Recovery Strategies for Parsing Extragrammatical Language 1 
Jaime G. Carbonell and Philip J. Hayes 
Computer Science Department 
Carnegie-Mellon University 
Pittsburgh, PA 15213 
Practical natural language interfaces must exhibit robust behaviour in the presence of 
extragrammatical user input. This paper classifies different types of grammatical deviations 
and related phenomena at the lexical, sentential and dialogue levels and presents recovery 
strategies tailored to specific phenomena in the classification. Such strategies constitute a 
tool chest of computationally tractable methods for coping with extragrammatieality in 
restricted domain natural language. Some of the strategies have been tested and proven 
viable in existing parsers. 
1. Introduction 
Any robust natural language interface must be capable 
of processing input utterances that deviate from its 
grammatical and semantic expectations. Many re- 
searchers have made this observation and have taken 
initial steps towards coverage of certain classes of 
extragrammatical constructions. Since robust parsers 
must deal primarily with input that does meet their 
expectations, the various efforts at coping with extra- 
grammaticality have been generally structured as ex- 
tensions to existing parsing methods. Probably the 
most popular approach has been to extend 
syntactically-oriented parsing techniques employing 
Augmented Transition Networks (ATNs) (Kwasny and 
Sondheimer 1981, Weischedel and Sondheimer 1984, 
Weischedel and Black 1980, Woods et al. 1976). Oth- 
er researchers have attempted to deal with ungrammat- 
ical input through network-based semantic grammar 
techniques (Hendrix 1977), through extensions to 
pattern matching parsing in which partial pattern 
matching is allowed (Hayes and Mouradian 1981), 
through conceptual case frame instantiation (Dejong 
1979, Schank, Lebowitz, and Birnbaum 1980), and 
through approaches involving multiple cooperating 
parsing strategies (Carbonell and Hayes 1984, Carbo- 
nell et al. 1983, Hayes and Carbonell 1981). 
1 This research was sponsored in part by the Air Force Office 
of Scientific Research under Contract AFOSR-82-0219 and in part 
by Digital Equipment Corporation as part of the XCALIBUR 
project. 
Given the background of existing work, this paper 
focuses on three major objectives: 
1. to create a taxonomy of possible grammatical devi- 
ations covering a broad range of extragrammaticali- 
ties, including some lexical and discourse phenom- 
ena (for example, novel words and dialogue level 
ellipsis) that can be handled by the same mecha- 
nisms that detect and process true grammatical 
errors; 
2. to outline strategies for processing many of these 
deviations - some of these strategies have been 
presented in our earlier work, some are similar to 
strategies proposed by other researchers, and some 
have never been analyzed before; 
3. to assess how easily these strategies can be em- 
ployed in conjunction with several of the existing 
approaches to parsing ungrammatical input, and to 
examine why mismatches arise. 
The overall result should be a synthesis of different 
parse-recovery strategies organized by the grammatical 
phenomena they address (or violate), an evaluation of 
how well the strategies integrate with existing ap- 
proaches to parsing extragrammatical input, and a set 
of characteristics desirable in any parsing process deal- 
ing with extragrammatieal input. We hope this will aid 
researchers designing robust natural language interfac- 
es in two ways: 
1. by providing a tool chest of computationally ef- 
fective approaches to cope with extragrammatical- 
ity; 
Copyright 1984 by the Association for Computational Linguistics. Permission to copy without fee all or part of this material is granted 
provided that the copies are not made for direct commercial advantage and the Journal reference and this copyright notice are included on 
the first page. To copy otherwise, or to republish, requires a fee and/or specific permission. 
0362-613X/83/030123-24503.00 
American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 123 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
2. by assisting in the selection of a basic parsing 
methodology in which to embed these recovery 
techniques. 
In assessing the degree of compatibility between 
recovery techniques and various approaches to parsing, 
we avoid the issue of whether a given recovery tech- 
nique can be used with a specific approach to parsing. 
The answer to such a question is almost always affirm- 
ative. Instead, we are concerned with how naturally 
the recovery strategies fit with the various parsing 
approaches. In particular, we consider the computa- 
tional tractability of the recovery strategies and how 
easily they can obtain the information they need to 
operate in the context of different parsing approaches. 
The need for robust parsing is greatest for interac- 
tive natural language interfaces that have to cope with 
language produced spontaneously by their users. Such 
interfaces typically operate in the context of a well- 
defined, but restricted, domain in which strong seman- 
tic constraints are available. In contrast, text process- 
ing often operates in domains that are semantically 
much more open-ended. However, the need to deal 
with extragrammaticality is much less pronounced in 
text processing, since texts are normally carefully pre- 
pared and edited, eliminating most grammatical errors 
and suppressing many dialogue phenomena that pro- 
duce fragmentary utterances. Consequently, we shall 
emphasize recovery techniques that exploit and depend 
on strong semantic constraints. In some cases, it is 
unclear whether the techniques we discuss will scale 
up properly to unrestricted text or discourse, but even 
where they may not, we anticipate that their use in the 
restricted situation will provide insights into the more 
general problem. 
