On the nature of adjectival resultatives--corpus-based evidence/Priroda adjektiviziranog pasiva u odnosu na gotovost.
In this paper we will demonstrate that the different derivations of passives are not language-specific characteristics. They are found in different languages, Croatian and English being among them. Some passives are verbal and some are adjectival. In order to show the lexical and syntactic characteristics of Croatian passives, we will first analyse the properties of participles, and compare the two passives. We will then focus on adjectival resultatives, a subtype of adjectival passives. Finally, we will compare Croatian adjectival resultatives with their English translation counterparts in order to detect similarities or differences between the two.
We will use corpus evidence for the occurrence of adjectival resultatives in two legislative texts, "Plan prihvata broda u nevolji" (2008) and its English translation "Plan for the Acceptance of a Ship in Distress", in order to present a description of the construction and raise questions about the nature of adjectival resultatives, and the terms in which such a construction can occur. The corpus is built of 23,140 words. Unless otherwise stated, all examples are from the mentioned corpus.
2. Relevant previous studies
The diversity of the passive voice phenomenon has been observed by many linguists. One early study of the passive was Wasow (1977) who was the first to note the difference between the two passives: verbal (eventive) passive and adjectival (stative) passives. According to Wasow's criteria, the rule deriving verbal passives is transformational, while the one for adjectival passives is a lexical redundancy rule. In other words, passives are called lexical when they display behaviour that is demonstrably adjectival, and are called transformational when their derived subjects are not their underlying objects (Wasow, 1977: 342-343). It follows that passives whose derived subjects are not their underlying direct objects should not be able to exhibit adjectival behaviour. Also, lexical but not transformational passives may undergo lexical rules, like category-changing rules. Wasow (1977: 343), working on the two passives, provides diagnostics for adjectival passives: prenominal adjective position; appearance as complements to verbs like to act, become, look, remain, seem and sound; prefixing of un, and degree modification by very. The requirements for transformational derivation are the following: passives of double object constructions; passives of the "accusative subject" construction; passives of idiom chunks; passives of to help and thank, and passives followed by predicative expressions like President which can appear after past participles and not ordinary adjectives. For example: Mary was elected President vs. * Mary was happy President (1977: 341-343).
The issue of distinction between verbal and adjectival passives is taken up again by Bresnan (1982) who centres the discussion on the three distributional diagnostic contexts and one morphological diagnostic test. The examples (1-3) are from Bresnan (1982: 29-31). The sentence in (1a) has a past participle considered with an adjectival complement. English adjectives do not take adjectival complements; the participle is hence a verb. The example in (1b) has the same form in the prenominal position of an adjective and is modified by very. Both characteristics support the adjectival reading.
(1) a Margaret's statement was considered Verbal passive. profound. b That was a very considered statement. Adjectival passive.
The sentence, as in (2a), contains a past participle (spared) followed by a second object noun phrase. English adjectives do not take noun phrase as objects; the participle is thus a verb. The participle in (2b) cannot take an object; it comes in a prenominal position and is clearly an adjective.
(2) a The prisoners were spared execution. Verbal passive. b The spared prisoners were freed. Adjectival passive.
The "reversative" prefix un, as in (3a), is attached to the verb to indicate the reversal of the action denoted by the verb. As, for example, in load / unload. If the prefix un is attached to an adjective, for example happy / unhappy, it would have a negative meaning of not having the feature represented by the adjective. Bresan (1982) argues that the negative adjectival prefix un is not attached to verbs, although there is a verb to unzip or to unload but not to untouch. Therefore, the participles occurring with the negative adjectival prefix un, as untouched, are not verbs but adjectives. Bresnan's (1982) examples, as in (3a & b), further show that both verbal and adjectival participles can be followed by a "by + agent".
(3) a The jacket was unzipped by someone wearing Verbal fingernail polish. b The jacket was untouched by human hands. Adjectival.
The generally accepted analysis of verbal passives is that the agent theta role is not assigned and accusative is absorbed. Following Levin & Rappaport (1986), the change of the verbal into adjectival passive accounts for the external role of the base form, the externalisation of the internal role, the absorption of case and the elimination of the [NP, VP] position. Levin & Rappaport (1986: 625-626) propose three diagnostic environments for adjectival passives. They repeat some of the arguments put forward by Wasow (1977) and Bresnan (1982) and they claim that:
* past participles prefixed with un are adjectival; * a past participle appearing as the complement of to seem, remain, sound and look is adjectival, not verbal; * only adjectives and not verbs occur as prenominal modifiers.
