (4 of which with varying scope and word order patterns), and 2 cases of FS focus in other types of constructions (impersonal constructions with haber ('there is/are') and sentence-final FS). 13  A subset of these FS sentences (12/25) had also been tested in Colombia in a previous study (Méndez Vallejo 2009).As in the acceptability judgment test in audio format, we decided to include these 12 sentences in the written test in order to facilitate future comparisons between dialects.The remaining 13 sentences were included here to examine the effects of word order, scope, and agreement on the acceptability rates of FS focus.Thus, we used 4 sentences to test word order and agreement (see category "Focus subject DPs and agreement patterns" in Appendix B), 4 sentences to test word order and emphatic adverbs like solo ('only') (see category "with emphatic adverbs" in Appendix B), 4 sentences to test the effects of word order and scope (see category "scope" in Appendix B), and 1 sentence to test sentence-final FS focus (see category "sentence-final" in Appendix B).
The decision to include these new sentences in the written test originated from observations made in previous studies, according to which word order and scope may influence the way in which FS is used in a given sentence (Méndez Vallejo 2014, 2009).For example, it was previously hypothesized that adverbs like solo ('only') may be located at a lower syntactic position than FS (i.e.FocusP > Spec vP), thus rendering more acceptable judgments when following FS ser.

Semi-production test 14
In this test, participants were shown a series of pictures from the story Frog, where are you? 15on a computer screen.After seeing each picture, participants were asked confirmation questions and their answers were voice recorded.The objective of this test was to record participants producing the FS construction in various contexts (e.g.FS-focused subject and object DPs, PPs, AdjPs, as well as word order alternations of such FS-focused phrases).Similarly to the other two tests, participants completed a brief tutorial before starting the task and they became acquainted with the characters of the story and the story-line.
The questions were formulated carefully in an effort to promote the production of the FS structure.For example, when participants saw a picture like the one illustrated in Figure 1, participants were asked questions such as, ¿El perrito está en el piso?(No focus: 'Is the dog on the floor?'),¿El perrito está es en el piso?[FS: 'Is it on the floor where the dog is?'], ¿Donde está el perrito es en el piso?[PC: 'Is it on the floor where the dog is?'].In turn, participants' 13 See Appendix B for a complete list of sentences tested.14 Thanks to Prof. María Luisa Zubizarreta for discussing with me methodological issues regarding the relationship between FS and scope/word order, and for revising earlier versions of the semi-production test.Mayer (1969).
Changing the focus: An empirical study of Isogloss 2015, 1/1 Focalizing ser ('to be') in Dominican Spanish 73 answers followed a specific pattern: they either confirmed or rejected the statement, and if they rejected it, they offered a corrected version of the events.For example, in the case of Figure 1, participants were expected to utter sentences such as, No, el perrito está en la cama (No focus: 'No, the dog is on the bed'), No, el perrito está es en la cama [FS: 'No, it is on the bed that the dog is'], No, donde está el perrito es en la cama [PC: 'No, it is on the bed that the dog is'].It is important to mention at this point that although FS may be a common structure in oral language, it is quite difficult to capture its uses by applying traditional methodological procedures, such as sociolinguistic interviews.When surveying the occurrences of FS in a wide arrange of syntactic structures, acceptability judgment tests have proved to be a more efficient tool (e.g.Méndez Vallejo 2012).Furthermore, given that FS depends on discourse-related factors (contrast, emphasis, discourse reference, etc.), the semi-production task provides us with a context that promotes the use of this form in the most natural way possible.In the following section, we will describe the results of these three tests.

Results
The results of the acceptability judgment tests and the semi-production test confirm that FS is a syntactic variety which is certainly present in Dominican speakers' linguistic inventory.Despite the fact that some participants anecdotally reported perceiving FS as a foreign form, their responses in the acceptability judgment tests (and even in the semi-production test) indicate that FS is a familiar and common form to them.
The results of each task are presented separately in the following subsections.As explained in the previous section, each task had a different methodology and participants only completed one of the three tasks (with the exception of four participants from Santiago who completed both the acceptability judgment test in audio format and the semi-production test).Given this, it is not our intention to make comparisons across tests.Furthermore, although we have included some statistical data, it is important to note that the main objective is to explore the ways in which FS is perceived and produced in the three Dominican communities.