Before proceeding with our discussion, the term 
extragrammaticality requires clarification. Extra- 
grammaticalities include patently ungrammatical con- 
structions, which may nevertheless be semantically 
comprehensible, as well as lexical difficulties (for ex- 
ample, misspellings), violations of semantic con- 
straints, utterances that may be grammatically accept- 
able but are beyond the syntactic coverage of the sys- 
tem, ellipsed fragments and other dialogue phenomena, 
and any other difficulties that may arise in parsing 
individual utterances. An extragrammaticality is thus 
defined with respect to the capabilities of a particular 
system, rather than with respect to an absolute exter- 
nal competence model of the ideal speaker. 
Extragrammaticality may arise at various levels: 
lexical, sentential, and dialogue. The following sec- 
tions examine each of these levels in turn, classifying 
the extragrammaticalities that can occur, and discuss- 
ing recovery strategies. At the end of each section, we 
consider how well the various recovery strategies 
would fit with or be supported by various approaches 
to parsing. A final section discusses some experimen- 
tal robust parsers that we have implemented. Our 
experience with these parsers forms the basis for many 
of the observations we offer throughout the paper. 
We also discuss some more recent work on integrating 
many of the recovery strategies considered earlier into 
a single robust multi-strategy parser for restricted do- 
main natural language interpretation. 
2. Lexical Level Extragrammaticalities 
One of the most frequent parsing problems is finding 
an unrecognizable word in the input stream. The fol- 
lowing subsections discuss the underlying reasons for 
the presence of unrecognizable words and develop 
applicable recovery strategies. 
2.1. The unknown word problem 
The word is a legitimate lexeme but is not in the 
system's dictionary. There are three reasons for this: 
• The word is outside the intended coverage of the 
interface (For example, there is no reason why a 
natural language interface to an electronic mail 
system should know words like "chair" or "sky", 
which cannot be defined in terms of concepts in its 
semantic domain). 
• The word refers to a legitimate domain concept or 
combination of domain concepts, but was not in- 
cluded in the dictionary. (For example, a word like 
"forward" \[a message\] can be defined as a com- 
mand verb, its action can be clearly specified, and 
the objects upon which it operates - an old mes- 
sage and a new recipient - are already well-formed 
domain concepts.) 
• The word is a proper name or a unique identifier, 
such as a catalogue part name/number, not hereto- 
fore encountered by the system, but recognizable 
by a combination of contextual expectations and 
morphological or orthographic features (for exam- 
ple, capitalization). 
In the first situation, there is no meaningful re- 
covery strategy other than focused interaction (Hayes 
1981) to inform the user of the precise difficulty. In 
the third, little action is required beyond recognizing 
the proper name and recording it appropriately for 
future reference. The second situation is more compli- 
cated; three basic recovery strategies are possible: 
1. Follow the KLAUS (Haas and Hendrix 1983) ap- 
proach, where the system temporarily wrests initia- 
tive from the user and plays a well designed 
"twenty questions" game, classifying the unknown 
term syntactically, and relating it semantically to 
existing concepts encoded in an inheritance hier- 
archy. This method has proven successful for verbs, 
nouns and adjectives, but only when they turn out 
to be instances of predefined general classes of 
objects and actions in the domain model. 
2. Apply the project and integrate method (Carbonell 
1979) to infer the meaning and syntactic category 
of the word from context. This method has proven 
124 American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 
Jaime G. Carbonel| and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
useful for nouns and adjectives whose meaning can 
be viewed as a recombination of features present 
elsewhere in the input. Unlike the KLAUS method, 
it operates in the background, placing no major 
run-time burden on the user. However, it remains 
highly experimental and may not prove practical 
without user confirmation. 
3. Interact with the user in a focused manner to pro- 
vide a paraphrase of the segment of input contain- 
ing the unknown word. If this paraphrase results in 
the desired action, it is stored and becomes the 
meaning of the new word in the immediate context 
in which it appeared. The LIFER system (Hendrix 
1977) had a rudimentary capacity for defining syn- 
onymous phrases. A more general method would 
generalize synonyms to classify the new word or 
phrase in different semantic contexts. 
2.2.2 Misspellings 
Misspellings arise when an otherwise recognizable 
lexeme has letters omitted, substituted, transposed, or 
spuriously inserted. Misspellings are the most common 
form of extragrammaticality encountered by natural 
language interfaces. Usually, a word is misspelt into 
an unrecognizable character string. But, occasionally a 
word is misspelt into another word in the dictionary 
that violates semantic or syntactic expectations. For 
instance: 
"Copy the flies from the accounts directory to 
my directory" 
Although "flies" may be a legitimate word in the do- 
main of a particular interface (for example, the files 
could consist of statistics on med-fly infestation in 
California), it is obvious to the human reader that 
there is a misspelling in the sentence above. 
There are well-known algorithms for matching a 
misspelt word against a set of possible corrections 
(Durham, Lamb, and Saxe 1983), and the simplest 
recovery strategy is to match unknown words against 
the set of all words in an interface's dictionary. How- 
ever, this obviously produces incorrect results when a 
word is misspelt into a word already in the dictionary, 
and can produce unnecessary ambiguities in other cas- 
es. 