Some studies have proposed subtypes of adjectival passives with different syntactic and semantic properties. Embick (2004: 355-356), for example, deems the distinction between adjectival and verbal passive imprecise. He distinguishes among eventive passives (the term roughly corresponds to the verbal, ibid.) and two types of adjectival (stative) passives: resultative and stative. Resultative passives imply a state that is the result of a previous event, while the stative is a simple state, like a simple adjective. In many cases the two types are identical in form (e.g. closed); in other cases they have different forms (e.g. open--stative, opened--resultative). Embick (2004) analyses the two types of adjectives and notes that both adjectives are created syntactically using different functional heads. Following this, the two types of adjectives involve different aspect heads (heads which are the locus of participial morphology). Embick (2004: 363-364) goes on to claim that statives lack eventivity and are derived by the merge of the Aspect head to the root itself, without any verbal head. Resultatives denote a state that results from a prior event and consequently their structure must include a verbal head. But, according to Embick, resultatives are not agentive and the verbal head involved in their derivation cannot have the feature AG (agentivity). The following examples (2004: 356) reflect Embick's interpretation of the participle forms.
[FORMULA NOT REPRODUCIBLE IN ASCII]
(4) The door was opened. a Eventive passive Someone opened the door. b Resultative The door was in a state of having become open. (It requires a state resulting from an event).
He adds a third form. The form describes a simple state and is thus referred to as a stative.
(5) c Stative The door was open.
Embick notes that the resultative and the stative have distinct forms, though this is not always the case. To make an argument, he provides a diagnostic test to differentiate stative from resultative participles (the below examples are from Embick, 2004: 357-359):
2. Adverbial modification
Unlike pure statives, as in (6b), resultatives, as in (6a), allow adverbial modification (manner and others). In some cases an adverbial is possible with a stative but a resultative with the same adverbial has an additional reading.
(6) a The package remained carefully opened. b * The package remained carefully open.
The only reading in (7a) is that the door was open at a recent point in the past and is (probably) no longer open. On the other hand, the reading in (7b) is twofold: 1. as in (7a): the door was in the opened state recently but probably is no longer; 2. the door is in the opened state, the opening having taken place recently.
(7) a The recently open door. b The recently opened door.
2. Resultatives and statives after verbs: to build, create, make
The second test places the resultative or stative participle in an environment after a verb of creation, for example: to build, create, make. There is no contradiction between open which defines a simple state and the environment in which it occurs, as in (8a). Embick, however, notes that when the complement denotes a state resulting from a previous event, there is a contradiction, as in (8b), and the sentence is ill-formed and thus grammatically unacceptable. However, closed, as in (8c), is grammatical in this environment. It is stative and has no event like the resultative in (8b). Open is not typically seen as a participle.
(8) a This door was built open. Stative b * This door was built opened. Resultative c The door was built closed Stative
3. Ability to serve as resultative secondary predicates
The third syntactic diagnostic test distinguishes statives and resultatives by their (in)ability to serve as resultative secondary predicates. This is another environment in which statives are possible but resultatives are not.
(9) a John kicked the door open. b * John kicked the door opened. c Bill drank the glass empty. d * Bill drank the glass emptied.
4. Prefixation with un
The fourth syntactic diagnostic test is prefixation with un, generally restricted with statives, but applies with resultatives. Although there are some statives with un (e.g. unshaven), the general pattern is that un-prefixation comes with resultatives but not statives. Open is a stative, opened is a resultative; unopened is acceptable while *unopen is not.
(10) unopened * unopen unrotted * unrotten
Kratzer (2000) has made a significant contribution to the discussion of statives. She (2000: 1-2) looks at the statives in German and proposes two subclasses that behave differently with respect to the adverb still (immer noch):
a. target state passives; b. resultant state passives.
Target state passives describe states that are reversible which is what the adverb still requires. On the other hand, resultant state passives convey that the event represented by the participle is over by now. The state is hence irreversible and is incompatible with the adverb still. Kratzer (2000: 14) concludes that the resultant state participles are less adjective-like than the target state participles given that the former are never gradable and never permit the degree modifier very. She, however, admits that the still diagnostic test is not absolutely reliable. Kratzer (2000: 2-3) refers to Parsons (1990) who draws parallel between the resultant state passives and the perfect construction, and who proposes the resultant state interpretation as an interpretation for the English perfect construction. If this is right, Kratzer writes, then the state passives in (11b) share the aspectual properties of the verbal passives in (11a). However, such a view is in conflict with the examples Kratzer gives, as in (12). She notes that there is a subtle difference in meaning. The meaning in (12a) is the following: the children have washed themselves, while in (12b) someone must have washed them. Kratzer concludes that "resultant state passives have perfect aspect, but they are not just elliptical versions of perfect forms of verbal passives" (2000: 4).
(11) a The mailbox has been emptied. Verbal passive b Der Briefkasten ist (* immer noch) Resultant state geleert. passive. The mail box is (* still) emptied. (12) a Die Kinder sind (* immer noch) Resultant state gewaschen. passive. The children are still washed. b Die Kinder sind gewaschen worden. Perfect form of the verbal passive. * The children are washed gotten. The children have been washed.