Acceptability judgment test in audio format
When looking at the results of this test for all participants (73 in total), it is evident that FS-focused PPs and FS-focused Complex VPs are the two types of sentences that are evaluated as most acceptable.On the contrary, FS constructions in negative or indefinite expressions and in complex DPs present the lowest ratings.
More specifically, Graph 1 shows the mean scores given by the 73 participants for the various categories of FS-structures.The error bars indicate 95% confidence intervals for these mean scores.For example, there is a 95% probability that the population mean of the acceptability rating for the FS-focused PPs presented in the acceptability judgment test lies between 3.04 and 3.44. 16 Graph 1: Mean scores for FS-focus categories given by all participants in the first test Graph 2 below indicates the average ratings for all categories of FSsentences in each community.Again, speakers from both Santiago and Santo 16 Given that the acceptability judgment scale ranges from 1 (least familiar/common) to 5 (most familiar/common), averages above 3.0 are here considered as favorable (FS is favorably accepted), whereas averages below 3.0 are viewed as unfavorable (FS is not accepted or considered ill-formed).
Changing the focus: An empirical study of Isogloss 2015, 1/1 Focalizing ser ('to be') in Dominican Spanish 75 Domingo favor the use of FS-focused prepositional phrases (PPs) and FS focus in complex verbal phrases (VPs).On the other hand, FS in complex determiner phrases (DPs) and in negative and indefinite expressions received the lowest acceptability ratings: Graph 2: Summary of average ratings of FS by focus type in the first test Although the data in Graph 2 presents some interesting tendencies in these Dominican communities, it is important to consider that this is just a general snapshot of FS perception.A more detailed view of the results (see Interestingly, the only two categories from which there are no favorable ratings for FS focus are precisely the same ones that showed the lowest ratings in Graph 2 (complex determiner phrases (DPs) and negative and indefinite expressions).In fact, previous work (Méndez Vallejo 2012) points towards a similar tendency in Colombian Spanish: whenever FS occurs within a complex DP, e.g.A Marcelino le gusta la música es rock ('It is rock music that Marcelino likes'), or in conjunction with negative polarity items, e.g.Pobre Pachito no comió fue nada ('It was nothing that poor Pachito ate'), the sentence is rendered unacceptable.Going into a deeper analysis of these cases escapes the scope of this paper, but it is worth mentioning that some of the sentences tested in these categories received better ratings than others, possibly because of semantic and pragmatic reasons (e.g.different types of predicates, more favorable preceding contexts, word order, etc.). 17 On the other hand, the sentences with the highest ratings (range: 3.9 -4.0) correspond to categories that did not particularly display a high acceptability rating in Graph 2 (e.g.FS-focused subject DPs and object DPs, FS-focused AdvPs, sentence-final FS focus).Again, these results resemble those obtained in Colombian Spanish (Méndez Vallejo 2012) where FS-focused subject DPs, object DPs, and AdvPs were overall rated as acceptable.
Importantly, FS-focused subjects are quite acceptable in both Santiago and See Appendix A for a complete list of sentences tested and their acceptability ratings.
Changing the focus: An empirical study of Isogloss 2015, 1/1 Focalizing ser ('to be') in Dominican Spanish 77 Santo Domingo, which stands in stark contrast from what has been reported in other studies (Toribio 1992(Toribio , 2002)).In fact, the addition of sentence-final FS in this acceptability judgment test was prompted by reports of such cases in previous investigations of Dominican Spanish (Toribio 1992, Alba 2004).As shown in Table 3, at least one of the three cases of sentence-final FS evaluated was rendered as highly acceptable by our participants.
Coincidentally, the other two cases of sentence-final FS included FSfocused subject and object DPs, and were in present tense as in Trabaja Rocío es ('It is Rocío who works'), Les traigo esto es ('It is this that I bring to you').Although similar sentences have not been tested in other Spanish varieties (e.g.Colombian Spanish), we predict that they would not be as acceptable as they are in these Dominican communities.