Superior results are obtained by making the spelling 
correction sensitive to the parser's syntactic and se- 
mantic expectations. In the following example: 
Add two fixed haed dual prot disks to the order 
"haed" can be corrected to: "had", "head", "hand", 
"heed", and "hated". Syntactic expectations rule two 
of these out, and domain semantics rule out two oth- 
ers, leaving "fixed head disk" as the appropriate cor- 
rection. Computationally, there are two ways to or- 
ganize this. One can either match parser expectations 
against all possible corrections in the parser's current 
vocabulary, and rule out spurious corrections, or one 
can use the parse expectations to generate a set of 
possible words that can be recognized at the present 
point and use this as input to the spelling correction 
algorithm. The latter, when it can be done, is clearly 
the preferable choice on efficiency criteria. Generating 
all possible corrections with a 10,000 word dictionary, 
only to rule out all but one or two, is a 
computationally-intensive process, whereas exploiting 
fully-indexed parser expectations is far more con- 
strained and less likely to generate ambiguity. For the 
example above, "prot" has 16 possible corrections in a 
small on-line dictionary. However, domain semantics 
allow only one word in the same position as "prot", so 
correction is most effective if the list of possible words 
is generated first. 
2.3. Interaction of morphology and misspelling 
Troublesome side-effects of spelling correction can 
arise with parsers that have an initial morphological 
analysis phase to reduce words to their root form. For 
instance, a parser might just store the root form of 
'directory' and reduce 'directories' to 'directory' plus a 
plural marker as part of its initial morphological phase. 
This process is normally driven by first failing to rec- 
ognize the inflected form as a word that is present in 
the dictionary, and then applying standard morphologi- 
cal rules (for example, -ies =) +y) to derive a root 
from the inflected form. If any root thus derived is in 
the dictionary, the input word is assumed to be the 
appropriate inflected form. 
There are several ways in which this procedure can 
interact with spelling correction: 
1. The same test, viz. not finding the word in the 
dictionary, is used to trigger both morphological 
analysis and spelling correction, so there is a ques- 
tion of which to do first. 
2. The root of the word may be misspelt (e.g. dircto- 
ties), even though the inflexion is correct, so that 
after the inflexion is removed, there is still no 
matching dictionary entry. 
3. The inflexion itself may be misspelt (e.g. director- 
ise), so that the standard morphological transfor- 
mations do not apply. 
The first kind of interaction is not usually a major 
problem. On the assumption that inflexion is more 
common than misspelling, the most straightforward 
and probably best strategy is to try inflexion first on 
unknown words and then if that does not produce a 
word in the dictionary, try spelling correction. Match- 
ing only against contextually appropriate words should 
avoid cases in which a misspelling produces an inflect- 
ed form of a different word. 
If the root of an inflected word is misspelt, it will 
be necessary to spelling correct all of the (possibly 
several) uninflected forms, which might be inefficient. 
Again, contextual sensitivity can help. 
American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 125 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
The third kind of interaction is most troublesome. 
Most inflexions are too short for spelling correction to 
be effective - letter substitution or omission on two 
letter sequences is hard to identify. Moreover, inflex- 
ion processing does not normally use an explicit list of 
inflexions, but instead is organized as a discrimination 
net, containing the inflexions implicitly. One solution 
may be to have a list of all misspellings of inflected 
forms, but even utilizing hash coding schemes, search- 
ing this set would be inefficient. 
A simpler solution to the entire problem of interac- 
tion between spelling correction and morphological 
analysis is to eliminate the morphological analysis, and 
just store all inflected forms in the dictionary. This 
has the disadvantages of being unaesthetic and being 
unable to deal with novel inflexions, but neither of 
these are major problems for restricted domain natural 
language interfaces. There is also a second order 
problem in that more than one inflected form of the 
same word could be found as candidate corrections 
through spelling correction, but this can be overcome 
by explicitly grouping the various inflexions of a given 
root together in the lexicon. 
2.4. Incorrect segmentation 
Input typed to a natural language interface is segment- 
ed into words by spaces and punctuation marks. Both 
kinds of segmenting markers, especially the second, 
can be omitted or inserted speciously. Incorrect seg- 
mentation at the lexical level results in two or more 
words being run together, as in "runtogether", or a 
single word being split up into two or more segments, 
as in "tog ether" or (inconveniently) "to get her", or 
combinations of these effects as in "runto geth er". 
In all these cases, it is possible to deal with such 
errors by extending the spelling correction mechanism 
to be able to recognize target words as initial segments 
of unknown words, and vice-versa. For instance, by 
spelling correcting "portdisks" against what is accepta- 
ble in the position it occupies in: 
Add two dual portdisks to the order 
it should be possible to recognize the initial segment 
"port" as the intended word, with "disks" as a left 
over segment to be inserted into the input string after 
the corrected word for further parsing, resulting in this 
case in the correct parse. Again, in: 
Add two dual port disks to the order 
an unrecognized (and uncorrectable) word "er" fol- 
lowing a word "ord" which has been recognized as an 
initial segment abbreviation should trigger an attempt 
to attach the unknown word to the end of the abbrevi- 
ation to see if it completes it. Correction of 
Add two du alport disks to the order 
would be somewhat harder. After failing in the above 
recovery methods, one letter at a time would be strip- 
ped off the beginning of the second unrecognizable 
word ("alport") and added at the end of the first un- 
recognizable word ("du"). This process succeeds only 
if at some step both words are recognizable and enable 
the parse to continue. Migrating the delimiter (the 
space) backwards as well as forwards should also be 
attempted between a pair of unknown words, stopping 
if both words become recognizable. Of course, the 
compounding of multiple lexical deviations (for exam- 
ple, misspellings, run-on words and split words in the 
same segment) requires combinatorially inefficient 
recovery strategies. Strong parser expectations amelio- 
rate this problem partially, but trade-offs must be 
made between resilience and efficiency for compound 
error recovery. 
