<?xml version="1.0" standalone="yes"?>
<Paper uid="J80-1003">
  <Title>Slot Grammars</Title>
  <Section position="4" start_page="0" end_page="0" type="intro">
    <SectionTitle>
ASLOTS = (SUBJ AUXL ADVL IOBJ OBJ).
</SectionTitle>
    <Paragraph position="0"> This treatment of verb-dependencies is more direct than the use of transitivity features or encoding in transition network states, because this initial ASLOTS list expresses more directly what the verb &amp;quot;needs&amp;quot; to be the head of the VP. The semantic interpretation of the VP should be built (partially) from these slots and their fillers, and the syntax of the VP is guided by the filling of these particular slots. Furthermore, this method ties in nicely with the middle-out construction of the VP; search proceeds outward from the item that sets the goals.</Paragraph>
    <Paragraph position="1"> Not only does the slot grammar system initialize ASLOTS appropriately, but it also updates ASLOTS as parsing proceeds. At any point, ASLOTS provides a natural expression of what remains to be adjoined to the VP. Most parsers (e.g. ATN and APSG parsers) keep track of what slots have been filled, but it seems reasonable also to keep track of what slots may yet be filled, and use these in the control mechanism. Then rules that might be applied to fill a slot like OBJ never become activated if OBJ is not available.</Paragraph>
    <Paragraph position="2"> For instance, Heidorn (1972) has a rule roughly like the following: 32 American Journal of Computational Linguistics, Volume 6, Number 1, January-March 1980 Michael C. McCord Slot Grammars VP(TRANS,-~OBJ) NP --&gt; VP(OBJ=NP).</Paragraph>
    <Paragraph position="3"> This says that when a transitive VP with OBJ slot unfilled is followed by an NP, then a new VP is formed with OBJ filled by the NP. The rule will be tested every time a VP is formed, and this will be fruitless if the verb is not transitive (cannot take an OBJ) or if it already has an OBJ. Notice that OBJ is (implicitly) mentioned three times (counting the TRANS) in the rule, whereas one feels somehow that OBJ should be mentioned only once, since the rule is about filling the OBJ slot. Furthermore, if one had a slot that could be filled by more than one kind of filler (not just an NP) then this sort of rule would have to be duplicated for each type of filler.</Paragraph>
    <Paragraph position="4"> The appropriateness of basing search on an available-slots list seems especially clear in a language like Japanese with a rather free order of VP constituents. Suppose a grammar is to be written which captures the simple idea that the verb comes at the end of the VP, and the preceding NP's have case markings and can come in any order. In a slot grammar, the verb can activate a VP frame which has an ASLOTS list appropriate for that verb. Then the VP frame &amp;quot;looks to the left&amp;quot;, filling slots in AS-LOTS, and removing non-multiple slots from ASLOTS as it goes. In a situation that starts with, say, four slots and removes all but one, only this one slot will be relevant for further expectations in looking to the left, and rules will not be attempted needlessly. null Still another reason for basing expectations on ASLOTS has to do with the way raising constructions can be treated in bottom-up, middle-out analysis. Many languages allow unbounded raising of items, as in  (1) Which chair does Mary believe John said  he was sitting in? Here the question arises as to what syntactic role the initial NP which chair fills. Two VP levels and a PP down, there is a slot OBJ which is the object of the preposition in. Does which chair fill OBJ directly? If we try to write rules which accomplish this, we have to make them search down VP chains of arbitrary length and be aware of possible branching due to conjunctions, as in (2) Which chair does Mary believe that AI bought and John was sitting in? It seems that the rule for filling the object of the preposition should not have to &amp;quot;know about&amp;quot; these complications. The complications are created by VP complementation of verbs like believe and by conjunctions like and. The constructions that create the complications should take responsibility and should smooth the way for the placing of which chair.</Paragraph>
    <Paragraph position="5"> In slot grammars this is handled by the operation of raising slots. Every slot has a procedure attached to it called its slot-rule, which can test for the sorts of fillers the slot might have and can perform actions. RAISE is a possible action, and is illustrated as follows. Consider a sentence like  (3) Which chair does Mary believe that AI bought?  The VP frame for believe has a slot COMP (verbcomplement) which can be filled by another VP. To the right of believe is a VP that AI bought. This VP is &amp;quot;incomplete&amp;quot; in the sense that its ASLOTS register still contains a slot OBJ. In the slot-rule for COMP there is an instruction to RAISE all members of the filler's ASLOTS that belong to a specified list. (Some slots, such as verb auxiliaries, are not raised by COMP.) Raising a slot means creating a new member of the matrix VP's ASLOTS which is a sort of &amp;quot;image&amp;quot; of the lower slot. It has the same slot-rule and it is marked as being associated with the lower slot. A slot may be raised through several levels, but a path showing its origin is maintained for the purpose of semantic interpretation.</Paragraph>
    <Paragraph position="6"> In sentence (3) when the COMP slot for believe raises the lower OBJ to a new slot OBJ1, this is available to be filled by which chair at a certain stage when the top VP is looking to the left.</Paragraph>
    <Paragraph position="7"> The WH-movement that appears in sentences (1),(2), and (3) is a special kind of unbounded left movement (the left-dislocated item can be moved out of an unbounded number of embedded VP's). Another kind is topicalization, as in (4) This chair, she said you could put in the room.</Paragraph>
    <Paragraph position="8"> Raising applies to unbounded left movement in general, and in fact the same RAISE operation invoked by the VP COMP slot is used for handling both (3) and (4).</Paragraph>
    <Paragraph position="9"> In ATN grammars, unbounded left movement is handled by the HOLD facility (Woods, 1970, 1973).</Paragraph>
