versão On-line ISSN 2312-3621
versão impressa ISSN 1010-9919
Old testam. essays vol.26 no.1 Pretoria Jan. 2013
Against Floccinaucinihilipilification of the counterfactual sense of the BH suffix conjugation - or an Explanation of why the "Indicative" Qatal expresses conditions, hypotheses and wishes1
University of Stellenbosch
The present paper demonstrates that the counterfactual value displayed by the BH qatal is a rightful and logical component of the total meaning of the suffix conjugation, understood as a network of conceptually and historically connected senses. The chaining procedure built on the framework of universal paths (viz. a theory of typologically highly plausible evolutionary scenarios) enables the author to relate six specific values of the counterfactual domain (real and unreal optative, real and unreal hypothetical, as well as real and unreal conditional), and to establish their diachronic and synchronic (i.e. conceptual) arrangement. Furthermore, by employing an analogical chaining method, the entire counterfactual block is linked to the dominant indicative type of the qatal and, in particular, to its perfect, perfective and past values. The results of the article also demonstrate that - contrary to widespread opinion -the optative use of the BH qatal does not reflect a loss or omission of an original apodosis. Quite the reverse, the optative value was the initial modal meaning that the Proto-Semitic *qatal- acquired in the vicinity of the particle *law and its negative varieties: the BH optative examples are, simply then, remnants of such an ancient usage.
1 The Problem of the Counterfactual Qatal
The Biblical Hebrew (BH) verbal formation qatal most commonly expresses indicative values, being extensively used as a perfect (present perfect and plu-perfect), perfective and past (PPP) gram. According to McFall,2 these PPP senses amount to 10,830 of all 13,874 cases (78%) of the suffix conjugation.3 In other less frequent indicative functions, the locution approximates the category of a stative present (especially when it is derived from adjectival and static roots; this occurs in 2,454 cases that constitute 18% of the total number of the examples where the gram4 is employed) or expresses certain future values (255 or less than 2%). The predominance of these perfect, perfective and past senses is so evident and marked that grammarians, in their classifications of the qatal, have regularly focused on this portion of the gram's semantics, labelling the form as a perfect,5 perfective6 or past tense.7
Furthermore, a great prevalence of the indicative uses which equal 98% of the instances where the gram appears - versus 1.5% where the modal force is available8 - has usually lead scholars to view the qatal as a prototypically indicative (or unmarked as far as modality is concerned) construction opposed to a more modally marked yiqtol.9 The inherent non-modal nature of the suffix conjugation has commonly been accepted to the point of constituting the systematic basis of the entire BH organisation. This understanding (already mentioned being extremely widespread) is even more evident in models that are explicitly built on the parameter of mood where the suffix conjugation is straightforwardly classified as realis, indicative or non-modal. For instance, Loprieno10 establishes an aspectual-modal model for Biblical Hebrew and the Semitic family in which the "perfective" qatal is defined as a positively marked realis. In a similar vein, Rattray11 classifies the qatal as a perfective realis of immediate reality. According to Zuber,12 the qatal is a "recto-Form" introducing indicative statements while Joosten13 classifies the qatal as an indicative anterior. Finally, Hatav14 understands the qatal as a non-modal form (non-modal, non-sequential, non-progressive, but positively marked for the feature 'perfect') and DeCaen15 identifies it with a past non-modal category, syntactically marked as occupying the clause's second position.
Despite this almost exceptionless unanimity concerning the indicative character of the qatal, grammarians are aware of certain cases where the formation offers modal readings. One of the most typical modal uses of the suffix conjugation corresponds to examples where the gram is employed with the particles and . In such instances, the construction invariably expresses counterfactual wishes (approximating the category of an optative) or - when used in conditional protases and apodoses - counterfactual conditions (corresponding to a protetic subjunctive16 category) or hypotheses (approaching an apodotic conditional). Put differently, the qatal with and seems to be a different grammatical creature, to an extent, semantically detached from its perfect-perfective-past and indicative equivalent. Consequently, a paradoxical -at least superficially - situation may be observed: a typical indicative formation is a regular means of conveying counterfactual optative, conditional (protetic) and hypothetical (apodocic) meanings.
2 Grammatical Tradition
In grammar books, the counterfactual optative-conditional-hypothetical function of the qatal has usually been marginalised and its discussion confined to a few general statements. This type of value of the suffix conjugation has principally been treated in sections devoted to conditional or volitional (optative) phrases, even being removed from chapters where the semantics of the suffix conjugation are presented.17 It appears as if grammarians do not pay attention to the optative-hypothetical qatal, with the intent to disconnect it - as unfitting to a chosen classification - from the meaning of the gram.
Already Driver18 observes that the qatal may appear in conditional protases where, introduced by the particle , it expresses non-realised events. According to this scholar, when the apodosis fails to be articulated, the construction denotes non-realisable wishes.19 Gesenius-Kautzsch-Cowley20 interestingly note that the suffix conjugation is used in conditional sentences preceded by with a force analogous to the Latin imperfect and pluperfect subjunctive: amaram and amavissem, respectively. This signifies that, in a past time frame, the formation expresses conditions which were not fulfilled while, in a present time frame, it introduces conditions that are viewed as unlikely to be fulfilled.21 Additionally, the authors identify a use where the locution conveys the meaning of unfulfilled desires or activities whose accomplishment in the past is regarded as possible but not as factual.22 According to Davidson,23 in cases where the qatal is found in hypothetical sentences headed by or - in negative - , it denotes actions that are not realised in the past or are not realisable (and hence, unlikely) in the present-future. Besides this conditional use, the locution may also appear with an optative force, invariably unreal with a past temporal reference.24 Joüon25 affirms that when the qatal is introduced by , it expresses wishes referred to past and/or counterfactual conditions and hypotheses. Watts26 argues that the sequence: + qatal indicates contrary to fact conditions, portraying a given situation as non-existent and impossible.27 In this function - he claims - the gram approximates the category of a subjunctive. However, Watts rejects any association of the qatal with an unreal modality and alleges that "the verb itself is never actually subjunctive" but, on the contrary, "retains the full force of its certainty."28 In his view, the qatal in conditional protases headed by and its varieties, invariably conveys the value of single, finished and certain events.29 In addition to the conditional context, Watts30 detects a "subjunctive" use of the suffix conjugation in optative environments where the gram denotes wishes.31 In such instances - by using the qatal form -, the enunciator expresses his/her desire concerning actions or conditions over which, however, he/she does not assume control.32 Grether33 maintains that the qatal in conditional phrases with introduces conditions that are seen as non-realised or unrealisable (viz. impossible to realise), either in the past or in the present-future. He also observes that the same locution may introduce unreal - that is non-realised or unrealisable - wishes.34 Lambdin35 argues that the qatal appears in contrary-to-fact conditional sentences headed by and providing an analogical counterfactual sense. Additionally, when a subsequent apodosis fails to be overtly expressed, the / + qatal sequence offers an optative value corresponding to the English idiom "would that."36 In a similar vein, Jenni37 observes that the qatal may be found in both conditional and optative clauses, denoting unreal events. On the contrary, Schneider38 - in a brief and general statement - alleges that the qatal with can introduce wishes which are not only unreal but also real. Waltke and O'Connor39 discuss the qatal of the "irreal mood" exclusively at the very end of their description dedicated to the suffix conjugation. The authors maintain that the qatal - a perfective aspect - does not denote mood, either indicative (real) or non-indicative (irreal). In their view, theses notions are conveyed by contextual features. On the other hand, they observe that the gram does appear in certain modal environments, for example in contrary-to-fact conditional sentences and in expressions of a wish that is not expected to be materialised. As Watts,40 Waltke and O'Connor41 allege that the qatal in counterfactual conditional and optative clauses preserves its perfective value even though - they immediately add - such a sense is not evident. Van der Merwe, Naudé and Kroeze42 superficially - and, again, in a section devoted to particles and not to verbal semantics - mention that the subordinating conjunctions and present positive or negative unreal conditions and unlikely situations. Van der Merwe and Naudé43 offer a significantly more exhaustive treatment of the counterfactual qatal, alleging that the qatal with introduces hypothetical conditions and, if the apodosis is left unexpressed, wishes that are impossible to fulfil. In their opinion, the optative use derives from the subordinate conditional (hypothetical). In contrast to the + qatal sequence, the qatal headed by exclusively expresses hypothetical conditions. This means that no overt optative use is detected. Moreover, van der Merwe and Naudé interestingly observe that the sequences with are also found in contexts of oaths and strong assertions. Finally, Cook44 notes that the qatal may be used to introduce counterfactual conditional statements, being preceded by "irreal conditional conjunction[s]" and .45 Cook explains the choice of the qatal - in his classification, a perfective gram - in expressions of counterfactual conditions as being motivated by the temporal distance granted by the inherent perfective-past value of the formation. Namely, the temporal distance - namely the perfective-past sense of the qatal - is used in order to express a lesser degree in actuality.46 It shall be noted that Cook fails to see in the counterfactual qatal a case of the context induced reinterpretation or modal contamination.47 This explanation, according to him, may uniquely be applied to contingent uses, namely to the values offered by the weqatal.48
From the above discussion it is evident that scholars have regularly limited themselves to a simple remark stating that the suffix conjugation may appear in conditional periods headed by the particles or and that it is also able to - when the apodoses is left unexpressed - introduce counterfactual wishes. The widespread opinion is therefore that the former use is more original and the latter arose due to the deletion of the second member of a conditional sequence.49 Furthermore, with a few exceptions,50 no intent has been undertaken in order to establish a semantic relation between the counterfactual qatal and its indicative, as already mentioned, prototypical variety. When scholars do formulate such an opinion and explain the nature of the counter-factual qatal in respect to the remaining more common uses, they regularly minimalise the modal character of the gram. They invariably and à toute force try to accommodate the counterfactual variety within the definition of the qatal they have proposed. Namely, Watts,51 Waltke and O'Connor52 and Cook53 jointly reject any explicitly and properly modal understanding of the counter-factual qatal: they explain this type of the suffix conjugation as a regular manifestation of its inherent core meaning, either perfect (single, finished and certain),54 perfective55 or perfective-past.56
This minimisation of the importance of the counterfactual qatal (i.e. a simple relegation to a contextual use), its non-relatedness to the predominant indicative PPP variety or, on the contrary, a strict derivativeness from an allegedly inherent main meaning of the suffix conjugation render all the descriptions offered thus far unsatisfactory and unacceptable. In order to properly understand these three weak points of the traditional models, we must explain how the meaning of a verbal form shall be represented. In particular, we must demonstrate the following: a) that all senses are to be treated with an equal relevance; b) that they are necessarily connected and hence their representation is required to explicitly specify a type of chaining which links superficially heterogeneous values; and c) that this chaining cannot be achieved by means of a derivation from the inherent meaning because such an inherent meaning fails to be a realistic linguistic phenomenon.