2.5. Support for recovery strategies by various 
parsing approaches 
In general, lexical level recovery strategies operate in a 
sufficiently localized manner that the variations in 
global behaviour of different approaches to parsing do 
not come into play. However, most of the strategies 
are capable of using contextual restrictions on what 
incorrect lexical item might be, and therefore are most 
effective when the constraints on the unknown word 
are strongest. This suggests that they will be most 
successful when used with an approach to parsing in 
which it is easy to bring semantic constraints to bear. 
So, for instance, such techniques are likely to be more 
effective using a semantic grammar (Hendrix 1977, 
Brown and Burton 1975) or case frame instantiation 
(Dejong 1979, Hayes and Carbonell 1981) approach, 
than in an approach using a syntactic ATN (Woods, 
Kaplan and Nash-Webber 1972), where the expecta- 
tions are never more specific than membership in one 
or more general syntactic categories. 
3. Sentential Level Extragrammaticalities 
Recovery from extragrammaticality at the sentential 
level is much more dependent on the particular kind of 
parsing techniques that are employed. Some tech- 
niques lend themselves to straightforward recovery 
methods, while others make recovery difficult. An 
initial examination of the requirements for recovery 
from various kinds of sentential level ungrammaticality 
will allow us to draw some general conclusions about 
the most suitable basic parsing approaches to build on. 
We examine ungrammaticalities in five categories: 
missing words, spurious words or phrases, out of order 
constituents, agreement violations, and semantic con- 
straint violations. 
3.1. Missing constituents 
It is not uncommon for the user of a natural language 
interface to omit words from his input, either by mis- 
126 American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
take or in an attempt to be cryptic. The degree of 
recovery possible from such ungrammaticalities is, of 
course, dependent on which words were left out. For 
instance in: 
Add two fixed head dual ported disks to my 
order 
omitting "dual" would be unrecoverable since all disks 
are ported and the discriminating information about 
the number of ports would not be there. On the other 
hand, if "ported" is omitted, all vital information is 
still there (the only thing dual about disks is the num- 
ber of ports) and it should be possible to recover. 
Also the omission of function words like prepositions 
or determiners is usually (though not always) recover- 
able. In practice, most omissions are of words whose 
contribution to the sentence is redundant, and are 
done consciously in an attempt to be cryptic or 
"computer-like" (as in "Copy new files my 
directory"). This suggests that techniques that fill in 
the gaps on semantic grounds are more likely to be 
successful than strategies that do not facilitate the 
application of domain semantics. 
In general, coping with missing words requires a 
parsing process to determine the parse structure that 
would have been obtained if those words had been 
there. If the information provided by the missing 
words is not redundant (as in the case of "dual" 
above), then this structure will have gaps, but the 
structure will convey the broad sense of the user's 
intention, and the gaps can be filled in by inference or 
(more practically and safely) by interaction with the 
user, focusing on the precise gaps in the context of the 
global parse structure (see Section 4.2 for further dis- 
cussion of focused interaction techniques.) 
A parsing process postulates a missing word error 
when its expectations (syntactic or semantic) of what 
should go at a certain place in the input utterance are 
violated. To discover that the problem is in fact a 
missing word, and to find the parse structure corre- 
sponding to the user's intention, the parsing process 
must "step back" and examine the context of the 
parse as a whole. It needs to ignore temporarily the 
unfulfilled expectations and their contribution to the 
overall structure while it tries to fulfil some of its oth- 
er expectations through parsing other parts of the in- 
put and integrating them with already parsed constitu- 
ents. In terms of a left-to-right parse of the above 
example (minus "dual"), this would mean that when 
the parser encountered "ported", it should note that 
even though it was expecting the start of a modifier 
suitable for a computer component (assuming its ex- 
pectations are semantic), it had in fact found the latter 
part of a modifier for a disk and so could proceed as 
though the whole of the modifier was there. A parser 
with greater directional freedom might find "disk" 
first and then look more specifically for qualifiers suit- 
able for disks. Again, the existence of a complete disk 
qualifier in the user's intended utterance could be as- 
sumed from finding part of the qualifier in a place 
where a whole one should go. 
Another way of looking at this is as an attempt to 
delimit the gap in the input utterance, correlate it with 
a gap in the parse structure (filling in that gap if it is 
uniquely determined), and realign the parsing mecha- 
nism as though the gap did not exist. Such a realign- 
ment can be done top-down by hypothesizing the oth- 
er expected constituents from the parse structure al- 
ready obtained and attempting to find them in the 
input stream. Alternatively, realignment can be done 
bottom-up by recognizing as yet unparsed elements of 
the input, and either fitting them into an existing parse 
structure, or finding a larger structure to subsume both 
them and the existing structure. This latter approach 
is essential when the structuring words are missing or 
garbled. 
Whether a top-down or a bottom-up method is best 
in any given instance will depend on how much struc- 
ture the parser can recognize before having to deal 
with the missing word. If the parser is left-to-right 
and the gap appears early in the input, there is likely 
to be little structure already built up, so a bottom-up 
approach will probably produce better results. Similar- 
ly, if the missing word itself provides the highest level 
of structure (for example, "add" in the example 
above), a bottom-up approach is essential. On the 
other hand, if the missing word corresponds to a spot 
low-down in the parse structure, and the gap is late in 
the utterance, or the parser is not b~und to a strict 
left-to-right directionality, a top-down approach is 
likely to be much more efficient. In general, both 
methods should be available. 
