Refunctionalization. First-Person Plural of the Verb Haber in the History of Spanish

Abstract

In this paper, the first-person plural diachronic behaviour of the verb form habemos with an existential value is analysed to explore its recovery in current Spanish as a case of refunctionalization. The latter is understood as timely cooptation of a form, which begins with any of the form’s characteristics. It is known that the cooptation’s origin might be directly, indirectly or not at all related to the previous or original use of the form. Results shown here are based on the analysis of constructions in which the first-person plural verb form of haber is used with a possessive meaning, as an auxiliary, and as existential between the 13th and 21st century. While grammaticalization theory pays attention to processes that culminate with grammatical enrichment of words or constructions, the verb form habemos with an existential meaning does not show that behaviour. It is explained as a case of refunctionalization or, at least, specialization.

1. Introduction

It is well known that the verb haber has played a main role in several changes throughout the history of the Spanish language. From Latin to Spanish, it gradually changed from being a possessive transitive verb (1), used with that same meaning in Old Spanish (2a), to being an auxiliary (2b) and an existential verb (2c), until it finally was reduced to an auxiliary level and a plain verb within the existence domain.

(1)	HOSTI	HABET		MUROS
	enemy-NOM SG	have-PRS 3SG		walls-ACS PL
	(Virgilio, Eneida, II, v. 290)
	“The enemy has walls.”
(2) a.	enviar	vos	quiero	a	Castiella
	to send-INF	you-ACUS2P	want-PRS1S	to	Castilla
	do	avemos	heredades
	where	have-PRS1P	inheritances
	(Cid, v. 1271)
	“I want to send you to Castilla where we have inheritances.”
b.	rogo-l	que	non	dixiesse
	beg.PST3S-DAT3SG	that	not	tell-PST.SBJC3SG
	que lo	auie	acabado
	that	it-ACS3SG	AUX-IPFV3SG	finished-PTCP
	fasta que	los otros	ouiessen
	until CONJ	the others	AUX-IPFV.SBJC3P
	cerca d-acabadas	sus	obras
	near of-finished-PTCP	POSS-3PL	works
	(CG, 12a)
	“S/he begged her/him not to tell that s/he had finished until the others were about to finish their works.”
	Nos	habemos	mandado	a
	we	AUX-PRES1P	ordered-PTCP	to-DAT
	nuestro	amado	clerigo	Michelet de	Mares
	POSS-1P	beloved	priest	Michelet from	Mares
	que	ficiesse	adobar	los dichos	altares
	that	make-PST.SBJC3S	to tan-INF	the mentioned	altars
	(Gastos de la catedral de Pamplona, 1400, Spain, corde, s.v. habemos mandado)
	“We have ordered our beloved priest Michelet of Mares to prepare the aforementioned altars.”
	Conosciendo	esto	e	sabiendo	que
	Knowing	this	and	realizing	that
	habemos	de	ir	a aquella	vida
	AUX-PRES1P	PREP	to go-INF	to that	life
	perdurable	[…] tenemos	por	derecho	de
	everlasting	AUX-PRES1P	by	law	PREP
	lo	emendar	a	Dios
	it-ACU3S	amend-INF	to	God
	(Real Monasterio de Santo Domingo, 1291, Spain, corde, s.v. habemos de)
	“By knowing this and realising that we must go to that everlasting life […] we consider (that) we must redeem ourselves before God.”

c.	ssi	en	la	iglesia	mayor	non
	if	in	the	church	main	not
	ouyesse	obispo	et	el cabillo
	there be- IPFV.SBJC3S	bishop	and	the chapter
	della	rogase	a	otro
	of.ART	beg-IPFV.SBJC3S	to	someone else
	que	por	y	pasasse	que
	that	by	there	go by-IPFV.SBJC3S	that
	lo	ffiziesse
	it-ACUS3S	do-IPFV.SBJC3S
	(Setenario, 161.24)
	“If in the main church there weren’t a bishop and if the council of the church begged someone else to go there, let him do it.”

