As a technical matter, the way in which FrameNet analyzes instances of a target predicate consists of marking up parallel aligned layers of annotation with appropriate label sets, as shown in Figure 3.1. The layers that are displayed in the FN Desktop can be manually selected by annotators. The number of layers and the kind of information that can be recorded on them is technically unlimited. But in FrameNet's current practice the four core annotation layers are the Target, frame element (FE), grammatical function (GF), and phrase type (PT) layers. On the first, the (parts of the) target predicate are marked while on the latter three, labels are applied to the constituents expressing the frame elements of the target.
The next-most important set of layers consists of the layers called Other; a layer called either Noun, Verb, Adj, Adv, or Prep depending on the part of speech of the target (this layer is also often called the part-of-speech-specific layer); and the Sent(ence) layer. The Other layer holds labels relating to certain special constructional contexts in which the target may occur, such as relative clauses, existential constructions, and extraposition constructions. The part-of-speech-specific layer holds labels that can occur only with predicates of a particular part of speech. For instance, the Copula does not occur with verbal targets; it can be applied on the Noun, Adj, and Prep layers of targets of the appropriate part of speech. The Sent(ence) layer is special in that it does not actually bear any annotation labels: when the layer is invoked, information about the sentence as a whole can be recorded on an appearing list of check-boxes.A final group of layers includes, among others, layers holding labels related to part of speech (POS) and Named Entity Recognition (NER). This information is derived automatically from our corpora and third-party software and is generally not modified by FrameNet annotators.3
We now turn to FrameNet's annotation process. The work can be divided into two kinds according to the way in which sentences are chosen for annotation. In the lexicographic annotation mode, our main focus is on the goal of recording the range of semantic and syntactic combinatory possibilities (valences) of each word in each of its senses. To that end, we extract sentences from the different texts of a corpus because they contain a predetermined target LU. Then we annotate a selection of the extracted sentences in respect to that particular LU.
In another kind of work, the annotation of running text (also called full-text annotation), the sentences are chosen for us, so to speak. Annotation of running text is technically possible thanks to the annotation layering technique: FN lexicographers can one by one declare each word in a sentence a target, select a frame relative to which the new target is to be annotated, get a new set of annotation layers (frame element, grammatical function, phrase type) and appropriate frame element tags, and then annotate the relevant constituents.
Before going further into the details of annotation, let us briefly consider the Revenge frame, which will figure as an example frame repeatedly in this chapter. The definition of this frame follows:
An Avenger performs some Punishment on an Offender as a response to an earlier action, the Injury, that was inflicted on an Injured_party. The Avenger need not be identical to the Injured_party but needs to consider the prior action of the Offender a wrong. Importantly, the punishment of the Offender by the Avenger is seen as justified by individual or group opinion rather than by law.
Thus, the frame elements in the Revenge frame are Avenger, Punishment, Offender, Injury, and Injured_party.
Lexical units in this frame include avenge.v, avenger.n, get even.v, retaliate.v, retaliation.n, retribution.n, retributive.a, retributory.a, revenge.v, revenge.n, revengeful.a, revenger.n, vengeance.n, vengeful.a, and vindictive.a. Some example sentences with the lexical unit avenge are given here.
(1) [His brothers Avenger]
avenged [him Injured_party].
(2) With this, [El Cid Agent] at once
avenged [the death of his son Injury].
(3) [Hook Avenger] tries to avenge [himself Injured_party] [on Peter Pan Offender] [by becoming a second and better father Punishment].
Regardless of the kind of annotation that is done, the following guidelines apply to the annotation relative to a particular instance of a target word.
(4) Smith was surprised when Lowry retaliated for the attack.
While the above guidelines apply equally to lexicographic and full-text annotation, there are also some clear differences between the two modes of annotation.
So far, we have sub-divided our work into two parts depending on whether we choose the sentences to annotate or whether they are chosen for us by a text. Another subdivision applies specifically to our lexicographic work where we produce annotation of two different types, reflecting two different kinds of target words:
Both kinds of annotation will be discussed. Since the annotation relative to frame-bearing syntactic governors is FrameNet's main task, it will be discussed first and in greater detail in sections 3.2-3.7. Annotation relative to slot-fillers is discussed in section 3.8.
In accord with FrameNet's goals, syntactic and semantic descriptions are tailored to lexicographic description only and may differ from regular linguistic analysis as well as from shallow semantic analysis in several ways.
We classify frame elements in terms of how central they are to a particular frame, distinguishing three levels: core, peripheral, and extra-thematic. A fourth possible value for this attribute, called core-unexpressed is also discussed below.
A core frame element is one that instantiates a conceptually necessary component of a frame, while making the frame unique and different from other frames. For example, in the Revenge frame, Avenger, Punishment, Offender, Injury, and Injured_party are all core frame elements, because an avenging event necessarily includes these participants. One cannot imagine an act of revenge that is not preceded by a (perceived) offense or one that is not directed against anybody.
In determining which frame elements are considered core, we also consider some formal properties that provide evidence for core status. These properties are typically co-present, although they need not be.
Frame elements that do not introduce additional, independent or distinct events from the main reported event are characterized as peripheral. Peripheral FEs mark such notions as Time, Place, Manner, Means, Degree, and the like. They do not uniquely characterize a frame, and can be instantiated in any semantically appropriate frame. In respect to the Revenge frame, any report of an event of revenge may also include explicit information about the parameters of time, place, manner, etc. of the revenge, an example of which is given below.
(8) The bereaved family retaliated [immediately Time].
Extra-thematic frame elements situate an event against a backdrop of another state of affairs, either of an actual event or state of the same type, as illustrated with Iteration, or by evoking a larger frame within which the reported state of affairs is embedded, as shown for Containing_event.7
(9) Thou shalt not exact revenge [twice Iteration] for the same offense.
(10) The Aussies took revenge [in a penalty shootout
before 2465 fans in Long Beach the next day Containing_event].
Note that extra-thematic frame elements are understood not to conceptually belong to the frames they appear in. We take them to properly be frame elements of other abstract frames that take them as well as the targets that they modify as arguments. Thus, in example (11), we take twice and the verb phrase eat to be arguments of a more abstract Iteration frame. Similarly, in example (12), cooked dinner and me are frame elements of a Benefaction frame. Note that, as shown by (12), the native frame of the extra-thematic frame element need not be evoked by lexical material, it may simply be evoked constructionally.
(11) Learn how to spend a few extra minutes planning complementary menus where you cook once and eat [twice]. (12) Lennert, another sweetie in my life, cooked [me] dinner, mmm mmm good.
The view of extra-thematic frame elements presented here entails that these frame elements are necessarily the same across all the `host' frames in which they appear. That is, unlike core and peripheral frame elements, extra-thematic frame elements do not have a frame-specific understanding. By comparison, although many core frame elements named Agent share properties with each other due to Inheritance and Using relations, they do not necessarily have identical properties. More importantly, one cannot predict the frame role Agent from the fact that an NP has the grammatical function Ext.
(For an overview of the most frequently occurring extra-thematic frame elements, the reader is referred to Appendix 8.)
The value "Core-Unexpressed" is a special notational shorthand. It is assigned to FEs that behave like core frame elements in the frame where they are marked as Core-unexpressed but which, counter to expectation, may not be used for annotation in descendants of that frame. Frame elements marked as core-unexpressed will thus not necessarily be listed among the FEs in descendant frames.
We do not consider core-unexpressed frame elements to be violations of our definition of full inheritance. Our reasoning can be illustrated with the example of the core-unexpressed Act frame element in the Intentionally_act frame, which is exemplified in (13).
(13) I'll do [the vacuuming Act].
In the many child frames of Intentionally_act such as Choosing, Perception_active, etc., the idea of an Act is as relevant as in the Intentionally_act parent frame. However, in the child frames the frame element is absorbed by the lexical units in the frame and cannot be separately expressed.
Marking the frame element Act as core-unexpressed in the Intentionally_act parent frame allows us to keep the frames that are lower in the hierarchy from including an inherited FE which for any lexical unit in the frame could at most be annotated on the target itself, but never be expressed separately. The sentence *I chose decision the blue one is simply ungrammatical.