3 Objective and Methodology
While certain senses offered by a given gram may be perceived as peculiar, if compared with other values displayed by the formation, they must somehow be related to the remaining semantic potential of the gram. This necessity -labelled as the relatedness principle of the polysemous components of a form -stems from the fact that different values conveyed by a single grammatical entity are conceptually (and as we will see below, also diachronically) connected.57 Polysemy - a phenomenon which is typical for all components of the grammar - is not a random collection of values. The extension of meaning from one sense in another is obligatorily based upon and governed by universal cognitive mechanisms that inversely impose constraints on the composition of a given polysemous entity.58
Under this cognitive view, the meaning of a form is understood as the gram's entire semantic potential - a potential that can be activated in an immense number of concrete empirical cases. Put differently, the meaning equals a set-theoretical union of all individual atomic senses available in specific contexts. Since all such concrete senses rely on context, the total meaning of a construction is inevitably affected by its contextual settings.59 Inversely, given the fact that the overall meaning - the entire polysemy of a form - is a context-induced phenomenon (i.e. it is composed of atomic contextual senses), individual empirical instances are equally relevant and important for this total meaning of the gram.
All of this implies that a traditional dichotomy between the inherent meaning and contextual variations is removed, being substituted by a more realistic distinction between concrete, specific, empirically recorded senses - the use of a form in a specific place and time - and their summation into a complex referred to as an overall meaning - the total semantic potential.60
How can we represent such a sematic potential of a gram? Cognitive scholars usually employ a model where polysemy and thus the total meaning space of a form is graphically portrayed as a network of interconnected specific senses. In this network, values - elements of the grid - expand by means of human cognitive mechanisms (such as conceptual metaphors and image schema transformations) into novel senses assuring the connection of all components of the map. This means that certain universal conceptual processes enable us to link all senses from more original to those located in the peripheral zone of the network.61 But the relation between a sense and its extension - a novel value arisen due to a given cognitive mechanism - is not only conceptual but also, and in fact necessarily, historical. Namely, a sense from which another value has conceptually been derived is per vim diachronically earlier. This signifies that the chaining between senses must reflect a diachronic progression from original values to advanced ones, progressively more and more distant from the historically initial sense.62 Since polysemy is a synchronic reflection of an evolutionary process, the representation of a given polysemous entity as a network of connected senses - the mapping - shall reproduce this historical expansion.63 Under this view, the chaining of components of the grid - the explanation of the extension from one sense into another - copies a real evolutionary process whereby archaic values gradually give rise to posterior ones.64
Linguistic typology bestows us with a cluster of universal tendencies -or, under a stronger assumption, deterministic laws - that govern the development of grams belonging to certain major types. These evolutionary principles determine, for instance, how aspects, tenses and moods are born and how they develop until their grammatical death.65 This is what has been referred to as paths of grammatical growth. With these common evolutionary scenarios -principally diachronic laws - we can order or connect components of a given synchronic network, that is, of a polysemous meaning. In other words, by employing typological developmental universals, we can establish a relation between the elements of the grid, linking each sense to another. This link per vim represents a change from a historical earlier stage to a posterior one. As a result, we can compare the synchronic variety of senses displayed by a form and arrange it so that it would match a given developmental universal progression.66 The meaning of a gram is thus represented as a diachronically chained network - a portion of a certain path.67
Obeying the principles of relatedness, equality of senses and non-derivability, the present paper aspires to correct the three major inaccuracies of the traditional models previously identified in section A.3. Namely, we shall propose a novel solution to the problem of the counterfactual qatal, portraying this type of the suffix conjugation as equally relevant for the entire meaning of the gram, as conceptually and diachronically connected to the dominant semantic sphere (i.e. to the indicative PPP values) but, on the other hand, without deriving it from the so-called inherent value. More specifically, in light of the cognitive and typological facts introduced above, the elucidation of the relation that necessarily exists between the counterfactual qatal and the remaining semantic potential of the gram, and hence the demonstration of the semantic coherence and homogeneity of the BH suffix conjugation, shall consist in the following. The counterfactual values of the qatal must be chained to the semantic network designed for the entire formation, in particular to its most common domain of an indicative perfect-perfective-past. This chaining must be achieved by employing a typologically plausible path (or cluster of trajectories) that would assure a conceptual and historical connection between the indicative qatal and its counterfactual uses. Furthermore, this chaining procedure should internally correlate and organise different senses of the counterfactual qatal, especially its hypothetical (conditional) and optative subtypes.
Since our explanation is built on empirical contextual cases, in both their analysis and classification, it is evident that any further discussion must be preceded by a detailed description of concrete examples where the counterfactual qatal appears. Therefore, in the following part of the article, we will present all instances where this variety of the suffix conjugation is found, proposing their systematic categorisation.
B COUNTERFACTUAL QATAL - BIBLICAL HEBREW EVIDENCE
1 Clauses with
There are eleven examples where the qatal appears with the particle In two of these instances, the word is spelled as
1a Counterfactual Conditional and Hypothetical
In 4 cases, the qatal headed by appears in counterfactual conditional protases. In such instances, it expresses counterfactual conditions or hypothesis. This counterfactuality signifies that a hypothesised activity is divergent from what actually happened or happens in the real world. When referring to an ongoing state of affairs, the counterfactuality also conveys the ideas of the unlikelihood and/or improbability of achieving an imagined situation. A given condition may, also, be unreal - it refers to the past and hence is impossible to be materialised (3 times).
(1) a. Deut 32:29
If they had been wise, they would have understood this, and would have been able to discern what will happen to them
b. Judg 8:19
If you had saved them alive, I would not kill you
c. Judg 13:23
If the Lord had meant to kill us, he would not have accepted a burnt offering and a grain offering at our hands, or shown us all these things, or now announced to us such things as these.
However, the situation expressed by the qatal in the protasis may likewise be real, referring to a present state of affairs. In this case, a hypothesised event - even though improbable - is still possible to be achieved:
(2) Gen 50:15
If Joseph would bear a grudge against us, he will certainly repay us the harm with had done him.
It shall be noted that in two cases, the qatal appears in the apodosis expressing counterfactual both unreal (1.c) and real (1.b) suppositions.
1b Counterfactual Optative
In 5 instances, the suffix conjugation employed in the same type of sequence fails to appear in overt conditional periods. It such cases, it conveys rather a counterfactual optative sense as opposed to a conditional one. Three times, the counterfactuality is unreal: the gram expresses wishes that are impossible to realise because they concern a past state of affairs. The English equivalent of such an expression can be an optative locution would that + Pluperfect (e.g. would that something had happened) or a periphrasis corresponding to the protasis of an unreal conditional period (e.g. if only it had happened). In both cases, the constructions indicate that the event was contrary to the desired situation (counterfactual) and that it is impossible to change, since it has already occurred (unreal):
(3) a. Num 14:2
Would that we had died in the land of Egypt! Or would that we had died in this wilderness!
b. Num 20:3
Would that we had perished when our brothers perished before the Lord!
c. Josh 7:7
Would that we had been content (If only we had been content) to dwell on the other side of the Jordan!
However, almost equally frequent are examples where the suffix conjugation is employed with the particle in to order express counterfactual and real - that is, referring to a present state of affairs and hence theoretically possible (although unlikely) to achieve - desires or wishes. In this function, the formation approximates the category of a counterfactual real optative:
(4) a. Gen 23:13
If you only will listen to me!
b. Isa 63:19
O that you would tear the heavens and come down
1c Controversial Readings
In certain cases, the two interpretations - namely the conditional (and hence, also a hypothetical one) and optative - are equally possible or arguable. The former represents the entire sentence as an overt conditional period, while the latter understands the first clause as an expression of a counterfactual whish. A close semantic relation of the clause that immediately follows creates a conditional reading where the optative use is "re-analysed" as a hypothetical protasis and the next clause as an apodosis:
(5) a. 1 Sam 14:30
O that the men had only eaten today some of the plunder they took from their enemies; how greater would the slaughter of the Philistines have been!
If the men had eaten today some of the plunder they took from their enemies; would not the slaughter of the Philistines have been even greater?
b. Isa 48:18
O that you had paid attention to my commandments! Then your prosperity would have been like a river, and your success like the waves of the sea
If only you had paid attention to my commands, your prosperity would have been like a river, and your success like the waves of the sea
2 Clauses with
Apart from being found with the particle , the qatal may also be headed by its negative equivalent, namely . This occurs ten times of which four employs the spelling
2a Counterfactual Conditional and Hypothetical
In eight cases, the qatal accompanied by belongs to the negative protasis of an undeniable conditional period and expresses unreal counterfactual negative conditions (cf. examples 6.a-e below as well as 7.a-c):
(6) a. Judg 14.18
If you had not ploughed with my heifer, you would not have found out my riddle
b. Gen 43:10
If we had not delayed, we would now have returned twice c. Isa 1:9
If Lord of hosts had not left us a few survivors, we would have been like Sodom, we would have become like Gomorrah
d. Ps 94:17
If the Lord had not helped me, my soul would soon have lived in the land of silence
e. Ps 124:1-3
If the Lord had not been on our side - let Israel say -
if the Lord had not been on our side when men attacked us,
then they would have swallowed us up alive, when their anger was kindled against us
It shall be observed that on three occasions, the + qatal sequence appears in a clear context of oaths or strong assertions, a fact that certainly approximates the locution to the category of an optative:
(7) a. 2 Sam 2:27
As God lives, if you had not spoken, then only in the morning the army would have stopped pursuing their kinsmen.
b. 1 Sam 25:34
If you had not come quickly to meet me, not one male belonging to Nabal would have been left alive by daybreak.
c. Gen 31:42
If the God of my father, the God of Abraham and the Fear of Isaac, had not been on my side, surely now you would have sent me away empty-handed.
It shall also be noted that the apodoses in all of the quoted examples employ the qatal form with a patent sense of unreal counterfactuality: the gram introduces imaginable counterfactual unreal events and thus functions as a counterfactual unreal hypothetical.
2b Counterfactual Optative
One example may be interpreted as a "false" counterfactual unreal negative optative. Under this reading, the speaker "misleadingly" wishes that a given past event would not have taken place. This wish, however, since "dishonest," offers the force of an admonition directed to the speaker himself. Put differently, the enunciator imagines (wishes) an unreal counterfactual situation in order to show how fatal and disastrous it would have been for himself.