The replacement process from haber to tener has merited detailed studies in which the change has been explained using grammaticalization theory (Garachana Camarero 1994, 1997; Hernández Díaz 1999, 2006a, 2006b), which has also been used to explain the extension and generalization of haber as the prototype of existential verbs (Hernández Díaz 2007). It is well known that the change of haber from a transitive to an existential verb has reduced its paradigm to the third-person singular, since the rest of the grammatical paradigm turned out to be unnecessary because it was opposed to the new impersonal meaning of haber. Once the paradigm was reduced, haber was classified in grammar books as a third person verb (Bello [1847] 1997, §781–84; Gili Gaya 1943, §62; Seco 1989, §8.2.2; RAE 1973, §3.5.7.c y d; RAE/ASALE 2009, §41.6a, b).

Notwithstanding the new morphologic characterization of the verb, agreement between the noun phrase, designating what I call ‘existing entity’ and the verb, has consistently been rejected by grammarians (Bello [1847] 1997, §781–784; Gili Gaya 1943, §62; Seco 1989, §8.2.2 footnote; RAE 1973, §3.5.7.c, d; RAE/ASALE 2009, §41.6b, c, d, e). This rejection proves that the noun phrase has been reanalysed as the syntactical subject, because, even though existential sentences using haber are not classified as intransitive structures, they are closer to that classification than to that of transitive structures in speakers’ minds. As we can see in example (3), agreement is a limited phenomenon in terms of frequency but it is a persistent phenomenon diachronically speaking.

(3)	Habían		algunos		hechiceros
	There are-IPFV3P		some		sorcerers,
	ministros		del		diablo		que		los
	ministers		of.ART.S.		devil		that		ACUS-3P
	inducían		en		algunos		supersticiosos		errores
	incite-IPFV3P		PREP		some		superstitious		mistakes
	(Apologética historia sumaria, 1527–1550, Spain, corde, s.v. habían)
	“There were some sorcerers, ministers of the devil who led them into some superstitious mistakes.”
	todavía		creo		que		habrán		muchos
	still		believe-PRES1S		that		thereareFUT3P		many
	que		se declararán		por		V. M.
	that		declare-FUT3P		in favor of		Your Majesty
	(Historia de Felipe II, c.1619, Spain, CORDE, s.v. habrán)
	“I still believe that there will be many who will declare themselves in favor of Your Majesty.”
	aunque	hubieron		algunos		indios
	thoug	therearePST3P		some		indians
	hicieron	sus		tiros,		nada		hizo
	make-PST3P	POS-3P		shoots,		nothing		do-PST3S
	(Diario de la sublevación del año 1781, c. 1781, Bolivia, CORDE, s.v. hubieron)
	“Though there were some Indians, [and] they shot, s/he did nothing.”

Apart from the agreement in example (3), there are cases in the first-person plural verb form of haber that use the presumably disappeared existential meaning of habemos. Some examples of these cases are displayed in (4), where reanalysis of the noun phrase as a subject is updated and there has been a refunctionalization, because the discussed verb form habemos experienced a change in function that goes beyond the syntactic description level. The phenomenon involves recovering an element from the brink of extinction in standard Spanish, which acquired a new semantic and pragmatic meaning, as I will show in this paper.1

(4)	Vea	esto	no		es
	Look-IMP2S	this	not		be-PRES3S
	Rionegro,	donde	sí		habemos
	Rionegro	where	indeed		there bePRES1P
	muchos blancos,	por	los		cuatro	costaos
	many whites	by	the		four	sides
	(La marquesa de Yolombó, 1928, Colombia, corde, s.v. habemos muchos)
	“Look: this is not Rionegro, where there are many of us whites indeed, entirely white.”
	la organización	no	solicitó		los permisos	para
	the organisation	not	request-PST3S		the permissions	for
	la venta de alcohol,	por tanto	hubimos		varios
	the sale of alcohol,	therefore	there be-PST1P		several
	expositores	que	nos quedamos		con	el	viaje
	expositors	who	keep-PST1P		with	the	trip
	(Blog “Identidad y futuro,” 2015, Chile)
	“The organisation did not request the permissions for the sale of alcohol; therefore, there were many of us speakers who did not make the trip.”
	¿cómo	se		consigue	un experto?	Aquí en	la
	How	IMPER		get-PRES3S	an expert	here in	the
	nube docente	habremos		varios
	cloud	teaching there be-FUT3P		many
	listos	a		apoyar-nos
	ready	to		support.INF-ACUS1P
	(Webpage, 2015, México)
	“How can someone get an expert? Here in the teaching cloud there will be many of us to support each other.”