Coreness marking makes the most sense for event and state frames, and in these frames we use all three statuses. Coreness marking is done at the level of the frame and is intended to be consistent for all lexical units in a frame. In frames whose LUs are artifacts or natural kinds, we only use a two-way distinction (core and peripheral), noting that the values do not have exactly the same meaning as with events. In such cases, typically there is just one core frame element which is marked on the target word. For instance, in the Clothing frame the FE Clothing is core, and all other FEs are peripheral.
In providing a semantic analysis of the combinatorial possibilities of our target predicates-rather than one stated only in terms of phrase types and grammatical relations-we have observed that in an important sense frame elements are not independent of each other. Frame elements are related to the frame and required by it, as well as interrelated directly in a number of ways. These interrelationships have a direct impact on annotation, as they license the absence of core frame elements, which must normally all be accounted for in every annotation set, or (more rarely) require frame elements that might otherwise be optional. FrameNet systematically records these interrelationships. The three types of frame element relations that we recognize are discussed in sections 3.2.2.1-3.2.2.3. Section 3.2.2.4 discusses some respects in which the current treatment of frame element sets needs further refinement.
In our annotation practice, we often find that some groups of FEs seem to act like sets, in that the presence of any member of the set is sufficient to satisfy a semantic valence of the predicator. We refer to such a group of FEs as a coreness set, or CoreSet. For instance, Source, Path, and Goal are core FEs in the various motion frames in the database. However, although possible, it is not necessary, and in fact unusual, for all three FEs to co-occur, as in example (14). Sentences in many motion frames can be informationally complete and pragmatically felicitous with just one or two of the FEs expressed, as shown in (15)-(18).
(14) Fred went [from Berkeley Source] [across
North America and the Atlantic Ocean Path] [to Paris
Goal].
(15) Martha hiked [from Berkeley Source] [to Oakland Goal].
(16) Elaine walked [to Monterey Goal].
(17) I saw Peter sneak [past the guard Path].
(18) Juan was walking [out of the office Source] when I arrived.
FrameNet's normal annotation practice demands that we account for all core FEs and we could keep track of the un-instantiated FEs in the example sentences above by using null-instantiation labels as described in section (3.2.3). However, we prefer to group the FEs in CoreSets and not mark null instantiation for each member FE in cases where the FEs have an informational and conceptual interdependence. Source, Path, and Goal, for instance, are clearly related via a notion that we might call 'full path'. By contrast, omission of the Ingestibles for eat in the Ingestion frame is not related to the presence or absence of any other frame element. The sentence I'm eating __ (with my friends/in the kitchen/now) is acceptable with or without any of the frame elements given in parentheses expressed.
In some cases, the occurrence of one core FE requires that another core FE occur as well. To illustrate, in the Attaching frame Item, Goal, and Items all are core FEs. If Item occurs, then Goal is required, as shown below, where the sentence without a Goal is unacceptable. In this situation, we mark a Requires relation between the two frame elements.
(19) The robbers tied [Paul Item] [to his chair Goal].
(20) * The robbers tied [Paul Item].
The Requires relation occurs in almost all frames that have a construal alternation between a symmetric construal, when a single frame element name is used, and an asymmetric construal, when two frame elements with names of the are used. In the former case a simple name of the form [FENAME]s is used and in the latter, two FEs of the form [FENAME]_1 and [FENAME]_2 are used. Some sample frames are Compatibility, Chatting, and Similarity; there are many more.
In some cases, if one of the FEs in a group of conceptually related FEs shows up, no other FE from that group can. Again, in the Attaching frame, if Items occurs, then Item and Goal are excluded. In this situation, we say that Items excludes Item and Goal.
(21) The robbers tied [his ankles Items] together.
The above Excludes-relation in the Attaching frame is an instance of a much more common pattern of alternation between a symmetric/reciprocal and an asymmetric construal of events or states involving two parties. In most frames, where the alternation is possible, the names of the frame elements reflect the underlying alternation between reciprocal and asymmetric construal. For instance, in the Similarity frame with lexical units such as similar, different, etc. we have the frame elements Entity_1 and Entity_2, and Entities. Usually, one is allowed to infer equal participation in the event or state by the grammatically less profiled participant (Participant_2). However, since in the case of Attaching, the Goal (which would be Item_2 under our normal naming scheme) is not readily understood as itself being secured or immobilized via attachment to the Item on the asymmetric construal, we selected a name that reflects the fact that the usual inference to equal status for Participant_2 in the asymmetric construal is not warranted.
The Excludes relation also manifests in frames where an event can be brought about either by an intentional Agent or by a Cause event. Consider the following examples from the Placing frame.
(22) [The same flood tide that had brought such a
good harvest of tiles Cause] heaped a mass of driftwood onto the Reach.
(23) [Bill Agent] deposited
the bag of croissants and the Financial Times carelessly on the hall table.
The two sentences represent two different construals of Placing scenes. Sentence (22) focuses on an event as causing the change in location of the Theme, whereas sentence (23) focuses on an Agent who through their involvement in an unspecified event, most likely an intentional action involving his hands and body, causes the change of location of the Theme. The two construals are incompatible (since there is only one subject slot) and the frame elements Agent and Cause stand in an Excludes relation to each other.8
Another clear instance of the Excludes relation between frame elements occurs in the Evading frame, where an Evader moves under its own power to thereby avoid Capture or contact with a Pursuer. The Capture is an actual or hypothetical event in which the Pursuer takes physical control of the Evader. The Capture frame element and the Pursuer are thus clearly interrelated but only one of them can appear as a dependent of a target in the Evading frame.
(24) Sheriff's officials said they apprehended a gang member after he evaded [them Pursuer]. (25) He had successfully evaded [arrest Capture].
Finally, note that the Excludes relation strictly applies only to the direct syntactic dependents of a target word, that is, to first layer annotation. Frame elements that exclude each other may co-occur in an annotation set if they appear on separate annotation layers.
(26) Perkins McLain evaded [capture [through Spain Pursuer] Capture] . (27) The discussions [between [Miller Interlocutor_1] and [the dean Interlocutor_2] Interlocutors ] went nowhere.
In (26), information about the Pursuer is expressed inside the Capture frame element in a prepositional phrase dependent of the noun capture. In (27), the two sides of the discussion, Interlocutor_1 and Interlocutor_2 are expressed within the coordinate NP that encodes the Interlocutors frame element.
The treatment of frame element relations sketched in the preceding sections is adequate for a large number of frames. However, two systematic problems remain.
One problem is that we have no explicit treatment of the idea of proto-frame elements, of which other frame elements are more specialized expressions. In cases like (28), we would prefer not to have to pick either specifically Agent or Cause as constructionally null instantiated, since the context might not provide enough information to resolve that question. Instead we would make reference to a superordinate frame element (call it *Force) that is vague about intentionality and the event-person distinction. Likewise in (29), where B answers a question about a new employee, we would prefer to use a superordinate frame element (which might be named *Field in this case), rather than choosing among the frame elements Role, Skill, Knowledge, or Focal_participant for constructional null instantiation relative to the predicate good in the Expertise frame.
(28) The car got damaged while parked outside of our house.
(29) A. How's Susan working out? B. She's very good.
Having an explicit representation of proto-frame elements would also be useful in dealing with certain linguistic expressions that seem to instantiate the superordinate proto-frame element rather than one of its more specific manifestations. Consider the phrase trench to trench in (30): it does not refer to the Source or Goal of a Path but neither does it refer to a middle Path which would be compatible with the specification of a final Goal. A proto-frame element Full_path would provide the most adequate treatment.
(30) He crawled [trench to trench], looking for some sign of Stephen.
Another use for Proto-frame elements involves inheritance relations. In some cases, an inheriting frame will allow only one FE from an Excludes or CoreSet group in the parent frame. Superficially, this violates the rule that requires child frames to have a corresponding FE for each core/peripheral FE of the parent. In a deeper sense, however, inheriting only one member of a frame element set should be permitted on the understanding of inheritance as subtyping. This is so because the child frame is fulfilling every constraint of the parent, merely adding a constraint that prevents one of the construals possible in the more generic case. If we state the frame element restriction on inheritance so that it pays attention only to the most generic level of FEs, then mappings from subsidiary FEs are allowable, but not required. This would make it possible, for instance, to have a *Murder frame (with only agentive causes) as a child frame of the Killing frame (which allows Causes or Agents). 9
The second major problem that remains concerns the treatment of subject selection constructions. The current treatment of coreness requires that all frame elements that can occupy the subject position be marked as core frame elements. In many frames this leads to Means and Instrument frame elements having core status and further being part of a CoreSet with Agent, since the two frame elements may co-occur.