(8) Ps. 27:13
Would that I had not believed / Had I only not believed to experience the goodness of Yahweh in the land of the living!
or in a more optative manner.
Might I only have not believed to experience the goodness of Yahweh in the land of the living! (but I did and therefore God saves me) (cf. Polish overtly optative translations Zebym tylko nie był uwierzył! and Niechze bym tylko nie był uwierzył!)
2c Controversial Readings
In one case, both a conditional and optative interpretation may equally be argued. Namely, the protetic + qatal can be viewed, in harmony with more common examples (cf. 6 and 7 above), as a counterfactual unreal protetic condition, on the accomplishment of which an event expressed in the following apodosis depended. However, it is also possible to understand it in an "optative" manner, namely as a false counterfactual unreal wish. As in example (8), explained in the previous section, the speaker "dishonestly" desires that a past action would not have taken place. This wish due to its "deceitfulness" gives rise to the sense of a warning. It shall be noted that the device to use an unreal counterfactual negative optative of false wishes in order to warn or admonish is not rare in languages. Observe that in Polish the negative past optative is regularly used to wish that something had not happened (though it did in fact happen; "ah, if only it had not happened" - an honest optative sense) or in order to caution ("ah, might you only have not done it!"), insisting on disastrous implications of such a falsely desired act. In fact, in the BH example, this optative reading seems to be more correct because the clause lacks a properly fientive apodosis.
(9) Ps 106:23
Therefore he said he would destroy them if only Moses, his chosen one, had not stood in the breach before him, to turn away his wrath from destroying them
Therefore he said he would destroy them - would that Moses, his chosen one, had not stood Moses in the breach before him, to turn away his wrath from destroying them / might only Moses have not stood ... (cf. Polish overtly optative translations: Żeby tylko Mojżesz nie był stanał w wyłomie! and Nichże by tylko Mojżesz nie był stanqł w wyłomie!)
3 Summing up the BH Evidence
In light of the provided data, we may state that the qatal headed by the particle conveys the meaning of counterfactual, both real and unreal, wishes (optative) and conditions (conditional). When it is preceded by the lexeme , the counterfactual value - again either optative or conditional - is invariably unreal. Furthermore, in various cases, when a conditional protasis contains the sequence of + qatal, the counterfactual apodosis also includes a verb in the qatal form. In such cases, the gram introduces real or unreal hypotheses (hypothetical).
One may detect a certain typological parallelism between the above described counterfactual uses of the qatal and functions of two Latin - proto-typically subjunctive - formations. Namely, the qatal in cases where it appears with and generally matches optative and conditional uses of the Latin past and pluperfect subjunctives. As the Latin past subjunctive (coniunctivus imperfecti; 10.a), it expresses conditional (protetic) and hypothetical (apodotic) counterfactual and real (possible although unlikely) situations. It can also introduce counterfactual real wishes (for the Latin example, see 10.b below). On the other hand, as the pluperfect subjunctive (coniunctivus plusquamperfecti; 10.c), the BH gram denotes counterfactual unreal events in conditional protases and apodoses. However, it may likewise be employed in order to convey counter-factual unreal wishes (for the Latin illustration, see 10.d).68
(10) a. Si id crederes, errares
If you believed (you do not believe but you still could), you would go wrong
b. Utinam veniret!
Would that he were coming (but he is not)
c. Si id credidisses, erravisses
'If you had believed that (but you did not), you would have gone wrong
d. Utinam venisset!
Would that he had come (but he did not)
C COMPARATIVE EVIDENCE
Having described the entire repertory of uses offered by the counterfactual optative-conditional-hypothetical qatal, we shall now introduce certain comparative facts. This evidence will be highly relevant for an appropriate mapping of this type of the BH suffix conjugation as well as for its truthful chaining to the gram's main indicative domain.
The Akkadian language possesses in its verbal system the parsāku gram - a cognate form of the BH qatal. This construction is diachronically less developed than the BH suffix conjugation.69 Namely, the Akkadian parsāku is a resultative proper formation that typically denotes a static situation acquired due to a previously performed activity.70 Aside from this prototypical value, one may detect other senses that correspond to more dynamic uses where the gram approximates the category of a perfect and pluperfect. Additionally, with adjectival roots, the formation indicates current or permanent qualities and, in rare instances, the gram is employed instead of the iptaras and iprus in narrative passages with the force of a present perfect or indefinite past.71 In all of these meanings, the gram offers an indicative character.
However, this Akkadian genetic equivalent of the BH qatal is also commonly found with an injunctive-optative particle lū.72 In such cases, the locution introduces positive wishes or orders, approximating an optative category, labelled as 'precative stative' (in contrast to a fientive precative liprus).73This optative value is real and factual - the accomplishment of a given desire or command is still possible and fully feasible.
(11) a. lū dannātunu
Be strong / May you be strong!74
May they be on their way / let them be on their way! (i.e. be sent)75
May you live!76
The particle lü may also be encountered in counterfactual unreal sentences, introducing counterfactual unreal wishes.77 In such cases, however, it appears with the iprus - a verbal form that most commonly functions as a perfect, perfective and simple past and pluperfect (cf. 12.a-c).78 A given counterfactual wish introduced by lū can likewise refer to a present state of affairs, thus being real. In this case, the particle is accompanied by the morpheme -man and followed by the iparras (cf. 12.d)79 - a present, future and imperfective past gram:80
(12) a. lū īdē
Ich hätte wissen sollen81
Du hättest mich erwarten sollen82
c. šittūta lū ēpuš itâ ša ili lū ētiq
Mag ich Sünde getan haben, die Grenze des Gottes überschritten haben83
d.lūman anāku ammaraššu
Wenn ich ihn doch sehen könnte!84
Additionally, the parsāku could also be employed in order to convey a negative optative sense. Namely when headed by the negative optative particle lā (or in Assyrian lū lā)85 or ē, it regularly introduces real and factual desires (i.e. that something may not occur).86 Thus, it functions as a stative counterpart of the fientive prohibitive lā iparras or vetitive ayyiprus, respectively.87
(13) a. kaspum lā nadin
The silver may / must not be given88
b. la enšēta
do not be weak / you may not be weak89
c. ē naš'āti
Du mögest nicht bringen90
It shall be emphasised that the parsāku is not used with lū and (lū) lā in order to introduce hypotheses. In Akkadian, conditional apodoses are most commonly introduced by the conjunction šumma (factual and real) or šummaman and its variants (counterfactual, both real and unreal). It is interesting to note that in the counterfactual conditional periods (that are semantically equivalent to BH conditionals with or ), the real hypothesis is conveyed by the iparras while the unreal one is expressed by the iprus. However, the parsaku form may also appear in this environment offering both real (if referring to the present) and unreal (if referring to the past) senses:
(14) šumma-min mētāku
If I was dead
If I had been dead91
Some important pieces of evidence may also be encountered in the Arabic language. Classical, as well as Modern Literary, Arabic include in their verbal systems the gram qatala. It is a cognate formation of the BH qatala and hence of the Akkadian parsāku. The qatala is a prototypical present and past perfect, perfective and simple narrative past formation.92 Moreover, as the BH qatal, when derived from adjectival and static roots, the construction indicates situations or qualities, either current or permanent functioning as a present or stative.93
However, besides these prototypical indicative functions, the qatala may also be employed with a counterfactual conditional, hypothetical and optative force. First, when headed by the particle law - a genetic equivalent of the BH and Akkadian lū - the gram sometimes expresses counterfactual real wishes or suggestions approximating an optative (15.a-b).94 The counterfactual unreal wises are also commonly conveyed by the law qatala locution (15.c; sometimes the qatal is additionally accompanied by the completive particle qad or by the auxiliary kana in order to derive an overt pluperfect form, namely kana qatala).95
(15) a. law 'amarta 'umara
I wish you ordered96
b. yā rasūla -llāhi law wassa'ata l-masjida
O apostle of God, if you only could enlarge the mosque97
c.law (kuntu) (qad) fa'altu hādā...
Only if I had acted that way!98
Significantly more common are uses of the law qatala in conditional periods, both in protases and apodoses. In fact, the law qatala expression (also law qad qatala or law kana qatala) is the most straightforward manner to introduce counterfactual (also denominated "unlikely") - either real (16.a-b) or unreal (16.c-d) - conditions and hypothesis.99 It must be noted that although these uses are predominant in the Classical and Literary language, they are diachronically secondary as having arisen from the optative use of the particle law for indicating a wish.100
(16) a. law qad māta 'amïru l-mu'minïna lakad bāia'tu fulānan
If the Prince of the faithful were dead, I would swear allegiance to So-and-so101
b.law kuntu malikan hakamtu bi'adālatan
If I were king, I would rule with justice102
c.law kāna fī-himā ālihatun 'illā -llāhi la-fasadatā
If there had been in them gods besides God, they would have
gone to ruin103
d.law 'akhadhta l-kitāba...wa qara'tahu, la-fahimta ārā'ī fahman tāman
If you had taken the book....and read it, you would have understood my ideas fully.104
In certain types of negative conditional clauses the protases with the qatala may be introduced by law lā - an equivalent to the BH . In such cases, the nominal subject precedes the verb (17.a)105 or is introduced by the particle 'inna (17.b):106
(17) a. lawlā qawmuki hadīthū 'ahdan bi-kufran, la-'assastu l-baiti 'alá qawa'idi 'ibrahïma
Had not your people ceased recently to be in a state of infidelity, I would raise the house on the foundations of Ibrahim107
b. law ' innaka baqīta la-mā jurihta
If only you had remained, you not have been wounded108
D DLACHRONIC AND TYPOLOGICAL EVIDENCE - RECONSTRUCTION
1 PS *Qatal- and its Anterior Path
The entity that appears as qatal in Biblical Hebrew, as parsāku in Akkadian and as qatala in Arabic was not originally modal. More specifically, the BH qatal and its homologues in the other Semitic languages are descents of a resultative intransitive (passive when derived from transitive roots) verbal adjective used with a predicative function.109 This Proto-Semitic (PS) construction - to which we will henceforth refer to as *qatal- - provides a resultative proper sense that is still preserved in Akkadian110 (cf. section C.1 above). Being an exemplary resultative gram, the formation developed in later idioms in accordance with a typologically universal law governing the life of resultative formations, namely the anterior path.