3.2. Spurious constituents 
Words in an input utterance that are spurious to a 
parse can arise from a variety of sources: 
legitimate phrases that the parser cannot deal with: It 
is not uncommon for the user of a restricted do- 
main interface to say things that the interface can- 
not understand because of either conceptual or 
grammatical limitations. Sometimes, spurious ver- 
bosity or politeness is involved: 
Add if you would be so kind two fixed head 
and if possible dual ported disks to my order. 
Or the user may offer irrelevant (to the system) 
explanations or justifications, as observed in pre- 
paratory experiments for the GUS system (Bobrow 
et al. 1977), for example, 
I think I need more storage capacity, so add 
two fixed head dual ported disks to my or- 
der. 
American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 127 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
Some common phrases of politeness can be recog- 
nized explicitly, but in most cases, the only reason- 
able response is to ignore the unknown phrases, 
realign the parse on the recognizable input, and if a 
semantically and syntactically complete structure 
results, postulate that the ignored segment was in- 
deed redundant. In most such cases, the user 
should be informed that part of the input was ig- 
nored. 
• broken-off and restarted utterances: These occur 
when people start to say one thing, change their 
mind, and say another: 
Add I mean remove a disk from my order 
Utterances in this form are more likely to occur in 
spoken input, but a similar effect can arise in typed 
input when a user forgets to hit the erase line or 
erase character key: 
Add remove a disk from my order 
Add a single ported dual ported disk from 
my order 
Again the best tactic is to discard the broken-off 
fragment, but identifying and delineating the super- 
seded fragment requires strategies such as the one 
discussed below. 
• unknown words filling a known grammatical role: 
Sometimes the user will generate an incomprehensi- 
ble phrase synonymous with a constituent the sys- 
tem is perfectly capable of understanding: 
Add a dual ported rotating mass storage de- 
vice to my order 
Here the system might not know that "rotating 
mass storage device" is synonymous with "disk". 
This phenomenon will result in missing words as 
well as spurious words. If the system has a unique 
expectation for what should go in the gap, it should 
(with appropriate confirmation from the user) re- 
cord the unknown words as synonymous with what 
it expected. If the system has a limited set of ex- 
pectations for what might go in the gap, it could 
ask the user which one (if any) he meant and again 
record the synonym for future reference. In cases 
where there are no strong expectations, the system 
would ask for a paraphrase of the incomprehensible 
fragment. If this proved comprehensible, it would 
then postulate the synonymy relation, ask the user 
for confirmation, and again store the results for 
future reference. 
The kind of recovery strategies required here are 
surprisingly similar to those required for missing 
words. Essentially, the parser must recognize that the 
input contains recognizable segments as well as unex- 
pected and unrecognizable words and phrases inter- 
spersed among them. The way that a parser (at least a 
left-to-right parser) would encounter the problem is 
identical to the way that missing words are manifested, 
viz. the next word in sequence does not fulfil the 
parser's expectations. Overcoming this problem in- 
volves the same notion of "stepping back" and seeing 
how subsequent elements of the input fit with the 
parsing structure built up so far. A major difference is 
that the word that violated the expectations, and pos- 
sibly other subsequent words, may not be incorporated 
into the resulting structure. Moreover, in the case of 
purely spurious phrases, that structure may not have 
any gaps. For a parser with more directional freedom, 
the process of finding spurious phrases may be simpler 
in that it could parse all the words that fit into the 
structure before concluding that the unrecognizable 
words and phrases were indeed spurious. When gaps 
in the parse structure remain after parsing all the rec- 
ognizable input, the unrecognizable segment may not 
be spurious after all. It can be aligned with the gap in 
the parse and the possible synonymy relations dis- 
cussed above can be presented to the user for approv- 
al. 
In the case of broken-off utterances, there are 
some more specific methods that allow the spurious 
part of the input to be detected: 
• If a sequence of two constituents of identical syn- 
tactic and semantic type is found where only one is 
permissible, simply ignore the first constituent. 
Two main command verbs in sequence (for exam- 
ple, as in "Add remove ..." above), instantiate the 
identical sentential case header role in a case frame 
parser, enabling the former to be ignored. Similar- 
ly, two instantiations of the same prenominal case 
for the "disk" case frame would be recognized as 
mutually incompatible and the former again ig- 
nored. Other parsing strategies can be extended to 
recognize equivalent constituent repetition, but case 
frame instantiation seems uniquely well suited to it. 
• Recognize explicit corrective phrases and if the 
constituent to the right is of equivalent syntactic 
and semantic type as the constituent to the left, 
substitute the right constituent for the left constitu- 
ent and continue the parse. This strategy recovers 
from utterances such as "Add I mean remove ...", 
if "I mean" is recognized as a corrective phrase. 
• Select the minimal constituent for all substitutions. 
For instance in 
Add a high speed tape drive, that's disk 
drive, to the order 
one desires "disk drive" to substitute for "tape 
drive", not for the larger phrase "high speed tape 
drive", which also forms a legitimate constituent of 
like semantic and syntactic type. This preference is 
based solely on pragmatic grounds and empirical 
evidence. 