Consider the verb open in the Closure frame. In (31), we have a canonical Agent subject but in (32) an Instrument fills the subject slot. (Note that the frame conceptually requires an Agent; uses of open involving a Cause such as The wind opened the door are handled in another frame.)
(31) [John Agent] opened the door.
(32) [The key Bill gave him Instrument]
opened the door right away.
Given the earlier discussion of Frame Development (in section 2), Instrument should be core in the Closure frame since the noun phrase realizing it in (32) appears in the subject position. However, were there a separate way of representing the fact that English allows frame elements that are situated between the end-points of a causal chain to occur as subjects, it would not be necessary to give Instrument core status. This would avoid introducing a coreness set of Agent and Instrument.
A separate treatment of these subject selection facts would be parsimonious and would also expose the essential lexical similarity between the English frames and the frames of languages such as Japanese, where subject selection is much more restricted and Instruments and Means rarely, if ever, appear as subjects.
Sometimes FEs that are conceptually salient do not show up as lexical or phrasal material in the sentence chosen for annotation. Nevertheless, we indicate their absence since it provides lexicographically relevant information regarding omissibility conditions. The FE that has been identified indicates which semantic role the missing element would fill, if it were present.
With respect to null instantiation, verbal, adjectival, and prepositional targets are treated identically. (For null instantiation with noun targets, see 3.4.3.) The following examples show omitted elements with each part of speech. (The name of the frame element is given in square brackets and the frame of the target is given in parentheses.)
(33) That will suffice . [DNI Standard] (Sufficiency)
(34) The result should be similar . [DNI Entity_2] (Similarity)
(35) I tried to put the toys back in . [DNI Ground] (Locative_relation)
Not all cases of frame element omission are alike. We recognize three different cases, one that is not lexically specific and two that are. We will now discuss the three types of omission, focusing on null instantiation with verbs where the phenomenon is clearest.
The first type of lexically specific null instantiation to be considered is the definite (or anaphoric) type. Cases of definite null instantiation are those in which the missing element must be something that is already understood in the linguistic or discourse context. In the following example, the Offender is not expressed overtly in the syntax, but its referent has to be known in the context.
(36) [The monkey Avenger] avenged [himself Injured_party] [by growing to the size of a giant and setting fire to the city Punishment]. [Offender DNI]
The indefinite cases (sometimes also referred to as existential) are illustrated by the missing objects of verbs like eat, sew, bake, drink, etc., that is, cases in which these ordinarily transitive verbs can be spoken of as used intransitively. (e.g. Molly rarely eats alone; Granny begins baking about a month before Christmas; Michael even drinks heavily on weeknights.) As is well known, there are often special interpretations of the existentially understood missing objects. For example, with eat the missing entity is likely to be a meal, with bake it is likely to be flour-based foods, with drink it is likely to be alcoholic beverages, etc. In contrast to anaphoric omissions, with existential omissions, the nature (or at least the semantic type) of the missing element can be understood given conventions of interpretation, but there is no need to retrieve or construct a specific discourse referent.
For example, in the Revenge frame, all lexical units happen to allow the frame element Punishment to be omitted under indefinite null instantiation. This is shown for avenge in (37).
(37) He took it out on Scarlet in the same way as [he Avenger] avenged [himself Injured_party] [on her Offender] [for the pressures at work and the demands of his first wife Injury] . [INI Punishment]
Note that both in the case of definite and indefinite null instantiation, the LUs in a frame may differ from each other in whether or not they allow the omission. For instance, while eat allows its object to be omitted, devour does not, even though they are both in the Ingestion frame.
Verbs that usually require an argument to be present (or only allow it to be omitted under conditions of definiteness) can be used in a generic construction with indefinite null instantion, as shown below.
(38) He takes and never gives back. [INI Theme]
Figure 38 shows the FrameNet Desktop opened for the annotation of a case of definite null instantiation. Notice the tab to the right of the frame element Offender in the picture, which shows that the FE was omitted under definite null instantiation. In addition, an appropriately colored DNI tag appears at the end of the sentence in the corpus viewing section of the FNDesktop.
Constructionally omitted constituents (also called structurally omitted) have their omission licensed by a grammatical construction in which the target word appears, and are therefore more or less independent of the LU. Cases of CNI include: the omitted subject of imperative sentences, the omitted agent of passive sentences, the omitted subjects of independent gerunds and infinitives (i.e., the PRO-elements of generative grammar), and so on. In each of the following two examples, the FE Avenger is tagged with the symbol CNI.
(39) Family feuds last for generations, and [slurs on honor Injury] are avenged [by murder Punishment] . [CNI Avenger]
(40) Get even [with her Offender] [for this Injury] [CNI Avenger]
In addition, we use CNI for missing objects in instructional imperatives such as exemplified below, even though the omission is not dependent on a particular construction, but rather on particular genres, such as cookbooks and product labels.
The experiential perfect also licenses object omissions that are not possible in simple assertions of frame instances.
(43) Have you ever fostered [CNI Child] before?
Note that particular constructions licensing argument omission specify particular intepretations, either indefinite (existential) or definite (anaphoric). For instance, the instructional imperative construction in (41) and (42) specifies a definite interpretation, whereas the experiential perfect exemplified in (43) specifies an indefinite interpretation. The CNI label thus collapses the interpretational distinction that we make among the lexically licensed omissions; it does not represent a separate kind of interpretation in addition to the definite and indefinite types.
In some cases, information about two frame elements is expressed in a single constituent, a situation we call conflation. For instance, the concept of ousting somebody from office requires an understanding of the incumbent of the office and the identity of the office; each can be represented separately in a sentence like We ousted Jones as mayor. But in We ousted the mayor, the direct object stands for both the office and the incumbent.
We also find examples of frame element conflation in the Revenge frame. In particular, the Injured_party may be contained as a possessive in the phrase that realizes the Injury frame element, as seen in the following example:
(44) [He Avenger] avenged [Pedro's death Injury] [by taking out the poker-faced Guards Officer Punishment].
Here, the possessive Pedro's realizes the frame element Injured_party, the person who suffered the Injury. In such cases, the annotation tool allows for the creation of an additional FE layer, enabling the secondary annotation of (parts of) constituents in the same frame, as shown in Figure 44.
Note that there is never a phrase type or grammatical function indicated for the frame elements on the secondary FE layer.Even with conflation, it is still possible for the FE annotated on the secondary layer to be instantiated as a constituent of its own in the syntax. In example (45), information about the person who was hit is encoded by the direct object me, as well as by the possessive determiner my inside the NP complement of the prepositionon.
(45) He hit me on my hand.
In such cases, we tag the possessive on the second layer with the same frame element label that is applied on the first annotation layer to the object of the verb.
In general, we select sentences for annotation where, with the exception of subjects, we find all frame elements realized by constituents that are part of the maximal phrase headed by the target word. There are two types of situations in which we annotate non-local constituents with frame element labels. In each case the motivation for annotating constituents that bear no syntactic relation to the target is lexicographic: the non-local constituents contain lexical material and as such are of interest to the study of collocations since they provide more information about the semantic type of the frame element than the locally occurring co-indexed phrases or empty elements do.
The first case in which we annotate non-local constituents consists of cases in which the target word is syntactically governed by a raising or control predicate. In such cases, the valence properties of the higher raising or control predicate guarantee that one of its arguments is also interpreted as an argument of the target, even though the relevant argument is not dominated by the maximal phrase headed by the target. Some of the most common types of control and raising predicates are illustrated below. The control or raising predicate that guarantees the interpretation of the non-local noun phrase as a frame element of the target appears in typewriter font.
Subject to object (46) We expect [John Avenger] to retaliate [against us Offender] [INI Punishment] [DNI Injury].