2. Purpose

The purpose of this paper is to analyse first-person plural diachronic behaviour of habemos, with an existential meaning, to show its resurgence in current Spanish language as a case of the refunctionalization and exaptation phenomena. I understand the latter as timely cooptation of a form, a process that begins with any of the form’s characteristics, according to Lass (1990, p. 80). It is well known that the cooptation’s origin might be directly, indirectly or not at all related to the previous or original use of the form; in any case, it constitutes a conceptual innovation.

Exaptation changes, unlike any other changes, are characterized by renewing or recycling so called ‘historic junk’ in languages. Recycling starts with abandoned forms—at some linguistic analysis level—because they have lost their function (Lass 1990; Narrog 2007; Traugott 2004). Although it is true that morphological material rarely stops fulfilling its function, there are cases in which its function is partially lost or is kept only marginally. This, as we shall see in this paper, is the case for the verb form habemos in the history of the Spanish language.

3. Corpus

Results shown here are based on the analysis of 47,883 constructions in which the first-person plural of haber (habemos/avemos, hemos) is used with a possessive meaning, as an auxiliary and as an existential, in the period between the 13th and the 21st century (cf. Appendixe A and Appendixe B). However, as we know, possessive meaning was lost during the 16th century and, contrastingly, the verb form habemos is hardly registered as existential verb in writers preceding the 20th century.2 This difficulty is undoubtedly related to the generalized disapproval of haber concordance by normative grammar; it is also due to the textual typology represented in the linguistic corpus and the methodology used in the creation of the linguistic atlases (cf. Castillo Lluch and Octavio de Toledo 2016).

Cases were extracted from the Corpus Diacrónico del Español (corde) by the Real Academia Española (RAE). Additionally, cases corresponding to the 20th century were complemented with data from the Corpus de Referencia del Español Actual (crea), also by the RAE. 21st century cases were gathered using Google and the webpage Socialmention.com (www.socialmention.com), which is used to perform real time searches in social networks and internet. The plurality in the corpus is caused by the difficulty faced in registering the habemos form/expression in standard Spanish, though the form/expression is highly productive in oral exchanges and in chats, which have a very similar format to that of a conversation and support it despite being in written form.

The data considered in this study are 6457 cases of habemos or avemos, as a possessive or as an auxiliary verb, 41,167 cases of hemos expressing the same meaning and 259 cases of the verb form habemos as an existential verb.

As I already have mentioned, not all of them correspond to an existential use of haber. For contrasting purposes, in order to prove what I have proposed, I also include possessive, auxiliary and what I have called ‘prescriptive uses,’ as well as ‘other uses’ that I will explain in detail later on.3 This study’s scope is vast and heterogeneous in order to compare the diverse meanings of the verb form habemos from a diachronic perspective, for I believe that the wide variety of examples will play an important part in evidencing a contrast in quantitative terms. As we shall see, quantitative and qualitative differences in the meaning of the verb form habemos are crucial, for they help us understand why the existential use of this verb form is a case of refunctionalization and, in some way, a case of exaptation.

4. Reanalysis, Refunctionalization and Subjectivization of the Verb Form Habemos

As I have mentioned before, one of the changes experienced by haber was from being a transitive verb to an impersonal one. This change caused its paradigm to be reduced to the third-person singular although, by participating in compound tenses and in verbal periphrases as an auxiliary verb, it managed to keep its whole paradigm identical to the one it used to have as a transitive verb.

Once habemos became an auxiliary verb, the first-person plural experienced a formal reduction. The change avemos cantado > hemos cantado (i.e., the phonetic reduction of the fourth person or the first-person plural of the perfect) has been carefully analysed in Bustos Gisbert and Moreno Bernal (1992) and, more recently, in Rodríguez Molina (2010, 2012). Rodríguez Molina (2012), based on Girón Alconchel (2004, p. 870), states that the alternation hemos/habemos (avemos) became less and less frequent once the transitive use of haber was lost. The shortened form was a more effective iconic manifestation of its use as an auxiliary. According to Rodríguez Molina, the reduction started at the beginning of the 16th century and by the 18th century habemos disappeared from the standard variant, although it has dialectal presence (cf. Rodríguez Molina 2012, p. 181) and the data on which this work is based are consistent with this. From the first quarter of the 16th century, the long forms are obsolete and appear mainly in legal texts, pastoral poetry and stigmatized dialects (Girón Alconchel 2004, p. 866).