In addition to identifying and ignoring spurious 
input, a robust interface must tell the user what it has 
128 American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
ignored and should paraphrase the part of the input 
that it did recognize. The unrecognized input may 
express vital information, and if that information is not 
captured by the paraphrase, the user may wish to try 
again. Exceptions to this rule arise when the spurious 
input can be recognized explicitly as such. Expres- 
sions of politeness, for instance, might be treated this 
way. The ability to recognize such "noise" phrases 
makes them in some sense part of the expectations of 
the parser, and thus not truly spurious. However, 
isolating them in the same way as spurious input pro- 
vides the advantage that they can then be recognized 
at any point in the input without having to clutter the 
parser's normal processing with expectations about 
where they might occur. 
3.3. Out of order constituents and fragmentary 
input 
Sometimes, a user will use non-standard word order. 
There are a variety of reasons why users violate ex- 
pected constituent ordering relations, including unwill- 
ingness to change what has already been typed, espe- 
cially when extensive retyping would be required. 
Two fixed head dual ported disk drives add to 
the order 
or a belief that a computer will understand a clipped 
pseudo-military style more easily than standard usage: 
two disk drives fixed head dual ported to my 
order add 
Similar myths about what computers understand best 
can lead to a very fragmented and cryptic style in 
which all function words are eliminated: 
Add disk drive order 
instead of "add a disk drive to my order". 
These two phenomena, out of order constituents 
and fragmentary input, are grouped together because 
they are similar from the parsing point of view. The 
parser's problem in each case is to put together a 
group of recognizable sentence fragments without the 
normal syntactic glue of function words or position 
cues to indicate how the fragments should be com- 
bined. Since this syntactic information is not present, 
semantic considerations have to shoulder the burden 
alone. Hence, parsers that make it easy for semantic 
information to be brought to bear are at a considera- 
ble advantage. 
Both bottom-up and top-down recovery strategies 
are possible for detecting and recovering from missing 
and spurious constituents. In the bottom-up approach, 
all the fragments are recognized independently, and 
purely semantic constraints are used to assemble them 
into a single framework meaningful in terms of the 
domain of discourse. When the domain is restricted 
enough, the semantic constraints can be such that they 
always produce a unique result. This characteristic 
was exploited to good effect in the PLANES system 
(Waltz 1978) in which an input utterance was recog- 
nized as a sequence of fragments which were then 
assembled into a meaningful whole on the basis of 
semantic considerations alone. A top-down approach 
to fragment recognition requires that the top-level or 
organizing concept in the utterance ("add" in the 
above examples) be located first and the predictions 
obtainable from it about what else might appear in the 
utterance be used to guide and constrain the recogni- 
tion of the other fragments. 
As a final point, note that in the case of out of 
order constituents, a parser relying on a strict left-to- 
right scan will have much greater difficulty than one 
with more directional freedom. In out of order input, 
there may be no meaningful set of left-to-right expec- 
tations, even allowing for gaps or extra constituents, 
that will fit the input. For instance, a case frame 
parser that scans for the head of a case frame, and 
subsequently attempts to instantiate the individual 
cases from surrounding input, is far more amenable to 
this type of recovery than one dependent upon rigid 
word order constraints. 
3.4. Syntactic and semantic constraint 
violations 
Input to a natural language system can violate both 
syntactic and semantic constraints. The most common 
form of syntactic constraint violation is agreement 
failure between subject and verb or determiner and 
head noun: 
Do the order include a disk drives? 
Semantic constraint violations can occur because the 
user has conceptual problems: 
Add a floating head tape drive to the order 
or because he is imprecise in his language, using a 
related object in place of the object he really means. 
For instance, if he is trying to decide on the amount of 
memory to include in an order he might say 
Can you connect a video disk drive to the two 
megabytes. 
when what he really means is "... to the computer with 
two megabytes of memory". 
These different kinds of constraint violation require 
quite different kinds of treatment. In general, the 
syntactic agreement violations can be ignored; cases in 
which agreement or lack of it distinguishes between 
two otherwise valid readings of an input are rare. 
However, one problem that sometimes arises is know- 
ing whether a noun phrase is singular or plural when 
the determiner or quantifier disagrees with the head 
noun. It is typically best to let quantifiers dominate 
when they are used; for example, "two disk" really 
American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 129 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
means "two disks". And with determiner disagree- 
ment, it is often unimportant which reading is taken. 
In the example of disagreement above, it does not 
matter whether the user meant "a disk drive" or "any 
disk drives". The answer will be the same in either 
case, viz. a listing of all the disk drives that the order 
contains. In cases where the action of the system 
would be different depending on whether the noun 
phrase was singular or plural (e.g. "delete a disks from 
the order"), the system should interact with the user in 
a focused way to determine what he really meant. 
Semantic constraint violations due to a user's con- 
ceptual problems are harder to deal with. Once de- 
tected, the only solution is to inform the user of his 
misconception and let him take it from there. The 
actual detection of the problem, however, can cause 
some difficulty for a parser relying heavily on semantic 
constraints to guide its parse. The constraint violation 
might cause it to assume there was some other prob- 
lem such as out of order or spurious constituents, and 
look for (and perhaps even find) some alternative and 
unintended way of putting all the pieces together. 
This is one case where syntactic considerations should 
come to the fore. 
Semantic constraint violations based on the men- 
tion of a related object instead of the entity actually 
intended by the user will manifest themselves in the 
same way as the semantic constraint violations based 
on misconceptions, but their processing needs to be 
quite different. The violation can be resolved if the 
system can look at objects related to the one the user 
mentioned and find one that satisfies the constraints. 