Subject to subject (47) [John Avenger] seems to have avenged [the death of his brother Injury] [by luring Smithers into a trap Punishment].
Subject control (48) [They Avenger] are hoping to get even [with Smithers Offender] [for the insult Injury].
Object control (49) The commander ordered [the troops Avenger] not to retaliate [against the rebels Offender]
Note that Raising and Control cases are not restricted to verbal controllers, i.e. nouns may also serve that function, as illustrated here.
(51) Only a short few weeks ago, even [my Perceiver] hope of seeing [her Phenomenon] was just a dream [Subject control].
(52) Meanwhile, today, Americans, hungering for victory, are puzzling over the Pentagon's order to [the troops Agent]not to put [the Stars and Stripes Theme] [on their vehicles Goal]. (Object Control)
(53) The testers gave it a clear thumbs up both for [its Created_entity] ease of assembly and sail performance. [Tough movement]
The second case in which we annotate clearly non-local constituents with frame element labels concerns targets that occur inside relative clauses. Here our policy is to tag not only the constituent containing the relativizer (if there is one) as a frame element but to also repeat the FE/GF/PT triple on the antecedent, and to further mark the relative word and the antecedent phrase on the Other layer. Thus, our annotation for the simplest cases with an overt relative word is as shown in Figures 53 and 53.
Note that, in contrast to the examples above, the relative word is not always by itself a phrasal constituent. As Figure 53 shows, the constituent containing the relative word may be complex. Similarly, the antecedent phrase may be a rather complex phrase, as shown in Figure 53. When a target occurs in a relative clause without an overt relativizer, as in Figure 53, we only annotate the antecedent phrase and mark it with the label Ant on the Other layer. Since we do not assume any kind of zero or non-overt relativizer, the label Rel is not applied anywhere on the Other layer. If a relative phrase is governed by a preposition, we end up with identical FE/PT/GF triples applied to antecedent phrase and the prepositional phrase containing the relative word; the Ant and Rel labels are applied as usual, as shown in Figure 53. In sentences with preposition stranding, the same FE/PT/GF triple occurs on the antecedent phrase, the noun phrase containing the relative word (if there is one), and the preposition. The preposition never gets a Rel label assigned to it even in cases where there is no overt relativizer, as in Figure 53. Notice also that the above principles for relative clauses carry over to Gov-X annotation. When the target noun is the antecedent for a relative phrase that is an argument of a verb annnotated as a governor, we split the antecedent and relative phrases in the same ways as illustrated for cases of normal frame element annotation and also apply Ant and Rel labels on the Other layer in the usual fashion. An example of a governor occurring in a zero-marked object-relative clause modifying the target noun is given in Figure 53. Finally, infinitival relative clauses with an overt relativizer are treated just like finite relative clauses with an overt relativizer, which is illustrated in Figure 53. Infinitival relative clauses without relativizer, by contrast, receive no Ant-Rel marking at all. For instance, in books to read over the break the head that the relative clause modifies does not receive an Ant label.When annotating verbal targets, we do not record any predicates that may govern them. However, in the case of nominal, adjectival, and prepositional targets there exist several special classes of syntactic governors that we want to keep track of for lexicographic reasons.
In the case of Support expressions, Copulas and Controllers, it would have been theoretically justifiable to omit selecting phrases outside of the standard subcategorization frame of the target noun and to instead rely on automatic tools for syntactic analysis to identify phrases outside the target's maximal projection that give information about the filler of a frame element role. However, since one of our goals is to provide a database that includes samples of phrases capable of satisfying particular FE requirements of the analyzed words, our decision was to increase the scope of our annotation instead. A welcome by-product of this decision is that the FN database can also serve as a resource for identifying the Support verbs and prepositions, Copulas, Controllers, and X-Governors that FN annotators often find accompanying particular noun, adjective, and preposition targets.
As noted at the beginning of section (3.2), we have a special treatment for sentences in which the syntactic and the semantic head of a clause are different and where a noun target is the semantic head of the clause rather than the verb that governs it syntactically. In these cases, one or more syntactic core arguments of the support verb are necessarily understood as participants in the event or relation evoked by the target noun. These verbal arguments-typically the subject, in some cases the object, in others both the object and the subject-are tagged with labels appropriate to the noun's frame. Examples (54)-(56) exemplify support verb constructions:
(54) Aloha Gang, [Someone Speaker]
made a statement about my need two kerrect my
shpelling in ze
last newsletter.
(55) [Frances Patterson Patient] underwent an operation at RMH today and is expected to be hospitalized for a week ore more.
(56) [One of them] became my
successor in the professorship in the University of
Michigan and the presidency of Cornell.
By contrast, the verb-noun combinations in (57)-(59) do not involve support structures.
(57) Did you read about his latest mishap in the newspaper?
(58) A senior nurse observed the operation.
(59) John congratulated the new president.
In all three examples (57)-(59), the verbal predicate governing the target noun introduces a distinct event: reading about a mishap is completely independent from participating in it (57); observing something is independent from participating in it (58); and congratulating somebody is independent from the achievement at issue (59). (Actually, relational nouns like president never take support verbs, they project clauses only in combination with copular verbs such as be or become.)
Support verb+noun constructions are not to be equated with idioms. While support verb-noun combinations, too, may involve some measure of non-compositionality, it is normally much less than with true idioms whose meanings cannot be built up straightforwardly from the normal meanings of their parts (e.g. give walking papers/a pink slip/the boot). To classify a verb-noun combination as a support verb+noun construction, it is necessary for
Note that elements tagged as Support ultimately will also receive a separate treatment in their own right. Thus, any differences that exist between support predicates (including most saliently the introduction of causation) will be captured by describing these predicates as frame evoking elements in very generic frames and adding the semantic type Support (see further under example 13) to indicate their limited use. In this spirit, the verb lift as used in the sequence 'lift the UN sanctions on the country' was included in the Cause_to_end frame alongside the very general LU end.v.
Verbs are not the only part of speech that can `support' a noun. In some cases, prepositions combine with nouns to yield phrases that behave like predicative adjectives. That is, they can post-modify a head noun, as in (60), or combine with a copular verb to yield finite verb phrases, as is shown in (61)-(63).
(60) The "possessor" is the person in possession of the premises.
(61) Are health care workers at risk of getting HIV on the job?
(62) Soon, I was in possession of two dozen
Eagles cupcakes decorated with white icing,
green sprinkles and little plastic footballs and Eagles helmets.
(63) Some people might think that's out
of line with our "democratization" policy.
We may think of be and a few other verbs such as appear, seem, look etc. as a special subtype of support verb with a very minimal semantics when occurring in constructions of the form:
(i) NP1 Verb NP2/AJP/PP
Examples include:
(64) John is a sailor.
(65) This seemed a rather redundant effort to many.
(66) Tom appears smart enough.
(67) Massu looked without energy, he looked defeated seated with the towl on his face.
(68) Smithers is the vice-president of the armchair division.
(69) Sue is the mayor.
Appearing in sentence frame (i), these verbs are traditionally called copulas or linking verbs. Be occurs in many other sentence frames too and there are some cases of structural ambiguity where it is a copular verb under one reading, and an auxiliary under the other. In His pastime is annoying the girls, one can understand be either as a copula that pairs the role noun with the role filler, or as an auxiliary of the verb annoy used in the present progressive.
Note some uses of be with a target noun do not bear the label Copula. For example, when be occurs as part of the existential construction, it is tagged with the Exist label. In future data releases, there be occurring in the existential construction will be treated as multi-word Supp (cf. section 3.4.1.2).
As shown below in Figure 69, the label Copula appears on the part-of-speech specific layer of a (non-verb) target, in this case the adjective old.
Recently, FrameNet has begun to recognize a new category of syntactic governor called Controller, abbreviated in the data as Ctrlr. It covers verbs like merit and offer when they govern event noun targets, as in (70) and (71), as well as verbs like consider and find when they govern adjectival targets, as in (72) and (73). While these predicates introduce a distinct event from that of the target, they do share a frame element with the event of the target. For Controllers of noun targets the shared frame element is typically the subject of the Controller; for Controllers of adjectives of the shared frame element is typically the object of the Controller. The constituent expressing that shared participant is labeled with a frame element relative to the noun target.