Derived from extensive typological studies, the anterior path stipulates that present resultative proper constructions undergo a regular evolution whereby they are converted into past tenses.111 Of course, this transformation of resultative present inputs into definite past grams is gradual and consists of various intermediate stages. The anterior trajectory codifies the exact order of such consecutive stages that lead from a resultative present to a definite past tense. In that manner, it provides a representation of an entire archetypal grammatical life of resultative constructions. In accordance with the anterior cline, resultative inputs initially develop into present perfects, acquiring successively the following perfect values: inclusive, resultative, frequentative, experiential and indefinite.112 Afterwards they become acceptable in explicit past milieus, giving rise to definite past tenses. Once admissible in an overt past context, the locution increases its temporal distance from the enunciator's here-and-now and develops past functions in the following sequence: immediate, hodiernal, hesternal, recent, general and remote. Furthermore, the past tense uses are first generated in discourse from where they spread to narration. In certain languages, during the conversion of a present perfect into a definite past tense, one may likewise detect a stage where an upcoming past tense offers an explicit aspectual perfective sense. At a posterior stage, such perfective pasts develop into simple past tenses.113
In harmony with the anterior path, in Biblical Hebrew and Arabic, the successors of the PS input construction acquired values that correspond to subsequent developmental stages. In Biblical Hebrew, the qatal attained the phase of a perfect, perfective and simple past.114 In Arabic, the gram even developed a more advance value, namely the sense of a narrative past tense (cf. section C2).
2 Modal Contamination
The anterior path, by itself, does not account for the modal counterfactual senses provided by the BH gram - the cline does not predict any particular stage where a resultative formation could acquire such values.
Modal grams usually have their roots in explicit agentive modal constructions which express ideas of ability, obligation, desire and intention. These four - most commonly analytic - types of constructions represent four starting points of four modal clines (labelled, respectively: ability, obligation, desiderative and intentional path) that jointly specify how moods arise in languages of the world.115 However, since the BH qatal is a descent of a resultative input that failed to convey modal meaning in an explicit manner - more concretely, the PS form did not express any precise idea of ability, obligation, desire or intension -, the modal value of the suffix conjugation and its cognates in other Semitic tongues must have stemmed from "external" factors. Such factors most commonly correspond to modal lexemes (particles) or to syntactically modal contexts (certain types of clauses).116
Linguistic typology teaches us that modality - besides being a result of the four properly modal clines mentioned above - may also develop from originally indicative grams. According to this evolutionary principle, because of their frequent use in overtly modal milieus, indicative inputs gradually develop into grammatical moods. As the anterior path, this process, also, is gradual and consists of several consecutive steps. At the beginning, an original indicative locution starts being commonly employed in an explicit modal environment, providing a given modal sense imposed by this context. Due to its regular use in this milieu, the formation progressively assumes the meaning of the context as its own to the degree where the initially indicative form becomes entirely identified with a modal value generated by its own environment. At this stage, other non-modal uses of the formation are no longer acceptable - the gram is reanalysed as a mood.117 Afterwards, the "new" mood - that received its modal sense from the context where it was originally employed - may
become freed from that milieu and appear in other environments. It can even be used in clauses that do not possess any other overt modal markers. This means that the contextually determined mood may be employed, for instance, in main clauses where, without being accompanied by modal lexemes or used in explicitly modal syntactic structures, it will still provide the modal sense it has previously incorporated.118 This entire process of transforming indicatives into moods (especially into subjunctives) has been referred to as a "modal contamination"119 - a subtype of the common process of "conventionalisation of implicature,"120 "context-induced reinterpretation,"121 or "semanticisation."122 It shall be noted that most commonly, present indicatives develop into (or acquire senses of) real factual modality, present perfects into real factual perfect modality, past tenses into counterfactual real modality, and pluperfects into counterfactual unreal modality (cf. the modalisation of the French present, imperfective past and pluperfect in conditional protases: si tu viens and si tu est venu - real factual; si tu venais - counterfactual real and si tu avais venu -counterfactual unreal; cf. also "optative path" below). Thus, in agreement with the principle of compositionality, the interaction of a modal environment with originally non-modal grams shows that the sense of a contaminated expression is a product of the values of its components. More specifically, any type of modalisation of present, resultative present and present perfect constructions delivers a real factual modal sense. The modal contamination of a definite past generates a counterfactual real mood and the modalisation of a pluperfect triggers the value of a counterfactual unreal mood. Consequently, while the modal sense derives from contextual factors, the exact modal nuance (real factuality, real counterfactuality and unreal counterfactuality) stems from the meaning (i.e. the temporal and taxis load) of an initially non-modal verbal form.
3 PS *Law and Modalisation of the *Qatal-
As demonstrated by numerous examples in section B, the counterfactual qatal is invariable found in clauses with or its negative variety as well as in apodoses of conditional periods whose protases are headed by such particles. The same occurs in Arabic, where the qatala normally displays a counterfactual value in the proximity of the lexeme law.
The BH particle or Arabic law are descents of the PS lexeme *lau.123 As still demonstrated by Akkadian examples (cf. 11.a-c), this word (as well as its negative variant) was not a conditional conjunction. Quite the reverse, it functioned as an optative or voluntative particle.124 Conditional periods were usually headed by šumma (factual) or šumma-man (counterfactual). Furthermore, it was not limited to the counterfactual sense. It could introduce all type of wishes: real, unreal, factual and counterfactual.125 The exact value of a desire was specified by means of a verbal form employed with lū. The iprus introduced counterfactual wishes, the iparras connoted counterfactual real desires and the parsāku conveyed factual real requests or aspirations. This factual reading of the lū parsāku stemmed from a non-advancement of the Akkadian gram that continued to be mainly used as a resultative present or stative present category (cf. C.1). As explained above, presents or present perfects when modalised, regularly offer a factual real meaning. On the contrary, past tenses -simple, durative or pluperfect - commonly develop counterfactual senses.
Also the history of the Arabic language demonstrates that the word law was originally a wish particle. Indeed, the optative use of the periphrasis law qatala was still relatively common in Classical Arabic.126 Nevertheless, in Modern Literary Arabic the law lexeme is most frequently used as a conditional particle introducing counterfactual protases (cf. section C.2).
Consequently, we may affirm that when successors of the PS *qatal-advanced on the anterior cline and developed into past tenses (either discursive as in Biblical Hebrew or narrative as in Arabic), being also commonly employed as pluperfect, the factual and real readings were impossible while, on the contrary, counterfactual values became accessible. Thus, the original factual and real optative periphrasis *law + qatal- acquired counterfactual, real or unreal, values. The former was triggered by the optative past input (i.e. the simple past sense of the BH qatal or Arabic qatala in an optative context).127 while the latter was generated by the optative pluperfect value (i.e. the pluperfect sense of the BH qatal or Arabic qatala in an optative context). Thus, in conformity with the principle of compositionality, the sense of the counterfactual qatal is an aggregate of the values of its components: the modal optative value derives from the modal meaning of the particle *law that contaminated the non-modal resultative proper input. However, the specific type of the modality - namely real and unreal counterfactuality - stems from the sense of the BH successor of the PS *qatal-. The real counterfactual modal value is a product of the modalisation of the past tense qatal (i.e. of the qatal employed with the force of a definite past) while the unreal counterfactual value is a result of the modalisation of the pluperfect qatal (i.e. of the qatal that functions as a pluperfect).
4 Optative Path
It is important to emphasise that the BH qatal in the context of - as well as the Arabic qatala in the vicinity of law - does not behave anymore as a past or pluperfect of the "normal" or non-contaminated anterior path. It does not have the same temporal connotations - most importantly, the qatal in the original past optative sense does not offer a past meaning but, exclusively, a counter-factual real sense - it refers not to a past but to a present situation! The same holds for a pluperfect variety. It is not a genuine past perfect anymore but a counterfactual unreal gram - it refers to the past and not to events that precede other past actions. This change, as already explained, is a common outcome of a modalisation of originally indicative grams (cf. section D.2 above).
In our case, the modalisation is imposed by the overt optative environment. In Akkadian, since the parsaku is still a resultative present or stative present, its modalised output provides a factual real sense. However, in Biblical Hebrew and Arabic, in accordance with the rule of compositionality and due to the fact that the successors of the PS *qatal- frequently function as past tenses and as dynamic pluperfects the modal outcome is counterfactual, both real and unreal. This portion of the development - that harmonises with a regular modal contamination path, as described in section D.2 - explains the optative uses of the BH qatal. But how did the conditional and hypothetical meanings arise?
In the history of Semitic languages, we deal with a complex process that consists of two distinct - although connected - changes. On the one hand, the non-modal input PS *qatal- developed in overtly marked contexts a clear modal sense of real and unreal counterfactuality. On the other, the optative lexeme *law (that imposed a modal value of the context in which the *qatal- was employed and that consequently modally contaminated the original indicative form) developed into a conditional conjunction. This means that in Biblical Hebrew and Arabic, the entire periphrasis began to be employed as a counter-factual protasis, consequently enabling the use of the qatal in counterfactual apodoses. Thus, we face two phenomena: a change from an indicative (or non-modal) into modality and from an optative (sense or gram) into a conditional and hypothetical. This latter development, as the former, is typologically common and may be illustrated by the following Latin and Germanic examples.
According to a widespread opinion, most forms of the Latin present subjunctive are descents of the Proto-Indo-European (PIE) optative in *- įeh1- / - įh1-.128 It is therefore not surprising that the form rogemus, besides being most frequently used in conditional periods (Si habeat "If you should have") or subordinated syntactic milieus (impero ut facies "I order that you do"), could still be found independently, providing an optative sense of a wish (optative proper), command (jussive) or exhortation (hortative): Utinam veniat! "Would that he may come!" Rogemus "let us talk!" or Veniat "may he come!"129 Also the past subjunctive - as demonstrated in section B.3 is commonly used to indicate a counterfactual real mood in protases and apodoses - may have had its roots in original optative forms. Although there is no agreement as for the exact origin, scholars regularly posit a link to the PIE optative.130 Another highly instructive case may be found in the Germanic family. As in Latin, the present subjunctive derives from the PIE optative in *- įeh1- / - įh1-. Even more interestingly, the past subjunctive (a mood of counterfactuality) reflects etymologically the optative of the PIE perfect that, in Germanic, was used not only as a perfect but also as a simple past tense (the Germanic preterite is an amalgamation of the PIE perfect and aorist).131 This means that the optative of a past tense was transformed into a past subjunctive - a counterfactual gram. When Germanic languages developed a system of analytic perfect grams with the verb have, the past subjunctive offered values of real counterfactuality while its homologue with have (i.e. the pluperfect subjunctive) provided a counterfactual unreal sense, as still preserved in Icelandic.