In short, one of Kratzer's central arguments over the distinction between target state participles and resultant state participles rests on the following: resultant state participles are less adjective-like than the target state participles. They are never gradable and they never permit the degree modifier very.
Croatian does not mark the difference between verbal and adjectival passives overtly. For example, the same form of the verb biti (to be) + past participle is used in (13a-c). The sentences (13a-c) are taken from Belaj (2004: 18). The agent od klauna (by the clown), however, determines the verbal reading in (13a) while (13c) reads as a copula-adjective construction. Nasmijan covjek implies a quality or a feature typical of the person, independent of time where the external argument (1) is not accessible. On the other hand, the durative dverbial od jucer poslijepodne (since yesterday afternoon) stands for the adjectival passive reading. An important and recent contribution to the debate of passives has come from Belaj (2004). The understanding of adjectival passives is shaped by the cognitive domain, he notes. It is the cognitive domain that gives us an insight into the process of passivisation in the past. Following the importance of the cognitive domain for the analysis of adjectival passives, Belaj (2002, 2004) rests one of his central arguments on the distinction between a copula-adjective construction (a type of copula construction whose predicate is an adjectivised verbal participle) and adjectival passive, where the role of the durative adverbial, as in (13c), is seen as crucial. And if a state is the result of a prior event, then it also implies an external argument.
(13) a On je nasmijan od klauna. Verbal passive. b On je nasmijan covjek. Copula-adjective construction. c On je nasmijan od jucer poslijepodne. Adjectival passive.
This study is framed in the above theory, where, from this point on, the term adjectival resultatives is used to refer to participles that represent a state that is the result of a prior event but the context the past participle occurs in signals its currency. Adjectival resultatives are thus the construction we focused on in the study and the problem we shall focus on in the rest of the paper.
The aim of the study is to identify the syntactic features of adjectival resultatives in the source text (ST) and compare them with their translation pairs in the target text (TT) in order to detect any evidence of change on the syntactic level that occurred in translation.
Our descriptive data was obtained from two legislative texts and official documents: "Plan prihvata broda u nevolji" (2008) and its English translation "Plan for the Acceptance of a Ship in Distress". The document is an integral part of the "Ordinance on places of refuge". The "Ordinance on places of refuge" establishes the basic guidelines and the legal framework which applies to the procedure in the case of a request for a place of refuge (the content of the "Plan for the Acceptance of a Ship in Distress"), to the responsibility of the authorities and their accountability in procedures following request for a place of refuge, and procedures for securing financial warranties for liability in the event of damage.
The Croatian corpus was a point of departure. Our first step was to differentiate verbal (eventive) from adjectival (stative) passives (see Table 1). Since both verbal and adjectival passives bear the same morphology, we could not rely on their morphology alone. We decided to test the accessibility of an external argument and the role of the durative adverbial. Those that had an external argument, even when not realised syntactically, were verbal passives. Both (unedited) sample sentences are from the studied parallel texts.
Secondly, we focused on adjectival passives in order to distinguish adjectival resultative from stative participles in the ST. As already stated, in this study the term adjectival resultatives refers to participles that represent a state that is the result of a prior event but the context the participle occurs in signals its currency. Unlike statives which imply a simple state, like a simple adjective (Embick, 2004).
We tested the characteristics of adjectival resultatives based on transitive verbs in order to satisfy the criteria for the use as an adjective. We did it by applying syntactic test which provided evidence for the adjectival resultative status of the extracted samples:
1. negative adjectival prefix un (ne);
2. adjectival gradation;
3. coordination with prototypical adjectives;
4. degree modification by very;
5. presence of durative adverbial.
We then analysed morphosyntactic properties of the isolated adjectival resultatives. The next step was to look for translation equivalents in the English corpus and investigate if a change on the syntactic level occurred in translation. We wanted to see whether adjectival resultatives in Croatian match their translational pairs in English.
5.1. Adjectival resultatives in Croatian
In the following section we will present a representative sample and behaviour of adjectival resultative passives in the ST, based on transitive verbs but with respect to the applied test for the use as an adjective. Assigning the adjectival resultative status put emphasis on the response to the following diagnostics:
1. negative adjectival prefix un (ne);
2. adjectival gradation;
3. coordination with prototypical adjectives;
4. degree modification by very;
5. presence of durative adverbial.
The translation of the selected Croatian samples presented in the Results section is only to help understand the extracted samples. A translationally relevant analysis is presented in the paper later. All translated sentences are from the English corpus.