In the example above, this means going from the mem- 
ory size to the machine that has that amount of memo- 
ry. Clearly, the distance of the relationship over 
which this kind of substitution is allowed needs to be 
controlled fairly carefully - in a restricted domain 
everything is eventually related to everything else. 
But there may well be rules that control the kind of 
substitutions that are allowed. In the above example, 
it suffices to allow a part to substitute for a whole 
(metonymy), especially if, as we assumed, it had been 
used earlier in the dialogue to distinguish between 
different instances of the whole. 
3.5. Support for recovery strategies by various 
parsing approaches 
We now turn the question of incorporating the senten- 
tial level recovery strategies we have been discussing 
into the various approaches to parsing mentioned in 
the introduction. As we shall see, there are considera- 
ble differences in the underlying suitability of the vari- 
ous approaches as bases for the recovery strategies. 
To address this issue, we classify parsing approaches 
into three general groups: transition network ap- 
proaches (including syntactic ATNs and network- 
based semantic grammars), pattern matching ap- 
proaches, and approaches based on case frame instan- 
tiation. 
3.5.1. Recovery strategies using a transition 
network approach 
Although attempts have been made to incorporate 
sentential level recovery strategies into network-based 
parsers including both syntactically-based ATNs 
(Kwasny and Sondheimer 1981, Weischedel and Son- 
dheimer 1984, Weischedel and Black 1980, Woods et 
al. 1976) and semantic grammar networks (Hendrix 
1977), the network paradigm itself is not well suited 
to the kinds of recovery strategies discussed in the 
preceding sections. These strategies generally require 
an interpretive ability to "step back" and take a broad 
view of the situation when a parser's expectations are 
violated, and this is very hard to do when using net- 
works. The underlying problem is that a significant 
amount of state information during the parse is implic- 
itly encoded by the position in the network; in the 
case of ATNs, other aspects of the state are contained 
in the settings of scattered registers. As demonstrated 
by the meta-rule approach to diagnosing parse failures 
described by Weischedel and Sondheimer (1983) else- 
where in this journal issue, these and other difficulties 
elaborated below do not preclude recovery from extra- 
grammatical input. However, they do make it difficult 
and often impractical, since much of the procedurally 
encoded state must be made declarative and explicit to 
the recovery strategies. 
Often an ATN parse will continue beyond the point 
where the grammatical deviation, say an omitted word, 
occurred, and reach a node in the network from which 
it can make no further progress (that is, no arcs can be 
traversed). At this point, the parser cannot ascertain 
the source of the error by examining its internal state 
even if the state is accessible - the parser may have 
popped from embedded subnets, or followed a totally 
spurious sequence of arcs before realizing it was get- 
ting in trouble. If these problems can be overcome 
and the source of the error determined precisely, a 
major problem remains: in order to recover, and parse 
input that does not accord with the grammar, while 
remaining true to the network formalism, the parser 
must modify the network dynamically and temporarily, 
using the modified network to proceed through the 
present difficulties. Needless to say, this is at best a 
very complex process, one whose computational tract- 
ability is open to question. It is perhaps not surprising 
that in one of the most effective recovery mechanisms 
developed for network-based parsing, the LIFER 
system's ellipsis handling routine (Hendrix 1977), the 
key step operates completely outside the network for- 
malism. 
As we have seen, semantic constraints are very 
important in recovering from many types of ungram- 
matical input, and these are by definition unavailable 
130 American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
in a purely syntactic ATN parser. However, semantic 
information can be brought to bear on network based 
parsing, either through the semantic grammar approach 
in which joint semantic and syntactic categories are 
used directly in the ATN, or by allowing the tests on 
ATN arcs to depend on semantic criteria (Bobrow 
1978, Bobrow and Webber 1980). In the former tech- 
nique, the appropriate semantic information for re- 
covery can be applied only if the correct network node 
can be located - a sometimes difficult task as we have 
seen. In the latter technique, sometimes known as 
cascaded ATNs (Woods 1980), the syntactic and se- 
mantic parts of the grammar are kept separate, thus 
giving the potential for a higher degree of interpretive- 
ness in using the semantic information. However, the 
natural way to use this technique is to employ the 
semantic information only to confirm or disconfirm 
parses arrived at on syntactic grounds. So the rigidity 
of the network formalism makes it very difficult to 
bring the available semantic information to bear effec- 
tively on extragrammatical input. 
A further disadvantage of the network approach for 
implementing flexible recovery strategies is that net- 
works naturally operate in a top-down left-to-right 
mode. As we have seen, a bottom-up capability is 
essential for many recovery strategies, and directional 
flexibility often enables easier and more efficient oper- 
ation of the strategies. Of course, the top-down left- 
to-right mode of operation is a characteristic of the 
network interpreter, not of the network formalism 
itself, and an attempt (Woods et al. 1976) has been 
made to operate an ATN in an "island" mode, that is, 
bottom-up, center-out. This experiment was done in 
the context of a speech parser where the low-level 
recognition of many of the input words was uncertain, 
though the input as a whole was assumed to be gram- 
matical. In that situation, there were clear advantages 
to starting with islands of relative lexical certainty, and 
working out from there. Problems, however, arise 
during leftward expansion from an island when it is 
necessary to run the network backwards. The admissi- 
bility of ATN transitions can depend on tests that ac- 
cess the values of registers which would have been set 
earlier when traversing the network forwards, but 
which cannot have been set when traversing back- 
wards. This leads at best to an increase in non- 
determinism, and at worse to blocking the traversal 
completely. 