Consequently, it is possible to affirm that optatives may easily be transformed into subjunctives (both as conditionals in protases or as hypotheticals in apodoses).132 Again, counterfactual real optatives (or optatives built on past forms) generate counterfactual unreal conditional and/or hypothetical senses while counterfactual unreal optatives (or optatives built on pluperfect forms) give counterfactual unreal conditional and/or hypothetical values. The entire development of optatives from indicatives and their spread to conditional periods (henceforth referred to as an "optative path") may be schematised as follows:
As previously explained, the most prominent uses of the BH qatal (namely, present and past perfect, perfective and simple past) arose due to the progression of the PS resultative present expression in accordance with the anterior path. Thus, the PPP values of the BH gram correspond to stages of this evolutionary scenario. Namely, they are manifestations of consecutive phases of the anterior path whereby a resultative input evolves towards a past tense. Consequently, the PPP semantic potential of the BH suffix conjugation may be grasped in its totality and represented as a homogeneous and consistent whole, if we classify it as a portion of the anterior trajectory, spanning from the stage of a present perfect to the stage of a simple (although only discursive) past. In other words, by employing this universal evolutionary principle, we can chain various senses offered by the qatal and posit a solid - both conceptually and diachronically plausible - structured network.134
The counterfactual value of the BH qatal - composed itself of six basic senses (counterfactual real optative, counterfactual unreal optative, counter-factual real conditional, counterfactual unreal conditional, counterfactual real hypothetical and counterfactual unreal hypothetical) - may be networked as subsequent phases of two typologically plausible processes: a contamination path (a modalisation of the PS anterior-path gram *qatal-, imposed by the PS optative particle *law and its varieties) and an optative path (the spread from optative contexts to conditional periods).
More specifically, while the PS *qatal- normally developed along the anterior path, in the environment of the optative *law, it additionally acquired modal functions. The modal contamination by *law explains the optative sense of the BH qatal - a sense that, in harmony with the progression of the fientive entity along the anterior path, was transmuted from real and factual (as in Proto-Semitic and Akkadian) into counterfactual real and unreal. The modal contamination and acquisition of the counterfactual values (possible only because the BH suffix conjugation developed into a discursive past and pluperfect) conceptually and diachronically link the counterfactual qatal to its PPP variety. Again, the real and unreal counterfactual varieties of the modal optative meaning are results of the compositionality principle. While the modal particle introduced the modal optative sense, the qatal - developing along the anterior path - is responsible for the concrete values of this semantic domain. To be exact, when modalised, the past time qatal gave real counterfactuality while its pluperfect subtype produced unreal counterfactuality. On the other hand, the conditional and hypothetical senses may be chained to the optative value as corresponding to a further evolution of the counterfactual optative. Namely, in accordance with the optative cline, the PS optative construction developed certain subjunctive functions, being reanalysed as a counterfactual mood in protases and apodoses and hence giving rise to conditional and hypothetical uses, respectively. The entire networking of the counterfactual qatal and its chaining to the PPP qatal may be represented visually in the following manner:
Our study has demonstrated that the counterfactual value displayed by the BH qatal is a rightful component of the total meaning of the suffix conjugation understood as a network of conceptually and historically connected senses. The chaining procedure built on the framework of universal paths (a theory of highly plausible evolutionary scenarios) has enabled us to correlate all the six specific senses of the counterfactual domain, establishing a diachronic and synchronic (i.e. conceptual) order among them. Furthermore, the entire counter-factual block has been linked to the dominant indicative type of the qatal and its perfect, perfective and past values. Our chaining procedure has also liberated us from interpreting the counterfactual qatal as a derivation from its -supposedly - inherent taxis, aspectual or temporal value. The posited mapping mechanism conceptually and diachronically associates a sense with another: it designs a path leading from the meaning included in the proto-input to all possible later - superficially incoherent and heterogeneous - extensions. This evolutionary link both theoretically and empirically guarantees the solidness of the established network.
The results of the present research also indicate that - contrary to widespread opinion - the optative use of the BH qatal does not reflect a loss or omission of the apodosis. There is no need to posit any non-expressed apodosis in order to elucidate the optative sense provided by the gram. Quite the reverse, as explained, the optative value constituted the original modal meaning the PS *qatal- had acquired in the vicinity of the particle *law. The BH optative examples are thus remnants of such an ancient usage.
Finally our study provides further typological evidence for a diachronic and conceptual split of an originally resultative gram in an indicative (a variety that develops along the anterior path) and a mood (a variety that evolves following the modal contamination path). A typologically similar scenario has been posited for the BH "short" yiqtol morphology preserved in the shape of the jussive yiqtol and the wayyiqtol.136 This article shows that this type of evolution may have occurred in the Semitic family twice.
Aikhenvald, Alexandra. Evidentiality. Oxford: Oxford University Press, 2004. [ Links ] Andersen, David. "The Evolution of the Hebrew Verbal System." Zeitschrift für Althebräistik 13/1 (2000): 1-66. [ Links ]
Andrason, Alexander. "The Panchronic Yiqtol: Functionally Consistent and Cognitively Plausible." Journal of Hebrew Scriptures 10/10 (2010): 1-63. [ Links ]
____. "The 'Guessing' QATAL - the BH Suffix Conjugation as a Manifestation of the Evidential Trajectory." Journal for Semitics 19/2 (2010): 603-27. [ Links ]
____. "The Akkadian Iprus from the Unidirectional Perspective." Journal of Semitic Studies 55/2 (2010): 325-45. [ Links ]
____. Qatal, yiqtol, weqatal y wayyiqtol: Modelo pancrónico del sistema verbal de la lengua hebrea bíblica con el análisis adicional de los sistemas verbales de las lenguas acadia y árabe. Madrid: Publicaciones de Universidad Complutense, 2011. [ Links ]
____. "The Biblical Hebrew Verbal System in Light of Grammaticalization -the Second Generation." Hebrew Studies 52 (2011): 351-83. [ Links ]
____. "The Biblical Hebrew Wayyiqtol - a Dynamic Definition." Journal of Hebrew Scriptures 11/7 (2011): 1-50. [ Links ]
____. "The BH Weqatal: A Homogenous Form with no Haphazard Functions, Part 1." Journal of Northwest Semitic Languages 37/2 (2011): 1-25. [ Links ]
____. "The BH Weqatal: A Homogenous Form with no Haphazard Functions, Part 2." Journal of Northwest Semitic Languages 38/1 (2012): 1-30. [ Links ]
____. "The Dynamic Short Yiqtol." Journal for Semitics 21/2 (2012): 308-339. [ Links ]
____. "Future Values of the Qatal are "Logical" - how to Chain Future Senses of the Qatal to its Semantic Network?" Hebrew Studies 54 (forthcoming). [ Links ] Barnes, Oswald L. A New Approach to the Problem of the Hebrew Tenses. Oxford: Thornton, 1965. [ Links ]
Bauer, Hans and Pontus Leander. Historische Grammatik der hebräischen Sprache des Alten Testamentes. Hildesheim: Georg Olms, 1962. [ Links ]
Blau, Joshua. "Marginalia Semitica I: The Problem of Tenses in Biblical Hebrew." Israel Oriental Studies 1 (1971): 1-35. [ Links ]
Brockelmann, Carl. Grundriss der vergleichenden Grammatik der semitischen Sprachen. Hildesheim: Georg Olms, 1961. [ Links ]
Bybee, Joan, Revere Perkins and William Pagliuca. The Evolution of Grammar.Chicago: The University of Chicago Press, 1994. [ Links ]
Cohen, Marcel. Le Sytème Verbal Sémitique et l'expression du temps. Paris: Ernest Leroux, 1924. [ Links ]
Cook, John. "The Biblical Hebrew Verbal System." PhD diss., University of Wisconsin-Madison, 2002. [ Links ]
Couper-Kuhlen, Elizabeth and Margret Selting. "Introducing Interactional Linguistics." Pages 1-22 in Studies in Interactional Linguistics. Edited by Margret Selting and Elizabeth Couper-Kuhlen. Amsterdam: John Benjamins, 2001. [ Links ]
Croft, William and Alan Cruse. Cognitive Linguistics. Cambridge: Cambridge University Press, 2004. [ Links ]
Dahl, Östen. Tense and Aspect Systems. Oxford: Blackwell, 1985. [ Links ]
____. "The Tense and Aspect Systems of European Languages in a Typological Perspective." Pages 3-25 in Tense and Aspect in the Languages of Europe. Edited by Östen Dahl. Berlin/New York: Mouton de Gruyter, 2000. [ Links ]
____, ed. Tense and Aspect in the Languages of Europe. Berlin/New York: Mouton de Gruyter, 2000. [ Links ]
Danecki, Janusz. Gramatyka języka arabskiego. Warszawa: Wydawnictwo Akademickie Dialog, 1994. [ Links ]
Davidson, Andrew B. Hebrew Syntax. Edinburgh: T & T Clark, 1902. [ Links ]
DeCaen, Vincet. "Ewald and Driver on Biblical Hebrew 'Aspect.'" Zeitschrift für Althebraistik 9 (1999): 129-51. [ Links ]
Driver, Samuel R. A Treatise on the Use of the Tenses in Hebrew and some Other Syntactical Problems. Oxford: Clarendon, 1892. [ Links ]
Evans, Vyvyan and Melanie Green. Cognitive Linguistics: An Introduction. Edinburgh: Edinburgh University Press, 2006. [ Links ]
Francis, Augustus L. and Henry Tatum. An Advanced Latin Syntax. Cambridge: Cambridge University Press, 1919. [ Links ]
Gesenius, Wilhelm. Hebräische Grammatik. Halle: Renger, 1813-1842. [ Links ]
Gesenius, Wilhelm, Emil Kautsch and Arthur E. Cowley. Gesenius' Hebrew Grammar. Oxford: Clarendon Press, 1909. [ Links ]
Gildersleeve, Basil L. and Gonzalez Lodge. Gildersleeve's Latin Grammar. London: Macmillan and Co., 1895. [ Links ]
Gray, Louis. Introduction to Semitic Comparative Linguistics. Amsterdam: Philo Press, 1971. [ Links ]
Grether, Oskar. Hebräische Grammatik für den akademischen Unterricht. München: Claudius Verlag, 1967. [ Links ]
Hatav, Galia. The Semantics of Aspect and Modality: Evidence from English and Biblical Hebrew. Amsterdam/ Philadelphia: John Benjamins, 1997. [ Links ]
Haywood, John and Hayim M. Nahmad. A New Arabic Grammar. London: Lund Humphries, 1965. [ Links ]
Heine, Bernd, Ulrike Claudi and Friederike Hünnemeyer. Grammaticalization: A Conceptual Framework. Chicago: University of Chicago Press, 1991. [ Links ]
Heine, Bernd and Tania Kuteva. The Genesis of Grammar: A Reconstruction. Oxford: Oxford University Press, 2007. [ Links ]
Helasvuo, Marja-Liisa. "Emergent Grammar." Pages 66-73 in Grammar, Meaning and Pragmatics. Edited by Frank Brisard, Jan-Ola Östman and Jef Verschueren. Amsterdam: John Benjamins, 2009. [ Links ]
Hopper, Paul and Elisabeth Traugott. Grammaticalization. Cambridge: Cambridge University Press, 2003. [ Links ]
Huehnergard, John. "Assertive *la and Hypothetical *lu/law in Semitic." Journal of American Oriental Society 103/3 (1983): 569-593. [ Links ]
____. "Stative, Predicative Form, Pseudo-Verb." Journal of Near Eastern Studies 46/3 (1987): 215-32 [ Links ]
___. A Grammar of Akkadian. Winona Lake: Eisenbrauns, 2005. [ Links ]
Jasanoff, Jay. "The Origin of the Italic Imperfect Subjunctive." Historische Sprachforschung 104 (1991): 84-105. [ Links ]
Jenni, Ernst. Lehrbuch der hebraischen Sprache des Alten Testaments. Basel: Helbing & Lichtenhavn, 1978. [ Links ]
Joosten, Jan. "The long Form of the Prefix Conjugation Referring to the Past in Biblical Hebrew Prose." Hebrew Studies 40 (1999): 15-26. [ Links ]
Joüon, Paul. Grammaire de l'hébreu biblique. Roma: Institute Biblique Pontifical, 1923. [ Links ]
Kienast, Burkhart. Historische Semitische Sprachwissenschaft. Wiesbaden: Harrassowitz Verlag, 2001. [ Links ]
Kurylowicz, Jerzy. Studies in Semitic Grammar and Metric. Wroclaw: Zaklad Ossolińskich PAN, 1972. [ Links ]
Lambdin, Thomas O. Introduction to Biblical Hebrew. New York: Scribner, 1971. [ Links ]
Lambdin, Thomas O. and John Huehnergard. The Historical Grammar of Classical Hebrew: An Outline. Cambridge: Unpublished course hand-out, 1998. [ Links ]
Lewandowska-Tomaszczyk, Barbara. "Polysemy, Prototypes, and Radical Categories." Pages 139-69 in Handbook of Cognitive Linguistics. Edited by Dirk Geeraerts and Hubert Cuyckens. Oxford: Oxford University Press, 2007. [ Links ]
Lipiński, Edward. Semitic Languages Outline of a Comparative Grammar.