The analysis generated a major finding: a high number of occurrences of adjectival resultatives. There is ample evidence to confirm the occurrence and complexity of adjectival resultatives in the Croatian corpus. A total of 33 adjectival resultative passives was found. The findings are now organised into two subgroups: a) those confirming the diagnostics in a minimum of three tests (N=19), as in Table 2, and b) those confirming the fifth test, i.e. the presence of durative adverbial, while the number of positive response to the diagnostics is less than three, (N = 14), as shown in Table 4.
The findings below (Table 2) display the number of adjectival resultatives examined, and the ratio of the adjectival resultatives with relation to the applied test for the use as an adjective. Despite the difference among the aptitude of adjectival resultatives, the findings clearly show that the presence of durative adverbial (either explicit or implicit in the context) is the most common diagnostic for adjectival resultatives in Croatian. In the corpus sample of 33 (see Tables 2 & 4) one diagnostic test occurs with a frequency strikingly greater than all others: durative adverbial (N=33). Coordination with prototypical adjectives (N=17), for example: zaklonjeno i sigurno (sheltered and safe), comes second in the rank of frequency. It is another evidence for the adjectival nature of the past participles.
Significant is the negative prefix un (N=10), the meaning of not having the characteristic represented by the adjective. For example: ugodan / neugodan (pleasant / unpleasant), sretan / nesretan (happy / unhappy). The negative prefix un (ne) is not attached to Croatian verbs, therefore the past participles occurring with un (ne) are adjectives. The lowest proportion of adjectival resultatives is demonstrated when adjectival gradation diagnostic is applied (N=7). Even though the number of cases is not high, the found adjectival resultative passives display comparative morphology, which again proves that the participle
has been transformed into an adjective, for example: zaklonjen, zaklonjeniji (sheltered, more sheltered). Examples in Table 3 reinforce the interpretation of adjectival resultatives.
Durative adverbials are important in adjectival resultative readings in Croatian. The corpus data supported the adjectival resultative readings in a smaller but interesting number of instances (N=14) that passed the fifth diagnostic while the number of responses to diagnostics is less than three (see Table 4). The durative adverbial, for example jos uvijek (still), implies that the time for which the past participle stands and extends over the relevant (current) reference time.
Neprikazan (not shown), for example, is interesting. Its reading is twofold: not having the characteristic represented by the adjective, and the negation of prikazan (shown), which is the meaning conveyed in the text. Je li sidro spusteno? (Anchor released?) accounts for the acceptability of the adjectival resultative reading and for increased acceptability of reading in a different context. The sentence, as in (15a), can have both verbal and adjectival reading. If, however, we read the sentence as embedded in the context "Checklist - Ship accidents and breakdowns" (15b), then the reading is verbal, the meaning is 'the job is done'. Yet, if we read the sentence in a different natural setting, if the context provides an interpretation with a clear resultant state, then its adjectival resultative reading, as in (15c), improves significantly. Also, if we transform the interrogative into declarative statement, then the sentence Sidro je (jos uvijek) spusteno (The anchor is still released) is acceptable, as in (15d).
(15) a Je li sidro spusteno? Verbal and adjectival passive. Anchor released? b Je li sidro spusteno? Verbal passive. Anchor released? c Je li sidro (jos uvijek) Adjectival resultative spusteno? passive. Is the anchor (still) released? d Sidro je (jos uvijek) spusteno? Adjectival resultative passive. The anchor is (still) released.
5.2. Croatian adjectival resultatives in translation
The analysis of the change / retention of the adjectival resultative in translation is shown in Tables 5-9. Sixteen out of 19 scrutinised adjectival resultatives in the ST (Table 2) were translated into adjectival resultatives in the TT (Table 5). The count was restricted to those cases for which an adjectival resultative-to-adjectival resultative counterpart would be available, that is, an adjectival resultative-to-adjectival resultative form in the ST was rendered as such in the TT, 16 examples in total. The un-prefixation and durative adverbial ranked highest when the diagnostics was applied in the TT, followed by coordination with prototypical adjectives. Adjectival gradation scored lowest.
Three adjectival resultatives from the ST did not result in the translation pair. One adjectival resultative was translated into an adjective (difficult), one into a verb (shall pass), and one was translated into the participle (released). This study analysed the occurrence of to be + participle forms.
Before we turn to more specific focus, it is worth considering interesting findings shown in Table 6. Table 6 shows how extracted samples in the ST match their counterparts in the TT. They account for behaviour concerning the same diagnostics applied in the ST and the TT.
While neizlozen, as in (16), is not accepted in Croatian, unexposed is entering English (the meaning of not having been exhibited or brought to public notice, e.g. 'Andy Warhol's Unexposed Exposures at Steven Kasher Gallery').