Acceptability judgment test in written format
When looking at the results of the written test for all participants (41 in total), it is evident that FS-focused IPs and CPs, and FS-focused object DPs (with various agreement patterns) are some of the types of sentences that are evaluated as most acceptable.On the contrary, FS-focused subject DPs present the lowest ratings.
More specifically, Graph 3 shows the mean scores given by the 41 participants for the various categories of FS-structures.The error bars indicate 95% confidence intervals for these mean scores.For example, there is a 95% probability that the population mean of the acceptability rating for the FS-focused IPs and CPs presented in the written acceptability judgment test lies between 3.21 and 3.62.
Graph 3: Mean scores for FS-focus categories given by all participants in the second test In Graph 4 below we show that speakers from each community (Santiago, Sosúa, and Santo Domingo) tend to favor the use of FS when it focuses embedded CPs and IPs, when it occurs with impersonal verbs such as haber ('there is/are'), when it occurs sentence-finally, and when it focuses object DPs (and there is alternation in agreement and morphological coordination).On the other hand, FSfocused subject DPs (including those with alternating agreement patterns), and sentences in which FS occurs with emphatic adverbs and with varying scope (and word order) receive lower acceptability ratings: Graph 4: Summary of average ratings of FS by focus type in the second test The results outlined in Graph 4 also indicate an interesting tendency by Sosúa speakers to evaluate sentences at a much higher rate than the speakers from the other two groups.For example, FS-focused subject DPs presenting different agreement patterns were evaluated by Sosúa speakers as highly acceptable (3.1), whereas the speakers from Santiago and Santo Domingo rendered them as unacceptable (2.2 and 1.8, respectively).Conversely, the speakers from Santo Domingo appear to be more conservative with their ratings than speakers from the other two groups, given that they gave the lowest scores to five of the eight categories.Speakers from Santiago, however, seem more consistent with their judgments: they are not as generous as the speakers from Sosúa or as strict as the speakers from Santo Domingo.
Although this study cannot offer an in-depth explanation of these particular tendencies, we hypothesize that a possible cause for this significant disparity between the Sosúa and the Santo Domingo groups may be related to the format of the test.As mentioned in section 2.1, the majority of the participants from Sosúa had minimal formal education, in comparison to those from Santiago and Santo Domingo.In fact, many of the participants from Sosúa had trouble reading the mini-dialogues and they required help writing their answers in the written questionnaire.Thus, it is possible Changing the focus: An empirical study of Isogloss 2015, 1/1 Focalizing ser ('to be') in Dominican Spanish 79 that since many of them heard the mini-dialogues (we had to read them because they could not do it themselves), the sentences sounded much more acceptable to them that way than if they had seen them in writing.
Despite the propensity of Sosúa speakers to accept FS sentences at a higher rate than the other speakers, it is clear that some focus types are more restricted than others (e.g.FS-focused IPs and CPs seem more acceptable overall than FS-focused subject DPs).As shown in Sentences with embedded FS-focused IPs and CPs, for example, generally receive favorable ratings from all speakers (range: 2.8 -4.1), which is consistent with similar ratings provided for the same sentences by Colombian speakers in previous studies (Méndez Vallejo 2012).Similarly, cases of sentence-final FS and of FS in sentences containing haber ('there is/are') are also rendered as acceptable by Dominican speakers.However, these results should be viewed with certain reservation given that only one case was tested in each of these categories.
terms of FS-focused object DPs, it is important to point out that most cases (no matter what agreement pattern was established between FS ser and the focused object) were evaluated as acceptable (range: 2.6 -4.6).Looking at all the sentences tested in this category, it appears that the sentence Tenemos son un gato y dos perros ('It is a cat and two dogs that we have') may have received the lowest acceptability rating due to the fact that FS ser ('be') is in the plural form (son: ser-3PL) and the most adjacent item in the focused object DP does not match this morphological feature (un gato: 3SG).Incidentally, although these same sentences received a much lower acceptability rating by Colombian speakers (range: 1.8 -2.8) 18 , the sentences with the highest and the lowest acceptability scores in the Colombian data are the same ones that received such scores in the Dominican data.
As for FS-focused subject DPs, the results of this test generally indicate that Dominican speakers do not favorably accept this type of FS focus.Nonetheless, it is possible to see a slight preference for cases in which FS ser ('be') agrees in person and number with the focused pronoun nosotros ('we'): En la fiesta cocinamos fuimos nosotros ('It was we who cooked at the party'), Sacamos la basura fuimos nosotros ('It was we who took the garbage out').Just as in the case of number agreement with FS-focused object DPs, it is possible that this form is more acceptable when FS ser ('be') presents person and number agreement with the focused element.However, it is noteworthy that a similar case in which there is such person and number agreement (Tengo un perro soy yo 'It is I who has a dog'), does not render similar acceptable ratings.Interestingly, these same sentences, when tested among Colombian speakers 19 , showed a slightly different tendency: a sentence such as Tengo un perro soy yo ('It is I who has a dog') was rated as more acceptable than a sentence such as En la fiesta cocinamos fuimos nosotros ('It was we who cooked at the party').
The relationship between FS and emphatic adverbs, such as sólo ('only') and pero ('rather') 20 indicates a remarkable tendency: when these adverbs follow 18 See Méndez Vallejo (2012) 19 Ibid. 20 As pointed out by a reviewer, it is important to mention here that pero ('rather') may not be used as an emphatic adverb in some varieties of Spanish.To my Changing the focus: An empirical study of Isogloss 2015, 1/1 Focalizing ser ('to be') in Dominican Spanish 81 FS (they are probably located at a lower syntactic position than FS), the sentences become slightly more acceptable.These results seem to confirm previous theoretical proposals, according to which elements that convey informational features (i.e.contrast, emphasis, evidentiality, etc.) may be preceded by FS, as they would be syntactically located between FocP and vP (Méndez Vallejo 2009).
Finally, four variations in scope and word order were tested in such a way that each sentence was only semantically logical in a specific context: (  4) and ( 6)) surpassed their wide scope counterparts (see examples (3) and ( 5)).Although these are not conclusive results, they may suggest that the FS construction is more common in contrastive contexts than in non-contrastive ones.Furthermore, sentences with IO-DO word order (see examples ( 5)-( 6)), are slightly more accepted than sentences with DO-IO word order (see examples (3)-( 4)).Again, these results should be viewed with caution, especially because word knowledge, it may be used emphatically in certain dialects, such as Colombian, Venezuelan, Dominican, and Peruvian.
Isogloss 2015, Vol. 1 No. 1 D. Catalina Méndez 82 order may be susceptible to prosodic factors that are not tested when presenting information in a written format.