The dialect variation is important in order to track the spread of change. Rodríguez Molina’s data (Rodríguez Molina 2012, p. 207) suggest that the short form (hemos cantado) emerges in Aragon and spreads from there. According to this author, the causes of change can be grouped as: (1) phonetic reduction after the grammaticalization of compound verb tenses; (2) essentially phonetic factors (the bilabial occlusive voiced sound loss in the intervocalic position and the syncope of the low central vowel); (3) an analogy process to preserve the reduction in the verbal paradigm.4

According to our results, reduction must have happened around the 16th century. The corde registers 98 cases of hemos used as an auxiliary verb in the 13th century, a number that increased in the following centuries and reached 4174 cases by the 16th century. Throughout this period, the frequency of habemos (avemos) and hemos as an auxiliary is similar but two centuries later, in the 18th century, the frequency of hemos is recorded at 2162—and increases to 13,000 in the 19th century and 16,503 in the 20th century. On the other hand, the frequency of habemos decreases; there were 81 recorded cases in the 18th century, 68 cases in the 19th century and there was a small rebound in the 20th century, when there were 145 recorded cases. These results seem significant to us.

Based on the results of

Table 1. Possessive meanings vs. habemos (avemos) and hemos as an auxiliary verb.

	Habemos/Avemos		Hemos
	Possessive	Auxiliary	Possessive	Auxiliary
xiii	7%	93%	8%	92%
	(27/373)	(346/373)	(8/106)	(98/106)
xiv	10%	90%	6%	94%
	(13/124)	(111/124)	(12/197)	(185/197)
xv	21%	79%	1%	99%
	(106/496)	(390/496)	(2/459)	(457/459)
xvi	8%	92%	1%	99%
	(306/3715)	(3409/3715)	(29/4203)	(4174/4203)
xvii	10%	90%	2%	98%
	(135/1411)	(1276/1411)	(105/4512)	(4407/4512)
xviii	2%	98%	7%	93%
	(2/83)	(81/83)	(13/2175)	(2162/2175)
xix	28%	72%	<0.5%	99%
	(29/97)	(68/97)	(7/13,007)	(13,000/13,007)
xx	8%	92%	<0.5%	99%
	(13/158)	(145/158)	(5/16,508)	(16,503/16,508)

, we can state that habemos was kept in relation to the possessive meaning and, residually, as an auxiliary verb. On the other hand, hemos was generalized as the prototypical auxiliary verb (see the numbers in bold print in Table 1).

Table 1 shows very different use frequencies for habemos (avemos) vs. hemos as auxiliary verb forms in the 18th, 19th and 20th century. If one compares only the auxiliary meaning of these forms, as in

Table 2. Habemos vs. hemos as auxiliary verbs.

	Habemos Auxiliary	Hemos Auxiliary
xviii	4% (81/2243)	96% (2162/2243)
xix	<1% (68/13,068)	99% (13,000/13,068)
xx	<1% (145/16,648)	99% (16,503/16,648)

, it appears that habemos (avemos) is marginally used as an auxiliary verb as compared to hemos.

Considering these results, it appears that habemos is also bound to disappear as an auxiliary verb form, especially during the 19th and 20th century, even if one considers its presence in some Spanish dialects. As a possessive verb, its situation is similar; during the 20th century it was used sporadically in fixed phrases like the ones in example (5a). It is clear that its use as an auxiliary or as a possessive verb form has no major relevance in terms of frequency. Habemos was registered in cases like the ones in example (5b), which, although recent, seem to be marginal.

(5) a.	Y	cuánta	sabiduría	habemos menester
	And	how much	wisdom	need-PRES1P
	para	retener	nuestros	prisioneros
	for	hold back-INF	POS-1P	prisoners
	sin	destripar-los
	without	gut.INF-ACUS3P
	(Hace tiempos, 1935–1936, Colombia, corde, s.v. habemos menester)
	“And how much wisdom must we need in order to retain our prisoners without gutting them.”
	Si	de	todo	lo	que
	If	of	everything	PRON	REL-3S
	habemos menester	hubiese	copia	sobrada
	need-PRES1P	thereisIPFV.SBJV	copy	extra
	(Misión de la Universidad, 1930, Spain, corde, s.v. habemos menester)
	“If there were an extra copy of everything we need.”