(70) [What I paid and the satisfaction received] merits high praise. (71) [The minister] offered help to get the various agencies coordinated. (72) She understood that he considered [the trip] too expensive for them both. (73) I found [her scenes] very funny.
In (70) the complex NP What I paid and the satisfaction received is tagged as the Reason frame element of the Judgment_communication frame evoked by praise and in (71), the minister is tagged as the Helper frame element of the Assistance frame evoked by help. In (72), the NP the trip is labeled as the FE Goods in the Expensiveness frame evoked by expensive. In (73), the object of the Controller find, the phrase her scenes, is labeled as the Stimulus of the Subject_stimulus frame evoked by funny.
The Controller label is applied on the Noun and Adj layers just like the Supp label is.
In the context of slot-filler annotation (see 3.8), FrameNet also uses a category Governor, which marks predicates that have a semantic connection to the qualia structure of the target artifact nouns they govern. For instance, stab is treated as a Governor of knife in the Weapon frame. The dependent constituent of the Governor that is headed by the target noun is called X for lack of a better term. No dependents of the Governor other than the X constituent are annotated relative to the Governor. Any frame elements of the frame evoked by the artifact noun that are realized within the X-phrase are annotated with frame elements as usual. As with other kinds of special governors, the Governor label, as well as the X label, are applied on the part of speech specific Noun layer.
In some cases, the same FE label appears multiple times relative to a given target. There are two cases: multiple separate instances of the same frame element, as when several Path segments are described for a motion event (Josh ran [across the meadow PATH], [along the creek PATH] and right up to Bill's barn); a single instance of a frame element is realized in two discontinuous pieces, rather than as a single constituent. Here, we consider the latter type of discontinuos FE.
(76) [The Peacock Throne NP.Ext] has been sat [on PP.Dep] by the Iran monarchy since the days of Nadir Shah.
(77) [Where she had been PP.Dep] she didn't talk [about PP.Dep] (Topicalization) (78) [The man PP.Dep] [that PP.Dep] I lent my phone book [to PP.Dep] made off with it. (Relativization) (79) [What PP.Dep] was Bill angry [about PP.Dep] when he got the back spasm? (Wh-extraction)
(80) About this time [a guy NP.Ext] appeared [who had only a pair of boxers on NP.Ext]. (81) [Mrs Fiorentina NP.Ext] spoke next, [the chairwoman of Huelit-Peccard NP.Ext].
(82) [Mr Burns NP.Ext] stepped out of the shower, {buck-naked}. (83) [Smithers NP.Ext] showed up later {with a funny hat on}
(84) [Looking stunned Depictive], Mr Burns dialed his secretary. (85) Mr Burns dialed his secretary, [looking stunned Depictive]. (86) Mr Burns, [looking stunned Depictive], dialed his secretary. (87) *Who looked confused, Mr Smith came in. (88) *The former mayor of Denver, Wellington Webb was spotted at the buffet.
(89) Quite bluntly, you're a [dumber Descriptor.Dep.AJP] man [than I Descriptor.Dep.PP]! (90) The general has committed a [similar Descriptor.Dep.AJP] blunder [to our entry into the eastern valley Descriptor.Dep.PP]. (91) Yet what your school did to them [so Degree.Dep.AVP] irked them [that they'd rather pay the new school than pay the old school Degree.Dep.Sfin]. (92) Real people sometimes prove to be [more Degree.Dep.AVP] annoying [than anyone anticipated Degree.Dep.Sub].
Frames can be evoked by words in any of the major lexical categories of noun, verb, and adjective, as well as by adverbs and prepositions. We will begin our discussion with verbs.
Annotation is easiest when all and only the core
frame elements (the conceptually necessary participants of the frame that a syntactic governor evokes) find syntactic expression in the sentence as separate immediate syntactic dependents of the governor. Under such circumstances, we simply annotate each of the syntactic dependents for the three kinds of information: Frame Element (that is, semantic role), Grammatical Function, and Phrase Type.
The different kinds of information are recorded on separate annotation layers. The separation of layers makes it possible to represent many complex situations, such as when the constituent that realizes one frame element is contained within the constituent that realizes another, or when the semantic and syntactic constituency don't match. Usually, however, most tagged constituents consist of triples of information, with coterminous tags on three layers, and they are displayed as such in the FrameNet Desktop. Figure 3.3.1 is a screen shot of the middle window of the annotation software in which an example sentence has been annotated. The names of the layers appear in the left-most column of the bottom frame: FE (Frame Element); GF (Grammatical Function); and PT (Phrase Type); Other (labels that deal with a small set of special syntactic constructions); and Verb (a layer named after, and with labels specific to, the part-of-speech of the target)
In practice, annotators only need to apply a Frame Element label; Grammatical Function and Phrase Type are derived algorithmically based on position relative to the verb and patterns of part-of-speech labels, but may require manual correction.In addition to core frame elements, we also provide annotation for some non-core frame elements expressed in the sentences selected for annotation. Non-core frame elements are not necessary conceptually in the sense that they do not uniquely characterize the frame. Thus, in most frames denoting events or processes, Time and Place frame elements are not core, and therefore may not always be annotated. (For a more complete discussion of core vs. non-core, see Section 3.2.1). Similarly, actions often have a Purpose that the Agent intends to accomplish by performing the action indicated by the target word. However, having a purpose doesn't usually distinguish types of actions and so Purpose is often a non-core frame element, as in the following example:
(93) They wake you up [to give you a sleeping pill Purpose].
Grammatically, non-core elements cannot be nuclear arguments (subject or object) of target verbs, and they frequently are adverbs or prepositional phrases.11
Unfortunately, not all sentences are as straightforward to annotate as the ones we have seen so far. We will now discuss how FrameNet deals with various challenging cases.
Some syntactic constructions require the presence of non-referential material in an argument position, even though the non-referential material has no semantic relationship to the predicate.12 In a subset of cases, this happens while a semantic argument is dislocated to a non-canonical position. The non-referential items appearing in such constructions are called expletives, with it and there being the two kinds of non-referential NPs in English. They never instantiate frame elements, so are not not given GF and PT tags. We do, however, record their presence with the Null label on the Other layer. Some typical instances of expletives are given in the following examples.
Subject extraposition (94) [It Null] is clear that we won't finish on time.
Object extraposition (95) I hate [it Null] when you do that.
Existential construction (96) [There Null] are more hats in the closet.
Subject requirement of zero-place predicates (97) [It Null]'s raining.
When verb targets co-occur with particles that are used productively to indicate aspectual information, we tag the particle with the label Aspect on the `Verb'-Layer. Some examples follow.
(98) They were chattering [away Aspect] in the kitchen, when the door bell rang. (99) Mo talked [on and on Aspect].
However, particles that express aspectual meanings and simultaneously form entrenched lexical units with the verb are not treated in this way. For instance, in examples such as (100), pull through would be treated as one lexical unit, rather than as an instance of the verb pull accompanied by aspectual through since the verb cannot occur by itself with the appropriate meaning.
(100) It's amazing what they can do these days and luckily the sick dog pulled through.
FrameNet also annotates relative to noun targets. In the course of our work, we have analyzed several types of nouns: nouns that denote events such as withdrawal and replacement; relational nouns such as brother or girlfriend; artifact nouns such as house and vest; and some others. Of these, event nouns and relational nouns are most clearly frame-evoking. Many of the issues discussed earlier for verbs, e.g. conflated frame elements, discontinuous frame elements, or coreness statuses apply to nouns in the same or in similar ways as to verbs. However, the annotation of nouns also brings some new challenges to the annotation enterprise.
One of the major challenges in annotating nouns is that many of them participate in lexicographically relevant relationships with their governing predicates. These relationships are of four major types. Support predicates are governors of event-denoting nouns that serve mainly to project a clause centered on the frame of the noun. Existential predicates occur with event noun targets in the there-existential construction. Copular verbs are a semantically very bleached type of support expression used to project clauses centered on the frames of event or relation-denoting nouns. Controllers refer to an event that is separate from the one denoted by the target noun but which shares a participant with it. X-Governors of artifact nouns are verbs that evoke frames that involve the qualia structure of the artifact nouns.