Leuven: Uitgeverij Peeters en Departement Oosterse Studies, 2001.
Loesov, Sergey. "T-Perfect in Old Babylonian: The Debate and a Thesis." Pages 83-181 in Babel and Bibel. Edited by Leonid Kogan. Moscow: Russian State University for the Humanities, 2005. [ Links ]
Loprieno, Antonio. Das Verbalsystem im Äyptischen und Semitischen: Zur Grundlegung einer Aspekttheorie. Wiesbaden: Harrassowitz, 1986. [ Links ]
McFall, Leslie. The Enigma of the Hebrew Verbal System. Sheffield: Almond Press, 1982. [ Links ]
Nedjalkov, Vladimir. "Resultative Constructions." Pages 928-940 in Language Typology and Language Universals: An International Handbook. Vol. 2. Edited by Martin Haspelmath, Ekkehard König, Wulf Oesterreicher and Wolfgang Raible. Berlin: Walter de Gruyter, 2001. [ Links ]
Nedjalkov, Vladimir and Sergey Jaxontov. "The Typology of Resultative Constructions." Pages 30-63 in Typology of Resultative Constructions. Edited by Vladimir Nedjalkov. Amsterdam/Philadelphia: John Benjamins, 1988. [ Links ]
Nikiforidou, Kiki. "Constructional analysis." Pages 16-32 in Grammar, Meaning and Pragmatics. Edited by Fran Brisard, Jan-Ola Östman and Jef Verschueren. Amsterdam/Philadelphia: John Benjamins, 2009. [ Links ]
O'Leary, De Lacy. Comparative Grammar of the Semitic Languages. Amsterdam: Philo Press, 1969. [ Links ]
Peled, Yishai. Conditional Structure in Classical Arabic. Wiesbaden: Harrasowitz, 1992. [ Links ]
Rattray, Susan. "The Tense-Mood-Aspect System of Biblical Hebrew with Special Emphasis on 1 and 2 Samuel." PhD diss., University of California, Berkley, 1992. [ Links ]
Schneider, Wolfgang. Grammatik des biblischen Hebraisch: Ein Lehrbuch. Munich: Claudius Verlag, 1982. [ Links ]
Sonnenschein, Edward. A New Latin Grammar. Oxford: Clarendon Press, 1912. [ Links ]
Squartini, Mario and Pier M. Bertinetto. "The Simple and Compound Past in Romance Languages." Pages 403-439 in Tense and Aspect in the Languages of Europe. Edited by Östen Dahl. Berlin: Mouton de Gruyter, 2000. [ Links ]
Tyler, Andrea and Vyvyan Evans. The Semantics of English Prepositions: Spatial Scenes, Embodied Meaning, and Cognition. Cambridge: Cambridge University Press, 2003. [ Links ]
Van der Auwera, Johan and Volker Gast. "Categories and Prototypes." Pages 166189 in The Oxford Handbook of Typology. Edited by Jae Jung Song. Oxford: Blackwell, 2011. [ Links ]
Van der Merwe, Christo and Jackie Naudé. A Biblical Hebrew Reference Grammar. Rev. ed., forthcoming. [ Links ]
Van der Merwe, Christo, Jackie Naudé and Jan Kroeze, A Biblical Hebrew Reference Grammar. Sheffield: Sheffield Academic Press, 1999. [ Links ]
Von Ewald, Heinrich. Ausführliches Lehrbuch der hebräischen Sprache des alten Bundes. Göttingen: Verlag der dieterichschen Buchhandlung, 1870. [ Links ]
____. Syntax of the Hebrew Language. Edinburgh: T & T Clark, 1879. [ Links ]
Von Soden, Wolfram. Grundriss der akkadishcen Grammatik. Rome: Pontificium Institutum Biblicum, 1952. [ Links ]
Watts, J. Wash. A Survey of Syntax in Hebrew Old Testament. Grand Rapids: Eerdmans, 1951. [ Links ]
Waltke, Bruce K. and Michael O'Connor. An Introduction to Biblical Hebrew Syntax. Winona Lake, Ind.: Eisenbrauns, 1990. [ Links ]
Weingreen, Jacob. A Practical Grammar of Classical Hebrew. Oxford: Clarendon Press, 1939. [ Links ]
Wright, William. A Grammar of the Arabic Language. Vol. 2. Cambridge: Cambridge University Press, 1964. [ Links ]
Zevit, Ziony. "Talking Funny in Biblical Hebrew and Solving a Problem of the Yaqtúl Past Tense." Hebrew Studies 29 (1988): 25-33. [ Links ]
Zuber, Beat. Das Tempussystem des biblischen Hebraisch. Berlin: de Gruyter, 1986. [ Links ]
Department of Ancient Studies
Private Bag XI
7602 Matieland, South Africa
1 This article is a result of the postdoctoral research project carried out by myself in co-operation with Professor Christo van der Merwe in the Department of Ancient Studies (University of Stellenbosch). I would like to thank Professor van der Merwe for encouragement in writing this article and for his invaluable suggestions on my linguistic ideas.
2 Leslie McFall, The Enigma of the Hebrew Verbal System (Sheffield: Almond Press, 1982), 186-187.
3 Of course, theses digits shall be understood only approximately, i.e. as illustrations of certain general tendencies. Most importantly, McFall (Enigma, 186-187) does actually not analyse the meaning of the BH verbal forms but refers to the translation equivalents of the qatal in the Revised Standard Version. The terms "qatal" and "suffix conjugation" will be employed as synonyms. It shall be noted that they do not refer to the weqatal which is treated as an entirely separate category.
4 The term gram is used as synonymous to "grammatical form," "grammatical formation," "grammatical construction" etc.
5 See, for instance, Marcel Cohen, Le Système Verbal Sémitique et l'expression du temps (Paris: Ernest Leroux, 1924), 10-16; Jerzy Kuryłowicz, Studies in Semitic Grammar and Metric (Wroclaw: Zakład Ossolińskich PAN, 1972), 80-82; and, to an extent, John Cook, "The Biblical Hebrew Verbal System" (PhD diss., University of Wisconsin-Madison, 2002), 269-272.
6 Compare Heinrich von Ewald, Ausführliches Lehrbuch der hebräischen Sprache des alten Bundes (Göttingen: Verlag der dieterichschen Buchhandlung, 1870), 349; Heinrich von Ewald, Syntax of the Hebrew Language (Edinburgh: T & T Clark, 1879), 1-3; Samuel R. Driver, A Treatise on the Use of the Tenses in Hebrew and some Other Syntactical Problems (Oxford: Clarendon, 1892), 1-5, 13-26; Andrew B. Davidson, Hebrew Syntax (Edinburgh: T & T Clark, 1902), 58-63; GKC: 309-313; Paul Joüon, Grammaire de l'hébreu biblique (Roma: Institute Biblique Pontifical, 1923), 289-292, 294-297; J. Wash Watts, A Survey of Syntax in Hebrew Old Testament (Grand Rapids: Eerdmans, 1951), 12-32; Bruce K. Waltke and Michael O'Connor, An Introduction to Biblical Hebrew Syntax (Winona Lake, Ind.: Eisen-brauns, 1990), 479-480, 486; Christo van der Merwe, Jackie Naudé and Jan Kroeze, A Biblical Hebrew Reference Grammar (Sheffield: Sheffield Academic Press, 1999), 68-70, 143-147; David Andersen, "The Evolution of the Hebrew Verbal System," ZAH 13/1 (2000): 1-66; and Cook, "The Biblical Hebrew Verbal System," 269-272.
7 See, for instance, early editions of Gesenius' grammar: Wilhelm Gesenius, Hebräische Grammatik (Halle: Renger, 1813-1842), 203-204. Compare also Joüon, Grammaire, 290-291, 296 and Jacob Weingreen, A Practical Grammar of Classical Hebrew (Oxford: Clarendon Press, 1939), 56. In more recent times, see Oswald L. Barnes, A New Approach to the Problem of the Hebrew Tenses (Oxford: Thornton, 1965), 7; Joshua Blau, "Marginalia Semitica I: The Problem of Tenses in Biblical Hebrew," IOSt 1 (1971): 24-26; Beat Zuber, Das Tempussystem des biblischen Hebräisch (Berlin: de Gruyter, 1986), 27; and Ziony Zevit, "Talking Funny in Biblical Hebrew and Solving a Problem of the Yaqtúl Past Tense" HS 29 (1988): 25-33.