ST TT (16) 1. * neizlozen 1. is unexposed Zaklonjeno podrucje manjem utjecaju; to; je svako podrucje u 2. ili je 2. is more exposed unutrasnjim morskim izlozeniji; to; vodama ili 3. ili je izlozen 3. is exposed and teritorijalnom moru i podlozan; open to; na kojem je brod 4. ili je jako 4. is very exposed; izlozen manjem izlozen; 5. is still utjecaju vjetra i 5. ili je jos exposed. mora, bez obzira uvijek izlozen. plovi li, pluta ili je usidren. A sheltered area is any area in the internal sea waters or territorial sea in which a ship is less exposed to the effects of wind and sea, regardless of whether it is navigating, floating or is anchored.
Table 7 displays an increase in the un-prefixation rate in the TT. Such examples (N=8) respond to the five-item diagnostics in the TT now.
The next step was to analyse the change / retention of adjectival resultative in 14 cases that complied with the durative adverbial while the number of responses to diagnostic testing was less than three in the ST (see Table 4). Seven did not have a counterpart in the TT, four cases retained the adjectival resultative reading (Table 8) and three kept the adjectival resultative reading while also increasing the number of diagnostics (Table 9).
As already pointed out, three cases (Table 9) showed a significant increase in diagnostics, from two in the ST (coordination with prototypical adjective and durative adverbial) to three in the TT (un-prefixation, coordination with prototypical adjective and durative adverbial). However, the presented sample in Table 9 can have a twofold reading. A sheltered area is listed, the job is done (verbal passive), but it can also be read as a state that is the result of a previous event, a sheltered area that is still listed.
In this study we investigated syntactic features of adjectival resultatives in the Croatian text (ST) and compared them with their translation pairs in the English text (TT) in order to detect any evidence of overlap on the syntactic level. Adjectival resultatives needed to fulfil the criteria for the use as an adjective. The criteria provided the acceptability for adjectival resultative reading. To do this we framed our analysis and discussion in the works of Wasow (1977), Bresnan (1982), Kratzer (2000), Embick (2004) and Belaj (2002, 2004).
Although verbal and adjectival passives are morphologically very close to each other, certain morphological and syntactic environments clearly distinguish between them. This study has provided evidence of adjectival resultatives and their occurrence in the two parallel legislative texts, the corpus sample of 33. We applied the following diagnostics: negative adjectival prefix un (ne), adjectival gradation, coordination with prototypical adjectives, degree modification by very, and presence of durative adverbial. The study shows that most tests for adjectival resultatives work in Croatian. However, the key to understanding the adjectival resultatives in the ST is the durative adverbial, already recognised by Belaj (2002, 2004). The action represented by the past participle is not set in the present time frame, the external argument is not accessible, but the action materialises in the cognitive domain of the speaker and signals a correlation between the passive and active. In other words, adjectival resultatives include in the interpretation an event which is reconstructed in order to accommodate the role of the external agent.
Looking at the evidence from this small corpus, and considering the findings of the analysis, we may say that our study differs from Embick (2004) who does not consider the inclusion of a verbal head with an agentivity feature in his interpretation of resultatives. At this point, however, we may agree with Kratzer (2000) who usefully distinguishes between target and resultant state passives applying the still (immer noch) diagnostic test where target state passives describe states that are reversible, which is what the adverb still requires. They are also gradable and they permit the degree modifier very.
The results of this study also show that the extracted sentences in the ST responded positively to adjectival resultative diagnostics. Apart from the durative adverbial, coordination with prototypical adjectives, prefixing of un (ne), degree modification by very and adjectival gradation, additional contextual support reinforced the adjectival resultative reading of the studied sentences. It was also important to observe the behaviour of un-prefixation in the TT. Their occurrence increased in the TT and as a result such adjectival resultatives responded to the five-item diagnostics for the use as an adjective in the TT.
The focus of the article was the nature of adjectival resultatives in two parallel texts: Croatian (ST) and English (TT). This study has found strong evidence that adjectival resultatives occur in both legislative texts. The evidence was provided by the crosslinguistic analysis of behaviour that adjectival resultatives display with regard to their responses to the diagnostics for the use as an adjective.
Following the findings presented in the Results section, there is every reason to conclude that the adjectival resultatives in the two texts imply a state that is the result of a previous event. In the ST, they mostly respond to the durative adverbial diagnostic testing and they also overwhelmingly respond to the coordination with prototypical adjectives diagnostic. While in the TT, they largely respond to the un-prefixation and durative adverbial diagnostics. Yet, the adjectival gradation diagnostic scores lowest in both texts.
Another point we would like to make, and it is a challenge for any study, is that adjectival resultative passives are generally shaped and determined by pragmatics. We can conclude that the formation depends more on the context than the grammar or lexicon and that it would be misleading to judge adjectival resultative passive formation on grammatical grounds only. The context, cognitive domain and interpretation are important. Furthermore, to understand the relation between grammar and pragmatics is essential when approaching semantics.