We define support verbs as those verbs that combine with a state noun or an event noun to create a verbal predicate, allowing arguments of the verb to fill the slots of the frame elements of the frame evoked by the noun. Consider some examples from the Revenge frame.
(101) The Americans must have felt as if he was [taking Supp] revenge on them for what had happened. (102) King Menephta [took Supp] awful revenge on a Libyan army he defeated around 1300 BC
Both of these examples report an act of revenge rather than an act of taking, the frame evoked by the noun revenge clearly being dominant. We treat verbs like take in a special way by marking them as support verbs on the Noun layer. The intuition behind this treatment is basically that support verbs do not introduce any significant semantics of their own. Constructions of nouns with their support verbs denote the same state of affairs that would be denoted by the noun occurring by itself.
In the examples above, recognizing take as a Support verb allows us to annotate its subject as the Avenger frame element. However, a frame element of the target noun that a support verb expresses does not have to be realized as the subject of the Support verb. Objects and other complements of support verbs may also express frame elements of the target word. This last fact allows us to circumvent the problem of certain disputed constituency decisions that come up in the case of support-verb constructions. For instance, in example (103) it is contestable whether the phrase to the press is a complement of the noun statement or the verb write. (In contrast, with the support verb make, the issue would not arise as make by itself does not take to-PP complements.)
(103) He [wrote Supp] a statement to the press about the bribery case.
Regardless of the analysis, FrameNet tags the phrase to the press with the FE Addressee as a complement of the target word statement. By allowing this phrase under either of the disputed analyses, we avoid the constituency decision completely.
The practice of annotating support verbs not only allows us to annotate their subjects as frame elements. It is also lexicographically necessary to record them. One reason for this is that support verbs are selected by the noun, rather than the other way around. For instance, while the noun question can take the support verb pose, other nouns in the Questioning frame such as query and inquiry take the support verbs make. Support verbs also vary with the sense of the noun, that is, a noun may take different support verbs depending on the frame it belongs to. Consider that in (104), the noun argument takes the support verb have, and has a meaning related to conversation, whereas in (105), the noun takes the support verb make, and has a meaning related to reasoning.
(104) John and I [had Supp] a terrible argument last night. (105) John [made Supp] a convincing argument that the project should be funded.
With regards to their semantic contribution, we recognize several types of support predicates:
Note that the above classification is an informal one that is not encoded in the database through the use of different support labels. Further, the above divsion is clearly much less finegrained than the distinctions that exist among Lexical Functions in the sense of Igor Mel'cuk[Mel'čuk, 1996]. An extension of FrameNet could be defined which sought to distinguish the full range of Lexical Functions.
In addition to support verbs, we recognize a second type of support expression, namely combinations of nouns with support prepositions. Support prepositions combine with certain target nouns to yield a phrase that is more or less equivalent to a predicative adjective.14 As in the case of support verbs, the frame of the noun is dominant and it is the noun that selects the particular support preposition, rather than the other way around. For instance, while the noun danger in the Run_risk frame is supported by the preposition in, the noun risk in the same frame requires the support preposition at.
When a target noun occurs with a support preposition and projects a finite clause, we also apply the Copula label to the copular verb that combines with the preposition and noun to form a verb phrase. Some examples are given below.
(106) The painting [is Cop] [on Supp] loan from Mr Smithers (107) About 650 oil wells [are Cop] [on Supp] fire. (108) This old book [is Cop] now [in Supp] my possession. (109) The documents [came Cop] [into Supp] my possession by inheritance.
Notice, though, that there need not be a copular verb for a target noun to occur with a support preposition. When the combination of support preposition and noun is used as a depictive or resultative secondary predicate, it is not accompanied by a copular verb. Illustrative examples from the Facial_expression frame are given here.
(110) His face was twisted [in Supp] a grimace. (111) He wrinkled his brow [into Supp] a frown.
Usually, such sentences are roughly parallel to sentences with a support verb. For the two preceding examples, make is a good candidate.
(112) His face was twisted, [making Supp] a grimace. (113) He wrinkled his brow, [making Supp] a frown.
Finally, notice that in accordance with their status as syntactic governors for target nouns, support expressions are not tagged with frame element labels. They are tagged on the Noun layer of the target nouns and have no grammatical function or phrase type assigned to them, as shown in Figure 113.
A subset of nouns can occur in a there-construction that serves to introduce an instance of the frame evoked by the noun target into the discourse. FrameNet so far has recorded the governing verbs occurring in the there-construction by applying the label Existential (abbreviated Exist in the data) to the relevant verb on the Other layer. Usually the verb is be, but others, such as come, occur, and ensue are found as well. Examples are given in (114) and (115).
(114) There [was Exist] an argument between the players. (115) There [ensued Exist] a discussion about the current program.
The word there that occurs before the existential verb is marked with the Null tag, which like the Exist tag is applied on the Other Layer, as is shown in Figure 115 .(See section 3.3.2 for more information on the Null tag).
The Exist label was introduced earlier and separately from the Supp label. In the release following Release 1.3, we intend to convert all instances of the Exist label to Supp. The reason for doing so is that with the exception of be all other verbs occurring in the there-existential construction can also occur as Support verbs of nouns outside of the there-construction, as is shown by examples (116) and (117).(116) Another private wrestled the revolver away from Czolgosz while a melee ensued. (117) The robbery occurred on Smithfield road six miles east of Campbellton.
The combination of there and be will be treated as an idiomatic multi-word support-it is actually the same as the lexical unit there be in the Existence frame. That is, in there be, the word there will in the future be labeled as Supp on the Noun layer rather than as Null on the Other layer.
As pointed out above in section 3.2.6.2, verbs like be, seem, appear and a small number of others can be considered to be a special, rather bleached kind of support verb. However, unlike the semantically heavier support verbs, copular verbs can occur with relational nouns such as chairperson (118), not just with event-denoting nouns such as proposal in (119).
(118) Jeanne Hossenlopp will be the next chairperson of the department. (119) His proposal was that Carthage should be destroyed.
Copular verbs occur in two distinct constructional contexts. This is illustrated by the examples in (120)-(123).
(120) Jim is a tailor.
(121) This seemed a silly idea to me.
(122) Jeidels is the director of the physics department.
(123) Paul is the treasurer.
The four example sentences, though superficially the same, instantiate two different constructions. The first two examples, (120) to (121), are predicative or ascriptive. In (120), the property of being a tailor is ascribed to Jim; in (121), the property of being a silly idea is ascribed to the action anaphorically referred to by this. The second use, illustrated in (122) and (123), is specifying or equational. In (122), the identity of the referents of Jeidels and the director of the physics department is asserted; in (123), the identity of Paul and the filler of the treasurer role is asserted. In specifying sentences, NP1 and NP2 can be switched, with the resulting sentence still meaningful and even having the same truth conditions. This is not possible with predicational sentences, as is shown by (120') and (121').
(122') The director of the physics department is Jeidels.
(123') The treasurer is Paul.
(120') *A tailor is Jim.
(121') *A silly idea seemed this to me.
FrameNet annotation practice does not distinguish the above two constructions.15 Be is tagged as copula in both cases.
Controllers are the third major kind of special governor that FrameNet records. They are different from Support expressions and Copular verbs in that they evoke a separate frame from the one evoked by the target noun while still having a shared participant. Consider examples (124) and (125).
(124) In this case it means that the bad deed deserves revenge. (125) Bill offered help in the case of an emergency.
Example (124) is not a report of actual or imagined revenge. It is only asserted that revenge is warranted for the bad deed referred to. However, the bad deed clearly would be the Injury that an act of revenge would be intended to punish. The situation is similar in (125): offering help is not the same as actually helping. But somebody who offers help would have to deliver help, that is, be a Helper in an actual instance of Assistance, if called upon.
When a Controller verbs like deserve and offer is present, then the constituent representing the shared participant, typically the subject of the Controller, is annotated with a frame element label relative to the noun target. Just like the Supp label the Controller label is applied on the (part of speech-specific) Noun layer.
(126) The Prime Minister encouraged revenge against Absurdistan. (127) We welcome the Elbonian decision to withdraw behind the lines of 1328.