Semi-production test
This test was run with 13 participants, which resulted in six hours of semidirected conversation.Unfortunately, despite the fact that the test was designed exclusively to promote the production of FS, only 4 cases were compiled.
The first case shows an FS-focused prepositional phrase (7).It is important to mention that the preceding question did not include any type of focus (it was a simple question): (7) A: ¿Dónde está la rana?¿La rana está afuera del frasco?'Where is the frog?Is the frog outside the jar?' B: No, la rana está es adentro del frasco.'No, it is inside the jar that the frog is.' On the other hand, the questions preceding the other three cases included FS focus.In (8), for example, the second question presents an FS-focused object DP and the answer maintains such structure, even when it does not describe the situation in the most accurate way: Isogloss 2015, 1/1 Focalizing ser ('to be') in Dominican Spanish

(8)
A: ¿A quién botó el venado por el abismo?¿El venado botó fue al búho?'Who did the deer throw out of the cliff?Was it the owl that the deer threw out?' B: No, el búho botó fue al niño y al perro por el abismo.'No, it was the boy and the dog that the owl threw out of the cliff.' In ( 9), the answer to the question indicates that both the object DP and the prepositional phrase are focused (these segments are both part of the new information of the sentence): (9) A: ¿Qué fue lo que el niño y el perrito encontraron y dónde?¿El niño y el perrito encontraron fue un par de marmotas dentro del árbol?'What was it that the boy and the dog found and where?Was it a couple of groundhogs that the boy and the dog found inside the tree?' B: No, el niño y el perrito encontraron fue una pareja de ranas detrás de un tronco.
'No, it was a pair of frogs that the boy and the dog found behind a log' Finally, in (10), although the first part of the answer does not follow the format of the preceding question, it includes a case of an FS-focused subject: (10) A: ¿Qué estaba haciendo el venado?¿El venado le estaba dando al niño era una patada?'What was the deer doing?Was it kicking the boy that the deer was doing?' B: Bueno, pues se presentó fue el niño encima del venado.En ningún momento el venado dándole una patada al niño.'Well, it was the boy above the deer that appeared.At no point do we see the deer kicking the boy.' The results of the semi-production test offer a valuable contribution to this study, despite the fact that only a few cases were found.In fact, being able to record the four cases transcribed above suggests that a similar type of test may be suitable for future studies and that this form is indeed present in naturallyoccurring conversation 21 .It is also prominent that these four cases comprise focus types that were also selected as favorably acceptable in the acceptability judgment tests (FS-focused subject and object DPs and FS-focused PPs).
In the next section, we discuss the results of the three tests to show that there are some patterns in the occurrence of FS in Dominican Spanish.