b.	alguna vez	nos	habemos de	apartar
	some time	ACU-1P	AUX-PRES1P	move away-INF
	del	común y simple	modo	de	decir
	of.ART	common and simple	way	of	say
	(Poesía española. Ensayo de métodos y límites estilísticos, 1950, Spain, corde, s.v. habemos de apartar)
	“Some time we must have to move away from the common and simple way of saying.”
	mandar	matar	a	un hombre ordinario,
	order-INF	kill-INF	to	a man ordinary
	pone	a	un hombre	tan grande	en
	place-PRES3S	to	a man	so great	in
	el	estrecho	que	habemos	visto
	the	strait	that	AUX-PRES1P	seen-PTCP
	(Discurso de recepción en la Real Academia Española: Pasión y muerte del Conde de Villamediana, 1964, Spain, corde, s.v. habemos visto)
	“To order an ordinary man to be killed, places a great man in the situation that we have seen.”
	ya	sabéis	la	voluntad	que
	already	know-PRES2P	the	will	that
	la Católica Reina mi Señora,	é	yo	habemos
	the My Lady the Catholic Queen	and	I	AUX-PRES1P
	tenido	é	tenemos	al	bien
	had-PTP	and	have-PRES1P	to.ART	good
	(Contrapunteo cubano del tabaco y el azúcar, 1963, Cuba, corde, s.v. habemos tenido)
	“You know already what My Lady the Catholic Queen and I have willed to be correct.”

Despite what we have previously said, the frequency of habemos grows in the 21st century, when the verb has an existential meaning, because, as is widely known, the agreement of haber in contemporary Spanish is a general and widely spread phenomenon, even though it is not always recorded in written form.

In my opinion, as stated in a previous study on haber’s agreement (Hernández Díaz 2005), it is necessary to make a distinction between the creation of two different kinds of agreement, for they depend on different semantic, and, more importantly, pragmatic reasons. I am referring to the contrast in example (6).

(6) a.	Lo	más	enriquecedor	fue	cómo
	The	most	enriching	be-PST3S	how
	contaban	las tradiciones	que	habían
	relate-IPFV3P	the traditions	that	there are-IPFV3P
	en	sus	pueblos
	in	POS-3P	hometown
	(Manuscript, c. 2000, México)
	“What was most enriching was how they told of the traditions that they had in their hometowns.”
	no	deben	haber	fueros	ni
	not	AUX-PRES3P	be-INF	exemptions	nor
	privilegios	respect	e	este	problema
	privileges	regarding	of	this	problem
	(TV, México)
	“There should not be neither exemptions nor privileges regarding this problem.”
	No	habían	copias,	entonces	reduje
	Not	there be-IPFV3P	photocopies	so	reduce-PST1S
	los horarios	para	hacer-los	en	la	computadora
	the schedules	for	make.INF-ACUS3P	on	the	computer
	(Spoken language, México)
	“There were no copies, so I reduced the schedules in order to make them on the computer.”

b.	Habemos	muchos	aquí	que	somos	inocentes
	There be-PRES1P	many	here	that	be-PRES1P	innocent
	(Socialmention, 2015)
	“There are many of us here who are innocent.”
	yo	sé	que	habemos
	I	know-PRES1S	that	there be-PRES1P
	muchos	que nos	sentimentos	así	y
	many	who	feel-PRES1P	like this	and
	no	tiene	nada	de	malo
	not	have-PRES3S	nothing	about	wrong
	(Socialmention, 2015)
	“I know that there are many of us who feel this way and there is nothing wrong about it.”
	habemos	personas	que	por	fuera
	there be-PRES1P	people	who	by	outsid
	aparentamos	y	fingimos	ser	humildes
	pretend-PRES1P	and	feign-PRES1P	be-INF	humble
	(Google, 2015)
	“There are many people of us who on the outside pretend and feign to be humble people.”

In both (6a) and (6b), the reanalysis of the element governed by the existential verb haber, as the subject of the sentence, is evident. This mechanism was defined as follows:

The change in the structure of an expression or kind of expression, which is not related to any immediate or essential modification in its superficial expression.

(Langacker 1977, p. 58)

As different authors have stated (Company 2010; García 1990; Narrog 2007), a reanalysis might or might not have repercussions in formal expression but will definitely have repercussions on a conceptual level. Reanalysis is completed, in cases like example (6), because it has happened at an internal and external level of expression and because of the way in which it manifests itself or becomes updated in the agreement.