Example (126) also does not assert an instance of the event evoked by the target noun. Encouraging somebody to do something is different from doing it and, crucially, the encourager need not even be a participant at all in the encouraged act. Likewise, in example (127), welcoming a decision is unrelated to making a decision and two such events normally share no participants, since people do not normally talk about the felicity of their own decisions. Verbs like encourage and welcome will most likely continue to go unlabeled, although we are considering the possibility of introducing a category Concomitant for verbs that evoke a frame that is related via a background scenario to the frame evoked by the event noun. For instance, to grade an exam is not the same as to give or take one. But since evaluation is a necessary part of the overall scenario of examination, it seems desirable to record the close connection between grade and exam. Another example of a Concomitant is the verb call when it governs the noun election (in the context of British politics). The person calling the election need not vote or stand for election but clearly the calling of the election is an integral part of the electoral process.
Our treatment of noun compounds recognizes, but does not easily cover, the full range of problems that they exhibit, which any theory of the grammar of English must address. In practice, we distinguish compounds that are treated as single LUs, for which FrameNet (at least currently) gives no account of their substructure (e.g. firing squad, sugar daddy, wine bottle), from compounds based on frame-evoking nouns whose modifiers happen to be nouns or relational adjectives (e.g. restoration costs, military supremacy). Almost all noun compounds are motivated in that their components have clearly defined semantic relations to each other, whether or not the pattern they exemplify is productive. In addition, the separation between conventionalized two-part words and examples of productive processes is not always clear, though there are clear cases. Compounds like fire risk and health risk would be treated with the noun risk, and the modifiers would be given FE labels that follow a description of the head noun's frame. Similar treatment is given for the compounds language pedagogy, animal sacrifice, and water flow, where the modifier is a noun, as well as presidential privilege, economic problem, educational crisis, etc., where the modifier is a relational adjective.
Note that in the context of noun-noun compounds, we never annotate the head noun as a frame element of a frame that may be evoked by the non-head. Thus, for a compound like weapons treaty we annotate weapons as a frame element relative to the noun treaty (which belongs to a Documents frame), but we do not annotate treaty as a frame element relative to weapons. While the non-head must figure in some frame evoked by the head, the reverse is not true in the same way; there is no sense that weapons saliently and consistently causes speakers to think about how their manufacture and possession is regulated. We consider as accidental any compounds in which the head noun superficially looks as if it fills a frame element role in a frame evoked by the non-head. Thus, although in the compound frying pan the noun pan appears to express the Container frame element in the Apply_heat frame evoked by fry(ing), we do not annotate the head pan as a frame element of the non-head frying. The policy exemplified here for noun-noun compounds also applies to compounds consisting of a relational adjective and a head noun. Thus, while we annotate the relational adjective relative to the head noun of the noun phrase in [economic] policy and [military] might, we do not annotate the head noun relative to the relational adjective (*economic [policy]; *military [might].)
Both in the case of noun-noun compounds and of compounds consisting of relational adjective and noun, we may still annotate the head as a Governor of the non-head target (cf. section 3.8 on slot-filler annotation). Also, in a compound, we always annotate any dependents of a non-head target that occur to its left. Consider the annotation of [food] processing facilities, where food is tagged as Material with respect to the Processing_materials frame evoked by processing but the head noun facilities is unlabeled. Notice that this contrasts with our normal annotation for pre-modifying uses of non-relational, predicative adjectives. For instance, when related in its Cognitive_connection sense occurs as a pre-modifier of a noun, we annotate both any preceding noun that may fill its Concept_1 role as well as the head that the adjective modifies: [work]-related [stress].
Figure 3.4.2 shows our treatment of the compound noun firing squad as a single lexical unit. Notice that the whole compound is the target (indicated with capital letters), and is annotated with the FE Executioner. (The annotation on the target is not visible on black and white print-outs.)
Figure 3.4.2 shows our treatment of the compound noun fertility rite as a head noun modified by a noun. Notice that the target word rite is modified by a noun that is annotated with the FE Desired_state.In the case of noun targets, null instantiation is very common, and it is much more difficult to decide what licenses the absence of the noun's conceptually necessary arguments. In particular, quantification and generic use often make the notion of DNI inapplicable since they tend to prevent the individuation of particular events or states and their participants. Consider the following pair of examples, which illustrate this effect of quantification and generic construal for a verbal and a nominal target.
(128) Every time Max did something like that, I knew how to get even. (129) Revenge is sweet.
However, even without quantification, there are no reliable clues from definiteness marking about how frame elements of a target noun are contextually known. Consider the following example sentence uttered in a courtroom context, phrased with a verbal target.
(130) Smithers was convicted after two hours of deliberation.
Now, consider an example with the noun conviction.
(131) His role in that conviction is now under investigation.
It is clear that in the Verdict frame, whose LUs include the noun conviction and the verb convict, the FE Charges is a core frame element. The example in (130) with a verbal target is felicitous only if the FE Charges is contextually recoverable. By contrast, a sentence like (131) with the nominal target conviction is felicitous even if the Charges are not fully recoverable in context. This is demonstrated by the fact that sentence (131) could be preceded by either one of the following discourses:
(131') We have reason to believe that this agent has previously tampered with evidence to get a conviction. Just two months ago the murder conviction of Howie Cheatham was overturned. Agent Smith also was a witness there. (131") We know that this agent has previously tampered with evidence to get a conviction. There is, for instance, the case of a man in Missouri that had to be released after it was found that agent Smith had manipulated fingerprints.
In the first example, the Charges (murder) are explicitly mentioned. In the second, they are not identified at all, just existentially bound to the event of the earlier trial and conviction. Thus, reference to an event with a definite event noun does not require that speaker and hearer can resolve all aspects of an event to entities that they know of independently of the event talked about by the target word under consideration.
Basically, it seems to be the case that there are no frame-evoking nouns that always require the expression of some or all of their frame elements. Still, there are situations where it is clear that a particular frame element has to be interpreted as omitted under DNI. For instance, relational nouns like the kinship terms brother, sister, etc., are most commonly used to refer to specific people in non-generic contexts and when this is so, we annotate omitted relata as DNI. An example of this is given in (132).
(132) The [brother ALTER] was found not guilty [DNI EGO].
Likewise, when an event noun projects a finite clause in combination with a support verb and the event is not construed generically, FN records frame elements that have to be recoverable in the discourse or the discourse setting as DNI. Thus, we record DNI for the Goal frame element in example (133), which refers to a specific occasion of entering, but keep no record, not even INI, of the frame element in the habitual sentence (134).
(133) He made his entrance singing "Oh, What a Beautiful Mornin" [DNI GOAL]. (134) He made his entrances and his exits like the man of the stage he was and never forgot his theatrical background.
While there may be other constructional and interpretational contexts in which a non-realized frame element of a target noun clearly receives a DNI interpretation, we currently only record DNI for noun targets in the above two cases: for specifically referring relational nouns and event nouns that are used in finite clauses with support verbs to refer to specific occasions.
Adjectives can also evoke frames, that is, be frame-evoking. Certain semantic areas such as emotion- or evaluation-related frames actually have a considerable share of adjectival lexical units. FrameNet annotates adjectives, both when they are used attributively and when they are used predicatively. For the most part, adjectives take the same kinds of phrases as dependents that verbs or nouns do: finite clauses (Phil is happy [that he passed]); infinitival clauses (Phil is eager [to finish]); gerundive forms (Phil is busy [writing]); prepositional phrases (Phil is curious [about the new student]); adverbial phrases (Phil is [rather] disappointed); etc. However, predicative adjectives are different from verbs in that they do not take direct objects, with the possible exception of worth (as in It's not worth [the trouble]). And attributive adjectives, of course, modify nominals rather than complete noun phrases (Eager [customers] snapped up items as soon as they were displayed).
As pointed out earlier, when adjectives are used predicatively, the annotation of verbs like be is the same as with predicate nominals: we tag them Cop(ula) on the part-of-speech specific layer, which is abbreviated Adj layer in the case of adjective targets.
There is a distinct subclass of adjectives like economic, medical, military, judicial that can never be used predicatively as is shown in the following examples.