Discussion
The results presented in section 3 exhibit some interesting tendencies regarding the production and perception of FS.First, after examining the results of the 21 From a purely impressionistic perspective, FS seems to be produced much more frequently in other Spanish dialects (e.g. in Colombian varieties).Future investigations should inquire whether or not FS carries social and/or regional stigma in the Dominican Republic.
doing') and Me trajeron una torta fue para mi cumpleaños ('It was for my birthday that they brought me a cake') received average scores above 3.0 in both Colombia and the Dominican Republic (average scores ranging between 3.3 and 3.7).
However, FS-focused CPs and IPs were not as favorably accepted in Colombia and Venezuela as they were in Panama and the Dominican Republic.The average acceptability score for all FS-focused CPs and IPs was 2.2 in Venezuela, 2.4 in Colombia, 3.1 in Panama, and 3.4 in the Dominican Republic.Sentences such as Ella trataba era de imponerle todo ('It was trying to impose everything that she was trying to do') and Sofía parecía era estar enferma todo el tiempo ('It was sick all the time that Sofía seemed to be') received unfavorable scores by Venezuelan speakers (2.4 and 2.1, respectively) and Colombian speakers (2.8 and 2.3, respectively), but highly favorable scores by Panamanian speakers (3.3 and 3.0, respectively) and Dominican speakers (3.9 and 3.5, respectively).
Although these perception patterns are intriguing and suggest intra-and inter-dialectal tendencies, it is essential that we further investigate how FS is used and perceived in various Spanish varieties by maintaining consistent methodological procedures.For example, it would be valuable to replicate the semi-production test in other varieties and to include (and adapt) the same aural stimuli in the acceptability judgment test.
Before continuing to the next section, it is crucial to revisit the results obtained in the acceptability judgment tests for FS-focused subject DPs.Despite the fact that this category did not result in the highest acceptability scores in the first test (see Graphs 1 and 2), two sentences with FS-focused subject DPs actually received some of the highest scores overall in that test: Tendrá dinero será la familia ('It might be the family who has money') and Salió fue Lucía ('It was Lucía who left') were given average scores of 3.9 and 3.5, respectively (see Table 3).These two sentences, when tested among Colombian speakers 23 , also rendered favorable scores: 2.9 for the former and 3.8 for the latter.
However, when looking at the results of the second acceptability judgment test, we find that FS-focused subjects are not accepted so favorably as in the first test: the average acceptability score for the six sentences tested was only 2.35 (see the categories "Focused subject DPs" and "Focused subject DPs & agreement patterns" in Appendix B).As discussed in section 3.2, the sentences Tengo un perro soy yo ('It is I who has a dog') and En la fiesta cocinamos fuimos nosotros ('It was we who cooked at the party'), displayed distinct score distributions: the former had an average rating of 1.7 and the latter had an average rating of 2.9.In spite of the fact that these two sentences also received low scores from Colombian, Venezuelan, and Panamanian speakers 24 , Colombian speakers rated the first sentence at a much higher rate than speakers from the other three countries (2.3 average score in Colombia, 1.9 average score in Panama, and 1.7 average scores in Venezuela and the Dominican Republic).The second sentence, however, was rated much more favorably by Panamanian and Dominican speakers (average scores of 2.6 and 2.9, respectively) than by Venezuelan and Colombian speakers (average scores of 2.1 and 1.8, respectively).
Changing the focus: An empirical study of Isogloss 2015, 1/1 Focalizing ser ('to be') in Dominican Spanish 87 One way to account for the enormous disparity between the highly favorable sentences in the first acceptability judgment test (Tendrá dinero será la familia and Salió fue Lucía) and the less favorable sentences in the second acceptability judgment test (Tengo un perro soy yo and En la fiesta cocinamos fuimos nosotros), may reside in the type of subject DP being focused.That is, in the first two sentences, the focused subject is someone other than the speaker (la familia 'the family' and Lucía) and it is not a personal pronoun (yo 'I' and nosotros 'we').
Another possible explanation may be related to the syntactic and semantic features of the predicate in which the focused element is embedded.For instance, the sentence with the lowest acceptability judgment score (Tengo un perro soy yo) is a transitive sentence with a marked word order (V-DO-S).Two of the other three sentences, however, are intransitive constructions with no particularly marked word order (Salió fue Lucía and En la fiesta cocinamos fuimos nosotros).The other sentence (Tendrá dinero será la familia) also occurs in a transitive sentence with a V-DO-S marked word order, but it has a slight different semantic meaning: instead of stating a fact about a present state of affairs (owning a dog), it expresses a hypothetical explanation about someone else 25 .
Needless to say, these syntactic and semantic constraints should be examined in future studies to elucidate the causes of the perception discrepancies found in FS-focused subject DPs.Incidentally, it is interesting that one of the cases recorded in the semi-production test was an FS-focused subject DP and that it occurs in an intransitive sentence where the focused subject is someone other than the speaker: Se presentó fue el niño encima del venado ('It was the boy above the deer that appeared').