As we know, reanalysis is a major mechanism for grammaticalization and for linguistic change in general, mainly because it is a prerequisite for the implementation of change through analogy: a process that modifies shallow evidences and that spreads reanalysis’ effects not only within the same linguistic system but also inside the speaking community. The analogy that caused the grammatical uses I am interested in was the intransitive mono-argument constructions’ formal structure, because whenever a predicate has only one argument, it will prototypically be the subject of the sentence. The first example in (6a) shows a very evident analogy because the existential verb habían is coordinated with existían, which makes the two sentences look syntactically equal in the eyes of the speaker.

The development of agreement in habemos is a case of refunctionalization, except for the one exhibited in (6a), because it does not only imply the reanalysis of an existing entity as the sentence’s subject but also the reuse of what is almost a junk form or a very marginal one in standard Spanish, as we have seen. Based on the results shown in Table 1, we can attest that the rotation of habemos and hemos seems to have been related, at some point, to the former’s possessive value and that, once it lost this value, habemos was occasionally kept as an auxiliary verb. Keeping both forms, then, turned out to be unnecessary. Therefore, habemos became a morphological archaic case of hemos but in terms of system and frequency the former was practically considered to be ‘junk.’ According to Lass (1988, p. 36), there are only three outcomes for a residual expression: (1) complete loss; (2) remaining as waste without a specific function; or (3) remaining and being systematically used for another purpose, instead of being left aside. The third option is defined by Lass as exaptation and, according to the results in

Table 3. Recorded meaning of habemos (avemos).

		Possessive	Auxiliary	Existential	Prescriptive Use	Other Uses
xiii	corde	7%	93%	-	-	-
		(27/373)	(346/373)
xiv	corde	10%	90%	-	-	-
		(13/124)	(111/124)
xv	corde	21%	79%	-	-	-
		(106/496)	(390/496)
xvi	corde	8%	92%	-	-	-
		(306/3715)	(3409/3715)
xvii	corde	10%	90%	-	-	-
		(135/1411)	(1276/1411)
xviii	corde	2%	98%	-	-	-
		(2/83)	(81/83)
xix	corde	28%	59%	-	-	13%
		(29/105)	(62/105)			(14/105)
xx	corde	8%	83%	8%	1%	-
		(11/143)	(119/143)	(12/143)	(1/143)
	crea	3%	39%	47%	2%	9%
		(2/57) †	(22/57)	(27/57)	(1/57)	(5/57)
xxi	Google††	-	2%	63%	34%	1%
			(4/160)	(100/160)	(55/160)	(1/160)
	SocialM	-	-	98%	2%	-
				(105/107)	(2/107)

^† Eventhough crea shows 58 results of habemos, one of these was left out because habernos was mistakenly typed as habemos (tú y yo podríamos habemos casado). ^†† I took the first 100 existential cases of habemos: 55 correction cases, 4 cases used as auxiliary and one more corresponding to the explanation of the Latin expression habemus papam (verb + accusative case).

, that is precisely what happened to habemos.

Even though grammarians in the 19th century identify the use of habemos with the meaning ‘existir’ (‘to exist’) as a mistaken agreement, results show that during the 20th and 21st century exaptation was the solution for habemos. This verb form was marginally used as an auxiliary or with a possessive meaning during the 19th and 20th centuries. However, during the 20th century and during the first fourteen years of the 21st century, it regained strength not only to express existence—for habemos with existential meaning does not merely mean that algo está o existe en alguna parte (‘something is or exists somewhere’)—but also as the verb form that allows the speaker to include himself as member of a group or class. Such a group exhibits a clear set of characteristics and is located inside space and time coordinates—real and metaphorical—as is shown in example (7). Thus, the recovery of habemos involved its refunctionalization, through subjectivization, as well. Moreover, since habemos, unlike other impersonal existential verb forms, was reused as an existential verb with the option to indicate the grammatical person, in some way, it might be seen as an exaptation process.