There is also the question of occurrence within particular types of written text: most passives occur in technical and legislative texts. The fact that this study is able to show passive is an integral characteristic of the genre, and as genres reflect different environments it is interesting to investigate the occurrence of specific grammatical categories in specific contexts.
The findings of the presented study are significant for several reasons. They stem from the corpus study and as such they provide solid evidence for the existence of adjectival resultative passives. But they also highlight the important role of cognitive domain in the interpretation of adjectival resultative passives. It is also an attempt to attach weight to the contextual conditions in which adjectival resultatives occur. Although the construction under consideration is complex, we hope that we have made a further contribution to the debate of past participles by offering evidence from a small corpus of legislative pair texts. This is another in a series of our attempts to study the positioning and frequency of past participles within a particular genre.
8. Directions for further research
The research results put forward areas for further research. Adjectival passives discussed in the paper are semantically close to se passives with qualitative-generic meanings, for example: Nakon smjestaja broda na mjesto zaklonista, clanovima posade broda nije dopusteno slobodno kretanje kopnom (After accommodating the ship in a place of refuge, the ship's crew is not permitted to communicate with the shore ...) / ne dopusta se slobodno kretanje brodom. The question for further research would be to investigate the nature of structures with qualitative meaning.
We would like to thank Anne Marie Foerster Luu for her invaluable comments on a draft of this paper.
Belaj, Branimir, 2004. Pasivna recenica. Osijek: Sveuciliste J. J. Strossmayera, Filozofski fakultet.
Belaj, Branimir, 2002. Kategorija gotovosti i vremenska vrijednost pasivnoga predikata. Jezikoslovlje. 3. 1-2: 1-16.
Bresnan, Joan, 1982. The passive in lexical theory in Joan Bresnan, ed. The Mental Representation of Grammatical Relations. Cambridge, MA: The MIT Press: 3-86.
Carter, Ron & Michael McCarthy, 2006. Cambridge Grammar of English. A Comprehensive Guide. Spoken and Written English Grammar and Usage. Cambridge: Cambridge University Press.
Embick, David, 2004. On the Structure of Resultative Participles in English. Linguistic Inquiry 35(3): 355-392.
Kratzer, Angelika, 2000. Building Statives. Berkeley Linguistic Society 26. Available from: [http:// semanticsarchive.net/Archive/GI5MmI0M/kratzer.building.statives.pdf] [Accessed 14 January 2010]
Levin, Beth & Malka Rappaport, 1986. The Formation of Adjectival Passives. Linguistic Inquiry, 17: 623-661.
Parsons, Terence, 1990. Events in the Semantics of English. A Study in Subatomic Semantics. Cambridge, Mass.: MIT Press.
Pravilnik o mjestima zaklonista. Plan prihvata broda u nevolji. 2008. Narodne novine broj 3.
Warhol, Andy: Unexposed Exposures. Available from: http://www.stevenkasher.com/html/exhibinfo. asp?exnum=1162 [Accessed 1 March 2010]
Wasow, Thomas, 1977. Transformations and the Lexicon, in Peter W. Culicover, Thomas Wasow & Adrian Akmajian, eds. Formal Syntax. New York: Academic Press: 327-360.
Faculty of Maritime Studies, Rijeka
(1) The term external argument is used as in Levinn & Rappaport (1986). "It appears external to the AP headed by the related adjectival passive participle (ibid: 624)".
(2) In the interpretation of the sample as an adjectivised passive with a durative adverbial jos uvijek (still) the importance of the context is observed. We can assume that the context refers to every ship and every crew. It is not explicitly stated that the ship has or has not been previously accommodated in a place of refuge. The context is also short of explicitly stating whether the crew was permitted to communicate with the shore if the ship had previously been accommodated in a place of refuge. Two contexts and two interpretations would thus be possible. One referring to the first accommodation of the ship in a place of refuge where the jos uvijek (still) interpretation would not be acceptable--the crew aboard a ship underway could not have communicated with the shore, and the other conveying the meaning that the ship has first entered a port but the crew was not permitted to communicate with the shore. The ship was then accommodated in a place of refuge where the crew was still not permitted to communicate with the shore.
(3) This context did not verify the diagnostics. In a different context, however, a ship could be exposed to a more or less intense threat. Such a case would then verify the diagnostics.