Conclusions
The main purpose of this investigation is to understand how FS is perceived and produced in Dominican Spanish.The results of the acceptability judgment tests and the semi-production test indicate that this form is certainly present in this Spanish variety and that it may not be perceived in the same way across dialects.In particular, both acceptability judgment tests show that certain communities are more inclined to favorably accept cases of FS focus than others.For example, in the first acceptability judgment test, speakers from Santo Domingo generally accepted cases of FS focus more favorably than speakers from Santiago, whereas in the second test, Sosúa speakers consistently accepted cases of FS focus much more favorably than speakers from Santo Domingo and Santiago.
Furthermore, a comparison between the data collected in the Dominican Republic and data collected in other countries (Venezuela, Panama, and Colombia) sheds light on issues of inter-dialectal variation.Although all these varieties display similar perception patterns, it is still possible to unravel dialectal peculiarities.For example, when examining cases of FS-focused subject DPs,

25
The verb tendrá (lit.'will have') in the sentence Tendrá plata será la familia should not be understood as a form of futurity, but rather as a modal verb that conveys probability.Although we explore the occurrence of FS in varieties where this phenomenon has received little attention, it is necessary to keep in mind that this is just an initial approximation to the study of this type of intra-sentential focus and dialectal variation.That is, the results obtained in this investigation should not be used to generalize the way in which FS is perceived and produced in Dominican Spanish.
In fact, this study should serve as a starting point for future empirical research so that focus structures, such as FS, can be analyzed in a more thorough and tangible way.Subsequent studies should develop, first and foremost, more accurate methodological mechanisms to ensure the reliability of the data and the uniformity of the data collection instruments so that speakers from a variety of FS communities can be surveyed and compared objectively.
Aside from these practical considerations, future studies should strive to clarify the semantic and pragmatic characteristics that allow certain cases of FS focus to be more accepted or more frequently produced.Deciphering these subtleties may allow us to explain why FS is used in the first place and why speakers select FS in some contexts and PC in others.

Figure 1 :
Figure 1: Sample picture used in the semi-production test Panamanian and Dominican) seem to accept some sentences more favorably than other dialects (Colombian and Venezuelan).

Table 2 :
Total number of participants by data collection task Table 3 below), indicates that out of the 61 FS sentences tested, 21 are rendered as favorably acceptable by both groups of speakers: * p < 0.05 † p < 0.01 ‡ p < 0.001 Table3: List of most accepted cases of FS focus in the first test

Table 4 :
Table 4, many sentences present the same acceptability patterns in all participant groups: List of all cases of FS focus evaluated in the second test Complete list of sentences tested in the acceptability judgment test in written format * p < 0.05 † p < 0.01 ‡ p < 0.001