8 The remaining 0.5% corresponds to so-called non-verbal uses (cf. McFall, Eningma, 186-187).
9 See already GKC: 313; Joüon, Grammaire, 304-307; Watts, Survey, 12-13, 33, 48; Waltke and O'Connor, Introduction, 493, 496-502, Van der Merwe, Naudé and Kroeze, Biblical Hebrew, 144, 148-149; and Cook, "Verbal System," 271.
10 Antonio Loprieno, Das Verbalsystem im Ägyptischen und Semitischen: Zur Grundlegung einer Aspekttheorie (Wiesbaden: Harrowitz, 1986), 110.
11 Susan Rattray, "The Tense-Mood-Aspect System of Biblical Hebrew with Special Emphasis on 1 and 2 Samuel" (PhD diss., University of Califonia, Berkley, 1992), 149-150.
12 Zuber, Tempussystem, 27.
13 Jan Joosten, "The long Form of the Prefix Conjugation Referring to the Past in Biblical Hebrew Prose," HS 40 (1999): 15-26 (see especially page 16).
14 Galia Hatav, The Semantics of Aspect and Modality: Evidence from English and Biblical Hebrew (Amsterdam/Philadelphia: John Benjamins, 1997), 29.
15 Vincet DeCaen, "Ewald and Driver on Biblical Hebrew 'Aspect,'" ZAH 9 (1999): 129-151 (see especially page 124).
16 In this paper, we will label the modal meaning offered in conditional protases as 'conditional' while to the value found in apodoses we will refer to as "hypothetical."
17 Cf. Driver, Treatise, 179-181; Joüon, Grammaire, 501; Ernst Jenni, Lehrbuch der hebraischen Sprache des Alten Testaments (Basel: Helbing & Lichtenhavn, 1978), 264; Van der Merwe, Naudé and Kroeze, Biblical Hebrew, 303-304.
18 Driver, Treatise, 179-180.
19 Driver, Treatise, 180-181.
20 GKC: 313.
21 GKC: 495.
22 22GKC: 313.
23 Davidson, Hebrew Syntax, 179-182.
24 Davidson, Hebrew Syntax, 183.
25 Joüon, Grammaire, 501, 515.
26 Watts, Survey, 47.
27 Watts, Survey, 48.
28 Watts, Survey, 47.
29 Watts, Survey, 48.
30 Watts, Survey, 57-58.
31 Watts, Survey, 57-58.
32 Watts, Survey, 58.
33 Oskar Grether, Hebräische Grammatik für den akademischen Unterricht (München: Claudius Verlag, 1967), 237.
34 Grether, Hebräische Grammatik, 233.
35 Thomas O. Lambdin, Introduction to Biblical Hebrew (New York: Scribner, 1971), 278.
36 Lambdin, Introduction, 279.
37 Jenni, Lehrbuch, 264.
38 Wolfgang Schneider, Grammatik des biblischen Hebraisch: Ein Lehrbuch (Munich: Claudius Verlag, 1982), 228.
39 Waltke and O'Connor, Introduction, 493-494.
40 Watts, Survey, 1951.
41 Waltke and O'Connor, Introduction, 493.
42 Van der Merwe, Naudé and Kroeze, Biblical Hebrew, 303-304.
43 Christo van der Merwe and Jackie Naudé, A Biblical Hebrew Reference Grammar (Rev. ed., forthcoming).
44 Cook, "Verbal System," 226-227.
45 Cook, "Verbal System," 226.
46 Cook, "Verbal System," 227.
47 On theses notions, see below in this paper.
48 Cook, "Verbal System," 226-227.
49 Cf. Driver, Treatise; Lambdin, Introduction; and Van der Merwe, Naudé and Kroeze, Biblical Hebrew.
50 Cf. Watts, Survey; and Cook, "Verbal System"; as well as, although only partly, Waltke and O' Connor, Introduction.
51 Watts, Survey, 1951.
52 Waltke and O' Connor, Introduction.
53 Cook, "Verbal System."
54 Watts, Survey, 12.
55 Waltke and O'Connor, Introduction, 493.
56 Cook, "Verbal System," 269-27. Watts (Survey) and Waltke and O'Connor (Introduction) argue that the qatal is never a "subjunctive" or an irrealis, respectively. Cook ("Verbal System"), on the other hand, alleges that the counterfactual use of the suffix conjugation stems from its temporal past sense.
57 Barbara Lewandowska-Tomaszczyk, "Polysemy, Prototypes, and Radical Categories," in Handbook of Cognitive Linguistics (ed. Dirk Geeraerts and Hubert Cuyckens; Oxford: Oxford University Press, 2007), 139-169. See especially page 140.
58 Vyvyan Evans and Melanie Green, Cognitive Linguistics: An Introduction (Edinburgh: Edinburgh University Press, 2006), 169-170 and 331-333.
59 Evans and Green, Cognitive, 352-353, 368 and Kiki Nikiforidou, "Constructional Analysis," in Grammar, Meaning and Pragmatics (ed. Frank Brisard, Jan-Ola Östman and Jef Verschueren; Amsterdam: John Benjamins, 2009), 16-32. See especially pages 17 and 26.
60 Cf. Östen Dahl, "The Tense and Aspect Systems of European Languages in a Typological Perspective," in Tense and Aspect in the Languages of Europe (ed. Östen Dahl; Berlin: Mouton de Gruyter, 2000), 3-25. See especially page 14. See also the following references: Elizabeth Couper-Kuhlen and Margret Selting, "Introducing Interactional Linguistics," in Studies in Interactional Linguistics (ed. Margret Selting and Elizabeth Couper-Kuhlen; Amsterdam: John Benjamins, 2001), 1-22 (see especially pages 4-5); William Croft and Alan Cruse, Cognitive Linguistics (Cambridge: Cambridge University Press, 2004), 258; Nikiforidou, "Constructional," 16; and Marja-Liisa Helasvuo, "Emergent Grammar," in Grammar, Meaning and Pragmatics (ed. Frank Brisard, Jan-Ola Östman and Jef Verschueren; Amsterdam: John Benjamins, 2009), 66-73 (see especially pages 70-72).
61 Evans and Green, Cognitive, 331-333.
62 Andrea Tyler and Vyvyan Evans, The Semantics of English Prepositions: Spatial Scenes, Embodied Meaning, and Cognition (Cambridge: Cambridge University Press, 2003), 344-346.
63 Lewandowska-Tomaszczyk, "Polysemy," 140.
64 Johan van der Auwera and Volker Gast, "Categories and Prototypes," in The Oxford Handbook of Typology (ed. Jae Jung Song; Oxford: Blackwell, 2011), 166189. Consult especially pages 186-188.
65 Joan Bybee, Revere Perkins and William Pagliuca, The Evolution of Grammar (Chicago: The University of Chicago Press, 1994); Östen Dahl, ed., Tense and Aspect in the Languages of Europe (Berlin/New York: Mouton de Gruyter, 2000); and Alexander Andrason, Qatal, yiqtol, weqatal y wayyiqtol: Modelo pancrónico del sistema verbal de la lengua hebrea bíblica con el análisis adicional de los sistemas verbales de las lenguas acadia y árabe (Madrid: Publicaciones de Universidad Complutense, 2011).
66 Cf. Alexander Andrason, "The Panchronic Yiqtol: Functionally Consistent and Cognitively Plausible," JHScr 10/10 (2010): 1-63; Alexander Andrason, "The Biblical Hebrew Verbal System in Light of Grammaticalization - the Second Generation," HS 52 (2011): 351-383; and Alexander Andrason, "The Biblical Hebrew Wayyiqtol - a Dynamic Definition," JHScr 11/7 (2011): 1-50.
67 Cf. Van der Auwera and Gast, "Categories," 186-188, 281 and Andrason, "Pan-chronic Yiqtol," 22; Andrason, Qatal, 69-73; and Andrason, "Wayyiqtol," 30-31. It must be observed that this path employed in order to explain a given concrete polysemy is both universal and realistic: universal since typologically plausible, but realistic because supposed to copy a real evolution of the gram under analysis.
68 Cf. Basil L. Gildersleeve and Gonzalez Lodge, Gildersleeve's Latin Grammar (London: Macmillan and Co., 1895), 385. Cf. a similar, although not duplicate, idea already in GKC: 313. These authors, however, consider only the link to conditional senses of the two Latin tenses.
69 This means that the Akkadian gram corresponds to a less advance stage of the development of an original Proto-Semitic input expression.
70 Wolfram von Soden, Grundriss der akkadishcen Grammatik (Rome: Pontificium Institutum Biblicum, 1952), 101 and John Huehnergard, A Grammar of Akkadian (Winona Lake: Eisenbrauns, 2005), 219-223.
71 Sergey Loesov, "T-Perfect in Old Babylonian: The Debate and a Thesis," in Babel and Bibel (ed. L. Kogan; Moscow: Russian State University for the Humanities, 2005), 83-181 (see especially pages 133-134) and Andrason, Qatal, 197.
72 Von Soden, Grundriss, 105, 176-177 and Huehnergard, A Grammar, 223.
73 Von Soden, Grundriss, 105.
74 Huehnergard, A Grammar, 223.
75 Huehnergard, A Grammar, 223.
76 Von Soden, Grundriss, 105.
77 Von Soden, Grundriss, 211.
78 Huehnergard, A Grammar, 119, 158 and 284.
79 Von Soden, Grundriss, 208.
80 Huehnergard, A Grammar, 98-99.
81 Von Soden, Grundriss, 206.
82 Von Soden, Grundriss, 206.
83 Von Soden, Grundriss, 211.
84 Von Soden, Grundriss, 208.
85 Von Soden, Grundriss, 106.
86 Huehnergard, A Grammar, 223.
87 Huehnergard, A Grammar, 146-147.
88 Huehnergard, A Grammar, 223.
89 Huehnergard, A Grammar, 223.
90 Von Soden, Grundriss, 107.
91 Von Soden, Grundriss, 214.
92 This means that the Arabic gram is slightly more advanced than the BH formation. It is used not only as a past tense in discourse but also in narration - a function that in Biblical Hebrew is carried out by the wayyiqtol.
93 Cf. William Wright, A Grammar of the Arabic Language (vol. 2; Cambridge: Cambridge University Press, 1964), 1-18, Janusz Danecki, Gramatyka języka arab-skiego (Warszawa: Wydawnictwo Akademickie Dialog, 1994), 153-154, Burkhart Kienast, Historische Semitische Sprachwissenschaft (Wiesbaden: Harrassowitz Ver-lag, 2001), 332 and Andrason, Qatal, 213-223.