Table 1 Verbal (eventive) and adjectival (stative) passives Diagnostics Sample sentence Mora biti Svako premjestanje broda iz Verbal odobreno od ... / mjesta zaklonista mora biti (eventive) must be approved by ... odobreno od dezurnog sluzbenika Nacionalne sredisnjice. (presence of an agent and the event Any relocation of a ship from represented by the the place of refuge must be participle implies approved by the on-duty officer that the task is of the RCC. performed). Clanovima posade Nakon smjestaja broda na mjesto Adjectival broda nije (jos zaklonista, clanovima posade (stative) uvijek) dopusteno ... / broda nije dopusteno slobodno the crew is (still) not kretanje kopnom, osim za permitted. obavljanje nuznih poslova na odrzavanju sigurnosti broda The state (the crew is odnosno sprecavanju ili still not permitted) uklanjanju oneciscenja. (2) is the result of a previous event: After accommodating the ship in permission to a place of refuge, the ship's communicate with the crew is not permitted to shore was not given at communicate with the shore, some point. except in the event of conducting essential activities on maintaining the safety of the ship or preventing or removal of pollution. Table 2 Overall response to diagnostics in the ST Diagnostics 1. 3. Negative Coordination Total adjectival 2. with number of prefix Adjectival prototypical sentences un (ne) gradation adjectives Number of cases 10 7 17 that pass the test: 19 10 7 17 4. Total Degree 5. number of modification Durative sentences by very adverbials Number of cases 11 19 that pass the test: 19 11 19 Table 3 Diagnostics of adjectival resultatives Diagnostics Sample sentence 1-5 1. ako je nezaklonjeno; Odredeno mjesto ili morsko 2. ako je zaklonjenije; podrucje valja smatrati 3. ako nije ni zaklonjeno neprimjerenim za smjestaj broda ni sigurno; ako: 4. ako nije jako zaklonjeno; * nije zaklonjeno od 5. jos uvijek nije prevladavajucih vjetrova i zaklonjeno. valova; ... A place or a sea area shall be considered unsuitable for accommodation of a ship if: * it is not sheltered from prevailing winds and waves; ... 2-5 1. * ako je zapovjednik Zahtjev za dodjelom mjesta neuvjeren; zaklonista valja smatrati potpunim 2. uvjereniji; ako je zapovjednik: 3. ako je zapovjednik uvjeren i siguran; * prikupio podatke o stanju broda, 4. ako je zapovjednik okoline i mozebitnoj prijetnji jako uvjeren; ljudima, brodu i morskom 5. ako je zapovjednik jos okolisu; uvijek uvjeren. * procijenio stupanj opasnosti i zakljucio da ne postoji neposredna i ozbiljna prijetnja ljudima; * uvjeren da bi nastavak putovanja znacio izlaganje broda povecanim opasnostima ... The request for granting of a place of refuge should be considered complete if the shipmaster: * has gathered data about the state of the ship, environment and possible threat to human lives, the ship and marine environment; * evaluated level of danger and concluded that there is no direct and serious threat to persons; * is convinced that continuation of the voyage means exposing the ship to more serious danger ... 1, 1. ili je neusidren; Popravak stroja, trupa ili druge 3&5 2. * ili je usidreniji; opreme obavlja se u pravilu u 3. ili je usidren i granicama teritorijalnog mora na siguran; brodu koji pluta ili je usidren, 4. * ili je jako usidren; prema zahtjevu zapovjednika. Kada 5. ili je od jucer usidren. je to moguce u slucaju potrebe plutanja duzeg od 4 sata ili tijekom noci brod treba sidriti. The repair of the ship's machinery, hull or other equipment on the ship floating or anchored shall normally be conducted within the borders of the territorial sea, according to the request of the shipmaster. When this is possible, in the case of floating longer than 4 hours or during the night, the ship should be anchored. Table 4 Presence of durative adverbials (selected samples) Diagnostics Sample sentence 3&5 1. * neprikazani su; Svi brodovi koji traze mjesto 2. * prikazaniji su; zaklonista moraju imati 3. prikazani su i vidljivi u osiguranje do granica prilogu; ogranicenja odgovornosti 4. * jako su prikazani; sukladno Protokolu iz 1996. 5. jos uvijek su prikazani u godine na Konvenciju o prilogu. pomorskim trazbinama, 1976. Primjeri potvrde o financijskom jamstvu i P&I osiguranju prikazani su u prilogu. All ships requesting a place of refuge must have insurance to the amount of the limitation of liability in accordance with the Protocol from 1996 on the Convention on limitation of liability for maritime claims, 1976. An example of a certificate of a financial security and P&I insurance is shown in the Supplement. 1. * Je li sidro nespusteno? Je li sidro spusteno? 2. * Je li sidro spustenije? 3. --; 4. * Je li sidro jako spusteno? 5. Je li sidro jos uvijek Anchor released? spusteno? Table 5 Overall responses to diagnostics in the TT Diagnostics 1. 