(7)	De	hecho,	habemos	un	equipo	grande
	In	fact,	there be-PRES1P	a	team	big
	que	seguimos	trabajando	en	ello
	that	AUX-PRES1P	working-gerund	on	it
	(Socialmention 2015)
	“In fact, we are a large team that continues to work on it.”
	No generalizo	que	todos	somos
	Not generalize-PRES1S	that	everybody	be-PRES1P
	asi (sic)	y	se (sic)	que	habemos
	like that	and	know-PRES1S	that	there be-PRES1P
	personas	que	estamos	orgullosos	de	quienes	somos!
	people	that	be-PRES1P	proud	of	who	be- PRES1P
	(Socialmention, 2015)
	I do not generalize that we all are like this and I know that we are people who are proud of who we are.”
	Con respecto	al	trato	que	se
	Regarding	to.ART	treatment	that	IMPER
	les	da	a	las personas	creo
	DAT-3P	give-PRE3S	to	the people	believe-PRES1S
	que	no	es	muy bueno,	ya que
	that	not	be-PRES3S	very good,	because
	habemos	personas	muy	ignorantes	y
	therebePRES1P	people	very	ignorant	and
	que	nos	consideramos	con	mucha
	that	AUX-1P	consider-PRES1P	with	so much
	suerte	como	para	no	tener	ese	tipo	de enfermedad
	lucky	as		not	have-INF	that	kind of	illness
	(Google, 2015)
	“Regarding the treatment that has been given to people, I think that it is not such a good thing, because we are very ignorant people and we consider ourselves lucky for not having that kind of illness.”
	y	si (sic)	habemos	personas	que
	and	yes	there be-PRES1P	people	that
	somos	como	somos	de	nacimiento
	be-PRES1P	like	be-PRES1P	of	birth
	y	no	podremos	cambiar
	and	not	can-FUT1P	change-INF
	(Google, 2015)
	“And, indeed, we are people who we are from birth and we won’t be able to change.”

Going back to the results of Table 3, I deem it necessary to explain the two far right columns: ‘prescriptive use’ and ‘other use.’ The former attracts my interest because, in Google, these cases equal 34%, that is, 55 of the 160 first reported examples. I decided to classify them as ‘prescriptive use’ because they are, in a sense, an expansion of suggestions by grammar books to avoid agreement between the verb haber and a mono-argument noun phrase. These results refer to articles, pages, blogs or chats that describe this agreement as incorrect and they censure it. This shows a genuine interest in the subject, as well as the frequency of the phenomenon, for we can assume that prescription occurs as often as there is a violation of a rule. Then, to the far right, the ‘other use’ column gathers fixed phrases where haber is used with possessive meaning but it does not make any sense to classify them as truly transitive uses. Taking this into account, the refunctionalization of the verb form habemos seems to be almost complete. Figure 1 shows the refunctionalization of habemos diachronically.

There is no doubt that the refunctionalization of habemos may be explained as a change through subjectivization, because grammatical alternatives, which are reduced to formal options dissociated from meaning in some theories, are studied in change theories as carriers of meaning. Such meaning is defined by different perception forms in a situation, rather than by different objective or fact conditions. Exposing the figurative condition of language allows us to approach rotation as part of the common pragmatic meaning of language, where we can see that change in shape also implies change in meaning.

Having observed an increasing tendency in correcting the agreement of habemos, we could ask ourselves why there is insistence on exaptation. The answer lies in the hypothesis ‘change in shape implies change in meaning’ because, subjectively, the advantage of habemos over hay or even over other verbs that can give constructions an existential sense, is that the speaker is able to include himself in the referred event as a directly affected member of the situation. It is worth mentioning that existential sentences with habemos in the corpus usually refer to events where the speaker, as part of the subject, plays the part of ‘theme.’ A theme that is somehow affected by the described situation, because sentences include theme as being part of a class with specific circumstances, frequently deemed negative (for example signs of vulnerability), as shown in italics in (8).