    <Paragraph position="10"> The ATN puts the left-dislocated item (like this chair in (4)) on a special stack by the HOLD action, and then at a later opportune time removes it from the stack while traversing a virtual arc --- in the case of (4), an arc parallel to the verb-object-NP arc --so that this chair becomes the object of put.</Paragraph>
    <Paragraph position="11"> The HOLD method does not mix well with bottom-up parsing, however, because it depends on using the complete left context at each point. (The item retrieved on a virtual arc could have been held anywhere from the beginning of the sentence.) Since bottom-up, middle-out analysis appears to be best for natural language (as this paper attempts to show), and since RAISE is a viable alternative to HOLD, we have an argument against HOLD.</Paragraph>
    <Paragraph position="12"> Furthermore, raising appears to be more generally applicable than HOLD. As hinted at in the discussion of (2) above, conjunction constructions should also involve raising. In that sentence, the American Journal of Computational Linguistics, Volume 6, Number 1, January-March 1980 33 Michael C. McCord Slot Grammars and frame should be responsible for creating the conjoined VP frame spanning that AI bought and John was sitting in, whose ASLOTS contains a slot OBJ1 which is related to both the object of bought and the object of id, by raising. This OBJ1 is further raised by the COMP slot of believe to a slot which is finally filled by which chair.</Paragraph>
    <Paragraph position="13"> The details for raising by conjunctions have not been completely worked out, but the general situation seems tO be roughly as follows. When a conjunction frame sees two frames of the same category on either side (the two conjuncts), it should construct raised slots corresponding to the intersection of the ASLOTS lists of the conjuncts. (In calculating the intersection, two slots that are already raised are considered equal if they originated from the same slot.) For example, in the sentence (5) John ate and slept.</Paragraph>
    <Paragraph position="14"> we could consider the ate frame to have ASLOTS = (SUBJ AUXL ADVL OBJ), but the slept frame would have ASLOTS = (SUBJ AUXL ADVL).</Paragraph>
    <Paragraph position="15"> The intersection would be (SUBJ AUXL ADVL), and these slots would be raised to slots (SUBJ1 AUXL1 ADVL1) in the conjoined VP ate and slept.</Paragraph>
    <Paragraph position="16"> Then John fills SUBJ1, to form the complete VP  (5). There is no object slot available in the conjoined VP. On the other hand, the conjoined VP cooked and ate would have both a subject and an object slot available, and we could get (6) John cooked and ate the pizza.</Paragraph>
    <Paragraph position="17">  In Woods (1973) conjunctions were handled by a system facility designed specially for conjunctions --meaning that the rules for conjunctions are not input by the grammar writer. The bottom-up, middle-out analysis with raising outlined above seems more straightforward and more controllable by the grammar writer. Consider a raising treatment possible for the following example discussed in Woods (1973): (7) John drove his car through and completely demolished a plate glass window.</Paragraph>
    <Paragraph position="18"> The and frame has on its left the VP drove his car through with ASLOTS = (SUBJ AUXL ADVL OBJ1), where OBJ1 is raised from the OBJ slot in the incomplete PP by ADVL. To the right is the VP completely demolished having ASLOTS = (SUBJ AUXL ADVL OBJ). The and frame creates the conjoined VP drove his car through and completely demolished, having raised ASLOTS = (SUBJ1 AUXL1 ADVL10BJ2) corresponding to the essentially identical ASLOTS lists of the two conjuncts. Then SUBJ1 is filled by John and OBJ2 is filled by a plate glass window, for the analysis of the complete sentence.</Paragraph>
    <Paragraph position="19">  3. States and slots  If all phrases had their heads at the beginning or end, and their other slots could be filled in any order, then all searching could be controlled by the unordered set ASLOTS. Many languages (including English) have an intricate combination of free placement of some slot-fillers with ordering restrictions on others. One conceivable method of controlling order would be to include tests in slot-rules for the position of the filler relative to other slot-fillers; but this seems to result in an unreasonable amount of testing, especially in languages in which there is a good deal of fixed order. It appears to be advisable to use some notion of &amp;quot;state&amp;quot; or &amp;quot;stage&amp;quot; in building phrases. In middle-out construction, another reason for using states is to control the direction in which the construction is proceeding; adjunctions might be made on the left, then the right, then switch directions again.</Paragraph>
    <Paragraph position="20"> In a slot grammar, each phrase frame has a register STATE, which contains an atom somewhat like an ATN state. Each state has a direction, LEFT or RIGHT, associated (permanently) with it, the idea being roughly that if a phrase is in state S, then it is looking for fillers in the direction associated with S. A restriction placed on states in slot grammars which makes their use much more constrained than in ATN's is that the set of states for a given phrase type (like VP) is linearly ordered. As a phrase gets built up, it can move ahead, but can never move back, in this ordering of states. Because of the linear order, the term stage might be more suggestive than state.</Paragraph>
    <Paragraph position="21"> In the grammar, slots are related to states in the following way. Each slot is specified to be attached to one or more states. To fill a given slot with a proposed filler, one must be able to advance (or not move back) from the current state of the matrix phrase (along the linear order of states) to a state to which the slot is attached, with the direction of the state corresponding to the direction of the proposed filler.</Paragraph>
    <Paragraph position="22"> The following diagram for a small VP grammar illustrates the use of states and slot attachment.</Paragraph>
  </Section>
class="xml-element"></Paper>