(135) The White House announced a new economic policy. (136) *The policy that the White House announced is economic .
We call these adjectives relational modifiers; other names used for them include pertainyms or domain adjectives. Relational adjectives are comparable to the modifying noun in noun-noun compounds (e.g. medicine in medicine man). They do not modify the referent or the extension of a head noun, but rather its sense or intension, hence the name. Dictionaries typically define them with phrases like having to do with, relating to, pertaining to, characteristic of [some abstract or concrete entity]. These adjectives are not frame-bearing, at least not in a way that is concrete enough to allow for a clear definition of a scenario and of a set of frame elements. Although they are placed in the frames with which they are broadly associated-the adjective retributory, for instance, lives in the Revenge frame-we do not provide full annotation for them, just as we do not annotate the heads of noun-noun compounds from the point of view of target nouns that pre-modify the head noun (cf. section 3.4.2).
Adverbs also evoke frames, that is, they are frame-bearing. In many cases, adverbial evocation of a frame alternates with adjectival evocation. Compare the following pairs:
(137) Bill was wise to sell the piano.
(138) Bill wisely sold the piano.
(139) Dillon was happy to carry the load.
(140) Dillon happily carried the
load.
Adverbially evoked frames are typically not the pragmatically dominant frames in a clause or sentence. For instance, whereas sentence (137) as a whole intuitively is an assessment of Bill's mental properties, (138) is foremost a report of a selling act. Support for this view comes from applying the so-called lie-test: if one challenges (137) by exclaiming That's a lie! one is contesting a different claim than somebody responding in the same way to (138). In the former case the wisdom of the act is challenged, in the latter case the act itself is challenged.
So far, the FrameNet project has not made adverbs such as wisely in (138) or happily in (140) targets of annotation; they were, however, annotated as dependents of the, typically verbal, semantic heads they modify.16
FrameNet largely concentrates on adverbs expressing speaker attitudes such as candidness or directness (cf. (141)-(142)), epistemic and evidential adverbs such as probably in (143), presumably in (144), and reportedly in (145).
(141) Frankly, my dear, I don't give a damn. (142) Honestly, I find this very disturbing. (143) In this case, Bayesian methods quickly reassure us that the Pope is probably not an alien. (144) An individual presumably fabricated a message purporting to be from me. (145) There are many herbal weight-loss supplements on the market today, and they utilize different mechanisms to reportedly aid in dieting.
Note that even if the clause that a target adverb modifies is broken up in two pieces by the adverb, the pieces still receive identical grammatical function and phrase type labels. The grammatical function will always be Head and the phrase type will mostly be Sfin. There are, however, cases such as (145), where the adverb only has scope over the infinitival purpose clause; the phrase type will then be of a different type, e.g. VPto in the case of (145).
Most frames do not include prepositions among the target lexical units. However, in a considerable number of frames, prepositions occur as semantically inactive markers of frame elements with some of the lexical units, as illustrated in (146) and (147).
While in each case the choice of a particular marker preposition must have some historical motivation, there is no clear understanding that the preposition evokes an identifiable frame that could also be evoked by the preposition when it heads a simple clause.
However, prepositions are appropriate targets of annotation in frames that cover the vocabulary of space, time, and motion, as shown in the following examples.
(148) The audience shouted that [Cinderella] was in [the cupboard]. (149) Lay [the insert] on [the table] and trace around it. (150) Sue dropped [her handbag] on [the table]. (151) [The beer] in [the fridge] was no good.
We annotate sentences such as (148) and (149) relative to the prepositions in and on, respectively, which are targets in the Locative_relation frame. The phrase following the preposition denotes the Landmark of the relation, and it is assigned the grammatical function Obj(ect). The Figure of the spatial/temporal relation is tagged as an Ext(ernal) argument. Note that this so even when the prepositional phrase post-modifies another noun as in (151). The phrase types for Figure and Ground expressions are chosen as appropriate. Very often, both Figure and Landmark will be NPs.
Note that in sentences like (148), we tag the verb be as a Copula on the (part-of-speech specific) Prep layer. Similar to what is found with nominal targets, some sentences with prepositions as targets allow reversal of the pre- and post-copular elements for pragmatic effect (Inside the drawer was a smoke alarm, still unopened in its box).17 FN annotation does not capture these differences; we always assign the label Copula (Cop).
Prepositions also often can take modifying phrases that immediately precede them as in (152) and (153).
(152) Bob took out a brand new sleeve of balls, teed one up and put it [right] into the water. (153) New Oxford is [10 miles] before Gettysburg.
These modifiers are assigned the GF Dep(endent) and whatever phrase type is appropriate, for instance, adverb in (152) and NP in (153).
Some nouns-for example, natural kind and artifact nouns like tomato, hammer, or pants-do not evoke frames by themselves, or do so only marginally. These nouns mostly just occur as slot fillers in frames evoked by verbs, adjectives, or other nouns. Nevertheless, we tag a governing verb or preposition on the Noun-Layer as a governor for some of these slot filler nouns.
The motivation for doing this is as follows: For certain entities, we would like to know in which frames they appear as slot fillers. Conversely, we also want to know, for a particular FE, what its typical slot fillers are. For example, consider the Building frame. We might wonder what kinds of things are usually built or constructed. Starting by thinking about various kinds of buildings, we might want to know in what kinds of events they participate. We may intuit that they are often mentioned as objects of build, but are there other kinds of building, for instance, that apply to buildings or their sub-parts?
We could attempt to automatically derive this kind of information from our annotation relative to governors like build, construct, assemble, etc. However, for that to be useful, we would have to annotate many more sentences for each of the governors than is current practice.18 For lexicographic purposes, it is not necessary to document that in addition to houses, towers, and many other things, castles can be constructed or put up, and so we have not done it. Thus, currently, looking for sentences with nouns such as castle, tower etc. and treating them as targets is the the easiest way to address questions like the ones above. We can then record what syntactic governors (Gov) take phrases containing the artifact noun targets as arguments, and what kind of syntactic constituents contain the target. (For lack of a better term, we call the constituent containing the target slot filler noun X.) Examples of what we call Gov-X annotation are given below.
(154) He [built Gov] [two tall towers X]. (155) Design and [construct Gov] [the castle X].
Notice that we tag both the verb and the particle as GOV when a phrasal verb is the governor of a target noun, whether or not the verb and particle are contiguous.
(156) Leslie [put Gov] [up Gov] [the towers X]. (157) Leslie [put Gov] [the towers X] [up Gov].
In several frames containing artifact nouns, we also have defined some frame elements reflecting a kind of qualia structure of the artifacts (see [Pustejovsky, 1995]). For instance, we may record modifiers that denote the material from which an artifact is made (constitutive quale) or the purpose for which it is used (telic quale). Here are some examples from the Clothing frame, which contains nouns denoting items of clothing.
(158) Dot always [slept Gov] [in [her Wearer] [vest Garment] and knickers X] (159) I [put Gov] [on Gov] [[my Wearer] [smart Descriptor] suit X] for the last of the interviews.
In addition to Gov-X annotation and qualia-type annotation, we can also attach a semantic type to some or all of the lexical units in a frame. For instance, we could assign the semantic type `Dress' to all items of clothing. Similarly we could give nouns like wood, concrete, ivory in a *Materials frame a semantic type `Material'. This kind of information would then give users of the FrameNet database information about classes of words that meet certain selectional restrictions.
One special kind of noun that received Gov-X annotation are what we call transparent nouns. Nouns like top, pound, bunch can appear as the first noun in N1-of-N2 constructions in contexts where the governing verb semantically selects N2 rather than N1, the syntactic head. In these contexts, we call N1 transparent.
(160) Sue drank a cup of hot coffee. (161) He pinned a square of fabric on the back. (162) She went to see her idiot of a husband.
Semantically, the nouns that can be transparent fall into the following classes:
While these words were annotated in appropriate frames such as Aggregate in the case of group, bunch, annotators added Gov-X annotation for those sentences where the target noun appears as a transparent N1:
(163) Leslie [ate Gov] [a [sliver Piece] [of lemon Substance]X]. (164) Robin [drank Gov] [[a Count] [pint Unit] [of beer Stuff] X].