(8)	habemos	personas		que		le
	there be-PRES1P	people		who		DAT-3S
	damos	fiebre		a		la	calentura
	give-PRES1P	fever		to		the	temperatura
	(Doña Bárbara, 1929, Venezuela, CORDE, s.v. habemos personas)
	“There are those of us who give fever to fever.”
	hay,	habemos		todavía
	there be-PRES3S	there be-PRES1P		still
	militares	pundonorosos,		para		defender
	militaries	honorable		for		defend-INF
	los intereses	del		pueblo		y	de la	Patria!
	the interests	of.ART		people		and	of the	homeland
	(El Papa verde, 1954, Guatemala, corde, s.v. habemos)
	“There are still honorable men, including me, to defend the interests of the people and of the homeland.”
	habemos	maestros		que		tienen (sic)	25años	de enseñanza
	there be-PRES1P	teachers		who		have-PRES3P	twenty-five years	of teaching
	(Proceso, 1997, México, CREA, s.v. habemos)
	“There are those of us teachers that have been teaching for 25 years.”
	habemos	muchos		venezolanos		que
	There be-PRES1P	many		Venezuelans		who
	estamo	inocentes		de		todo	estos	males
	be-PRES1P	innocent		of		all	these	misfortunes
	(Socialmention, 2015)
	“There are many of us Venezuelans who are innocent of all these misfortunes.”
	habemos		miles		de	profesionales
	There be-PRES1P		thousands		of	profesionales
	en		este sector		que	aunque	mal	pagados,
	in		this sector		that	even	bad	payed-PTCP
	exprimidos		por		los	empresarios	y
	used-PTCP		by		the	enterprise people	and
	con		convenios caducados		tenemos	un	trabajo
	with		agreements expired		have-PRES1P	a	job
	(Socialmention, 2015)
	“There are thousands of us professionals in this sector that though poorly payed, exploited by the businessmen with bad contracts, we still have a job.”

Thus, the refunctionalization of habemos is another way in which the Spanish expresses a change of mood within the syntactic subject. This is the reason why, even though the phenomenon is constantly condemned, its presence is a fact, due to the communicative advantages it has in pragmatics.

The cooptation of habemos to mean the ‘existence of a class to which the speaker belongs or includes himself’ is possible thanks to the personal inflectional morpheme. As we know, recycling of the form may be or may not directly related to its former or original use or can be used marginally. For habemos, we consider it is the second case because the marginal relationship lies in the personal form of the verb but it is not related to its previous possessive value. It is related to an existential value it rarely had in Latin and which increased in Old Spanish. The recycling and recovery of habemos to include the speaker as part of the subject is overwhelming, if we consider the most frequent combinations of habemos. In accordance with Google web browser results, in frequency order they are as follows: habemos personas que… (‘There are those of us people that…’) (393,000 cases), habemos gente que… (‘There are those of us that…’) (389,000), habemos algunos que… (‘There are some of us people that…’) (305,000), habemos algunas que… (‘There are some of us women that…’) (229,000), habemos quienes… (‘There are those of us who…’) (225,000 cases), habemos hombres que… (‘There are those of us that…’) (221,000), habemos mujeres que… (‘There are those of us women that…’) (219,000), habemos muchos que… (‘There are many of us that…’) (174,000 cases).

In approaching the end of this analysis, I would like to make it clear that my approach to haber has been a many-angled diachronic one, inserted in grammaticalization and reanalysis theory. I would also like to add that I did not find those theories useful in explaining refunctionalization regarding the development of agreement or the special case of habemos, because we are not dealing with a gradual one-way phenomenon in Spanish language history. It is surprisingly odd, within these theories, to find a practically useless obsolete word regaining expressive strength. However, under light shed by refunctionalization and by subjectivization, as an exaptation case, the change turns out to be natural and transparent. Thus, I consider that the development of agreement in haber, specifically in habemos, can be partially explained from on the grounds of grammaticalization. I also consider that both grammaticalization and exaptation allow us to explain a sudden phenomenon that neither has a unidirectional behaviour, nor ends in a grammatical or more grammatical form than the one at the beginning. It is a phenomenon with a form that reinserts itself, inside the verbal paradigm by building a new paradigm, through reanalysis and refunctionalization, due to subjective and pragmatic value assessment.

5. Conclusions

We can conclude that it is possible to account for the diachrony of habemos, in the history of the Spanish language, as a refunctionalization case. Through this concept, we can explain changes that do not follow a unidirectional trajectory, are not gradual and do not end with the creation of more grammatical forms than the ones from previous stages.

In accordance with data gathered in this analysis, change—or rather the recycling of habemos—has been possible due to communicative advantages that other existential verbs do not have; particularly, the possibility to include the speaker in an existential sentence and as member of a group or a class with well-defined characteristics. Thus, in order to explain the whole diachrony of the verb haber, from possession to existential domain, we must acknowledge the relevant role played by processes such as refunctionalization, subjectivization and to some degree, exaptation.