94 Wright, A Grammar, 8-9, 347 and Yishai Peled, Conditional Structure in Classical Arabic (Wiesbaden: Harrasowitz, 1992), 38 and 53.
95 Peled, Conditional, 53-56.
96 Peled, Conditional, 53.
97 Peled, Conditional, 53.
98 Peled, Conditional, 54.
99 Wright, A Grammar, 6-8, 347; John Haywood and Hayim M. Nahmad, A New Arabic Grammar (London: Lund Humphries, 1965), 290-291; and Peled, Conditional, 37-40.
100 Peled, Conditional, 37.
101 Wright, A Grammar, 7.
102 Haywood and Nahmad, Arabic Grammar, 290.
103 Wright, A Grammar, 6.
104 Haywood and Nahmad, Arabic Grammar, 291.
105 Wright, A Gramamr, 6-7.
106 Haywood and Nahmad, Arabic Grammar, 291.
107 Wright, A Grammar, 7.
108 Haywood and Nahmad, Arabic Grammar, 291.
109 Huehnergard, A Grammar, 221-223; Andersen, "The Evolution," 31, Thomas O. Lambdin and John Huehnergard, The Historical Grammar of Classical Hebrew: An Outline (Cambridge: Unpublished course hand-out, 1998); Edward Lipinski, Semitic Languages Outline of a Comparative Grammar (Leuven/Paris/Sterling: Uitgeverij Peeters en Departement Oosterse Studies, 2001), 336-337 and 341; Cook, "Verbal System," 209-219; Andrason, Qatal, 286-289; Alexander Andrason, "The BH Weqatal: A Homogenous Form with no Haphazard Functions, Part 2," JNSL 38/1 (2012): 1-30 (see page 3).
110 John Huehnergard, "Stative, Predicative Form, Pseudo-Verb," JNES 46/3 (1987): 215-232 (see especially pages 223).
111Vladimir Nedjalkov and Sergey Jaxontov, "The Typology of Resultative Constructions," in Typology of Resultative Constructions (ed. V. Nedjalkov; Amsterdam/Philadelphia: John Benjamins, 1988), 3-63; Bybee, Perkins and Pagliuca, Evolution; Dahl, Tense and Aspect; Vladimir Nedjalkov, "Resultative Constructions," in Language Typology and Language Universals: An International Handbook (vol. 2; eds. M. M. Haspelmath, E. König, W. Oesterreicher and W. Raible; Berlin: Walter de Gruyter, 2001), 928-940; Alexander Andrason, "The Akkadian Iprus from the Unidirectional Perspective," JSS 55/2 (2010): 325-345; Andrason, Qatal, 35-38; and Andrason, "Wayyiqtol," 11-13.
112 Fort an explanation of these labels, see Andrason, "Wayyiqtol," 11-13.
113 Compare Bybee, Perkins and Pagliuca, Evolution, 55-57, 98, 104-105; Mario Squartini and Pier M. Bertinetto, "The Simple and Compound Past in Romance Languages," in Tense and Aspect in the Languages of Europe (ed. Ö. Dahl; Berlin/New York: Mouton de Gruyter, 2000), 403-439 (see especially pages 406-407, 414-417 and 422). See also Dahl, "The Tense and Aspect," 15; Bernd Heine and Tania Kuteva, The Genesis of Grammar: A Reconstruction (Oxford: Oxford University Press, 2007), 151; and Andrason, "Wayyiqtol," 13-15.
114 Huehnergard, "Stative" 221-223, Andersen, "The Evolution," 31; Lipiński, Semitic Languages, 336-337 and 341; Cook, "The Biblical Hebrew Verbal System," 209-219; and Alexander Andrason, "The "Guessing" QATAL - the BH Suffix Conjugation as a Manifestation of the Evidential Trajectory," Journal for Semitics 19/2 (2010): 603-627 (see especially page 610); Andrason, Qatal, 305-307; Andrason,"The BH weqatal, Part 2," 3-5.
115 Bybee, Perkins and Pagliuca, Evolution, 240 and Andrason, "Panchronic Yiqtol," 26-28.
116 Bybee, Perkins and Pagliuca, Evolution, 26 and 235-326.
117 Östen Dahl, Tense and Aspect Systems (Oxford: Blackwell, 1985), 11; Paul Hopper and Elisabeth Traugott, Grammaticalization (Cambridge: Cambridge University Press, 2003), 82; Bybee, Perkins and Pagliuca, Evolution, 25-26.
118 Bybee, Perkins and Pagliuca, Evolution, 296.
119 Andrason, Qatal, 52-54; cf. also Alexander Andrason, "The BH Weqatal: A Homogenous Form with no Haphazard Functions, Part 1," JNSL 37/2 (2011): 1-25 (see especially pages 6-8).
120 Dahl, Tense and Aspect Systems, 11 and Bybee, Perkins and Pagliuca, Evolution, 25-26, 296.
121 Cf. Bernd Heine, Ulrike Claudi and Friederike Hünnemeyer, Grammaticalization: A Conceptual Framework (Chicago: University of Chicago Press, 1991), 71-72.
122 Cf. Hopper and Traugott, Grammaticalization, 82. However, in comparison to the above mentioned processes, the modal contamination is narrower, depicting the rise of modal formations, i.e. it is understood as one of the possible modal paths.
123 Carl Brockelmann, Grundriss der vergleichenden Grammatik der semitischen Sprachen (Hildesheim: Georg Olms, 1961), 74; Hans Bauer and Pontus Leander, Historische Grammatik der hebräischen Sprache des Alten Testamentes (Hildesheim: Georg Olms, 1962), 632; De Lacy O'Leary, Comparative Grammar of the Semitic Languages (Amsterdam: Philo Press, 1969), 275-276; and Louis Gray, Introduction to Semitic Comparative Linguistics (Amsterdam: Philo Press, 1971), 73.
124 Brockelmann, Grundriss, 642, 645.
125 Brockelmann, Grundriss, 30-31, 642. Cf., however, John Huehnergard, "Assertive *la and Hypothetical *lu/law in Semitic," JAOS 103/3 (1983): 569-593 (see especially, page 574). Huehnergard suggests that the factual use of the Akkadian lū is an innovation and corresponds to a weakening of the original counterfactual sense of this particle. Nevertheless, he admits that such a proposal is highly speculative.
126 Peled, Conditional, 37-38.
127 This proposal slightly differs from a reconstruction suggested by Huehnergard ("Assertive," 574-575). As already mentioned, according to his opinion, the PS *lü was originally a counterfactual particle. Thus, the modalisation did not involve a change from factual into counterfactual. Such a proposal (accepted even by its author with a great scepticism), however, seems to be less probable that the evolution posited in this paper due to an extensive factual use of the lü and lü parsaku in Akkadian (the oldest Semitic language). Also other diachronic and typological facts strongly support our reconstruction: in accordance with the typological rule presented previously, the resultative present sense of the PS *qatal- encouraged its use in factual modal contexts while the past sense, available in Biblical Hebrew and Arabic, entailed a coun-terfactual modal value.
128 Cf. Jay Jasanoff, "The Origin of the Italic Imperfect Subjunctive," HSF 104 (1991): 84-105 (see especially pp. 86-87).
129 Cf. Edward Sonnenschein, A New Latin Grammar (Oxford: Clarendon Press, 1912), 150-167 and Augustus L. Francis and Henry Tatum, An Advanced Latin Syntax (Cambridge: Cambridge University Press, 1919), 60-63 and 104-110.
130 Jasanoff, "Origin," 90 and 93.
131 Jasanoff, "Origin," 94.
132 It seems, however, that the development into a conditional (and thus in protases) is more widespread. There are numerous languages that employ old optatives in protases while in apodoses they use other formations. For instance in Spanish the verb in the protasis is an old Latin pluperfect, related to an earlier optative formation. In the apo-doses however, the language uses a novel construction - a type of a future in the past ("I had to do / I was going to do") Si lo pudiese hacer, lo haría "If I could do it, I would do it."
133 The optative sense may stem from contextual factors or can be included in a verbal morpheme or auxiliary. The past and pluperfect values or grams usually stem or derive from earlier resultative presents (cf. the anterior path in section D.1).
134 Andrason, Qatal, 281 and Van der Merwe and Naudé, Reference Grammar. Andrason (Qatal, 305-307) has shown that further senses conveyed by the qatal may be mapped and explained as manifestations of two remaining paths that, jointly with the anterior track, constitute the resultative trajectory - i.e. a comprehensive evolutionary scenario governing the grammatical life of all resultative constructions. In particular, resultative-stative, stative and present temporal values have been unified and explained by employing the network of the simultaneous path (see Andrason, Qatal, 282-283, 305-307; and Andrason, "Wayyiqtol," 42), while rare cases where the qatal offers an evidential sense have been rationalised as expressions of the evidential path (cf. Andrason, "The "Guessing" QATAL," 623-624; and Andrason, Qatal, 282; on the evidential path see Alexandra Aikhenvald, Evidentiality [Oxford: Oxford University Press, 2004]; and Andrason, "The "Guessing" QATAL," 604-609). Finally, certain modal functions of the gram have been classified as a manifestation of the modal contamination path of the original resultative input (Andrason, Qatal, 300-304; on the modal contamination, see Andrason, "The BH Weqatal, Part 1," 7-8). Analogical mappings have been posited for the Arabic qatala (cf. Andrason, Qatal, 223-228).
135 The networking of stative values and the chaining of prospective senses are disregarded in this chart. For their discussion and incorporation into the semantics of the gram see Andrason, Qatal, 282-283, 305-307; and Alexander Andrason, "Future Values of the Qatal are "Logical" - How to Chain Future Senses of the Qatal to its Semantic Network?" HS (forthcoming). The semantic domains in brackets "[ ]" refer to meanings that are not conveyed anymore by the BH qatal. These values, which correspond to more original stages of the paths, have been lost during the evolution of the BH formation. They may however be found in Akkadian where the parsāku, dia-chronically a less advanced resultative gram, offers both resultative proper (present and past) and factual real optative values. The blue spheres correspond to senses conveyed by the BH gram. The arrows represent diachronic (and conceptual) progression. The stages of the anterior path have been simplified in comparison with figure 1 -only perfect, perfective (past) and past phases are represented. The perfective past phase is represented as preceding the past although as explained the change from perfective into simple past is a development that co-occur with the change from a present perfect to definite past tense. The pluperfect sense corresponds to the anterior path of the original input in a past time frame (cf. see Andrason, Qatal, 45; see also Andrason "Wayyiqtol" 12-13).
136 Cf. Alexander Andrason, "The Dynamic Short Yiqtol," JSem 21/2 (2012): 308-339.