3. Negative Coordination Total adjectival 2. with number of prefix Adjectival prototypical sentences un (ne) gradation adjectives Number of cases 16 9 14 that pass the test: 16 16 9 14 4. Total Degree 5. number of modification Durative sentences by very adverbials Number of cases 10 16 that pass the test: 16 10 16 Table 6 Adjectival resultative-to-adjectival resultative translation (unchanged form) Total Diagnostics Sample sentence 7 1-5 1. is unsheltered; Odredeno mjesto ili morsko 2. is more sheltered; podrucje valja smatrati 3. is not sheltered nor neprimjerenim za smjestaj safe from ... ; broda ako: 4. is not very sheltered; 5. is still not sheltered. * nije zaklonjeno od prevladavajucih vjetrova i valova; ... A place or a sea area shall be considered unsuitable for accommodation of a ship if: * it is not sheltered from prevailing winds and waves; ... 3 1-4 1. is unexposed; Procjenu okolnosti obavlja 2. * (threats to which dezurni sluzbenik it) is more exposed Nacionalne sredisnjice. (context); Procjena okolnosti sastoji 3. (threats to which it) se od: is exposed and open; 4. (threats to which it) * prikupljanja podataka o is very exposed; brodu, njegovim opcim 5. it is still exposed. obiljezjima te prijetnjama kojima je izlozen ... (3) The assessment of circumstances shall be conducted by the on-duty officer of the RCC. The assessment of the circumstances consists of: * collecting data about the ship, its general characteristics and threats to which it is exposed ... 6 1, 1. is unfounded; Mjesto zaklonista uz obalu 3&5 2. * is more founded; nece se predloziti ako: 3. not founded nor complete; * zahtjev nije utemeljen 4. * the request is very ili zapovjednik broda founded; odbija pruziti potrebna 5. the request is still objasnjenja i podatke; founded. A place of refuge along the coast shall not be proposed if: * the request is not founded or ... Table 7 Un-prefixation in the ST and the TT Total Diagnosis Sample sentence ST TT 1. * ako je 1. is Zahtjev za dodjelom mjesta zapovjednik unconvinced; zaklonista valja smatrati neuvjeren; 2. is more potpunim ako je 2. uvjereniji; convinced; zapovjednik: 3. ako je 3. is zapovjednik convinced * prikupio podatke o uvjeren i and sure; stanju broda, okoline i siguran; 4. is very mozebitnoj prijetnji 4. ako je convinced; ljudima, brodu i morskom zapovjednik 5. is still okolisu, jako convinced. * procijenio stupanj uvjeren; opasnosti i zakljucio da 5. ako je ne postoji neposredna i zapovjednik ozbiljna prijetnja jos uvijek ljudima, uvjeren. * uvjeren da bi nastavak putovanja znacio izlaganje broda povecanim opasnostima ... The request for granting of a place of refuge should be considered complete if the shipmaster: has gathered data about the state of the ship, environment and possible threat to human lives, the ship and marine environment, * evaluated level of danger and concluded that there is no direct and serious threat to persons, * is convinced that continuation of the voyage means exposing the ship to more serious danger ... Table 8 Adjectival resultative-to-adjectival resultative translations from the ST into the TT Total Diagnosis Sample sentence ST TT 1. clanovima 1. the crew is Nakon smjestaja broda na posade unpermitted mjesto zaklonista, broda je to; clanovima posade broda nedopusteno 2. * is more nije dopusteno slobodno slobodno permitted; kretanje kopnom, osim za kretanje; 3. * it is obavljanje nuznih poslova 2. * nije neither na odrzavanju sigurnosti dopustenije; permitted broda odnosno sprecavanju 3. * clanovima nor ili uklanjanju posade acceptable; oneciscenja. broda nije 4. * it is dopusteno ni not very After accommodating the pozeljno permitted; ship in a place of refuge, slobodno 5. it is the ship's crew is not kretanje; still not permitted to communicate 4. * nije jako permitted. with the shore, except in dopusteno; the event of conducting 5. jos uvijek essential activities on nije maintaining the safety of dopusteno. the ship or preventing or removal of pollution. 4 Table 9 Adjectival resultative-to-adjectival resultative translation--increased diagnostics Total Diagnosis Sample sentence ST TT 1. * koje je 1. which is Mjesto zaklonista jest kao takvo unlisted luka, dio luke ili nenavedeno u as such; zaklonjeno pristaniste poglavlju 5; 2. * which is ili sidriste odnosno 2. * navedenije; more listed; drugo zaklonjeno podrucje 3. koje je 3. which is koje je kao takvo kao takvo listed and navedeno u poglavlju 5. navedeno i visible; ovog Plana. vidljivo u 4. * which is poglavlju 5; very listed; A place of refuge means a 4. * koje je kao 5. which is port, part of a port or takvo jako still sheltered pier or navedeno; listed. anchorage or other 5. koje je kao sheltered area which is takvo jos listed as such in chapter uvijek 5 of this Plan. navedeno. 3