3.5.2. Recovery strategies using a pattern 
matching approach 
A pattern matching approach to parsing provides a 
better framework to recover from some sentential- 
level deviations than a network-based approach. In 
particular, the definition of what constitutes a pattern 
match can be relaxed to allow for missing or spurious 
constituents. For missing constituents, patterns which 
match some, but not all, of their components can be 
counted temporarily as complete matches, and spurious 
constituents can be ignored so long as they are embed- 
ded in a pattern whose other components do match. 
In these cases, the patterns taken as a whole provide a 
basis on which to perform the kind of "stepping back" 
discussed above as being vital for flexible recovery. In 
addition, when pattern elements are defined semanti- 
cally instead of lexically, as with Wilks's (1975) ma- 
chine translation system, semantic constraints can 
easily be brought to bear on the recognition. Howev- 
er, dealing with out of order constituents is not so easy 
for a pattern-based approach since constituent order is 
built into a pattern in a rigid way, similarly to a net- 
work. It is possible to accept any permutation of ele- 
ments of a pattern as a match, but this provides so 
much flexibility that many spurious recognitions are 
likely to be obtained as well as the correct ones (see 
Hayes and Mouradian 1981). 
An underlying problem here is that there is no nat- 
ural way to make the distinctions about the relative 
importance or difference in role between one word 
and another. For instance, parsing many of the exam- 
ples we have used might involve use of a pattern like: 
(<determiner> <disk-drive-attribute>* <disk-drive>) 
which specifies a pattern of a determiner, followed by 
zero or more attributes of a disk drive, followed by a 
phrase synonymous with "disk drive". So this pattern 
would recognize phrases like "a dual ported disk" or 
"the disk drive". Using the method of dealing with 
missing constituents mentioned above, "the" would 
constitute just as good a partial match for this pattern 
as "disk drive", a clearly undesirable result. The 
problem is that there is no way to tell the flexible 
matcher which components of the pattern are discrimi- 
nating from the point of view of recognition and which 
are not. Another manifestation of the same problem is 
that different words and constituents may be easier or 
harder to recognize (for example, prepositions are 
easier to recognize than the noun phrases they intro- 
duce), and thus may be more or less worthwhile to 
look for in an attempt to recover from a grammatical 
deviation. 
The underlying problem then is the uniformity of 
the grammar representation and the method of apply- 
ing it to the input. Any uniformly represented gram- 
mar, whether based on patterns or networks, will have 
trouble representing and using the kinds of distinctions 
just outlined, and thus will be less well equipped to 
deal with many grammatical deviations in an efficient 
and discriminating manner. See Hayes and Carbonell 
(1981) for a fuller discussion of this point. 
3.5.3. Recovery strategies in a case frame 
paradigm 
Recursive case frame instantiation appears to provide 
American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 131 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
a better framework for recovery from missing words 
than approaches based on either network traversal or 
pattern matching. There are several reasons: 
• Case frame instantiation is inherently a highly inter- 
pretive process. Case frames provide a high-level 
set of syntactic and semantic expectations that can 
be applied to the input in a variety of ways. They 
also provide an overall framework that can be used 
to realize the notion of "stepping back" to obtain a 
broad view of a parser's expectations. As we have 
emphasised, this ability to "step back" is important 
when input deviates from the standard expecta- 
tions. 
• Case frame instantiation is a good vehicle for bring- 
ing semantic and pragmatic information to bear in 
order to help determine the appropriate parse in the 
absence of expected syntactic constituents. If a 
preposition is omitted (as commonly happens when 
dealing with cryptic input from hunt-and-peck typ- 
ists), the resulting sentence is syntactically anoma- 
lous. However, semantic case constraints can be 
sufficiently strong to attach each noun phrase to 
the correct structure. Consider, for instance, the 
following sentence typed to an electronic mail sys- 
tem natural language interface: 
"Send message John Smith" 
The missing determiner presents few problems, but 
the missing preposition can be more serious. Do we 
mean to send a message "to John Smith", "about 
John Smith", "with John Smith", "for John 
Smith", "from John Smith", "in John Smith", "of 
John Smith", etc.? The domain semantics of the 
case frame rule out the latter three possibilities and 
others like them as nonsensical. However, prag- 
matic knowledge is required to select "to John 
Smith" as the preferred reading (possibly subject to 
user confirmation) - the destination case of the 
verb is required for the command to be effective, 
whereas the other cases, if present, are optional. 
This knowledge of the underlying action must be 
brought to bear at parse time to disambiguate the 
cryptic command. In the XCALIBUR system case 
frame encoding (Carbonell, Boggs, Mauldin, and 
Anick 1983), we apply precisely such pragmatic 
knowledge represented as preference constraints 
(cf. Wilks 1975) on case fillers at parse time. 
Thus, problems created by the absence of expected 
case markers can be overcome by the application of 
domain knowledge. 
• The propagation of semantic knowledge through a 
case frame (via attached procedures such as those 
of KRL (Bobrow and Winograd 1977) or SRL 
(Wright and Fox 1983)) can fill in parser defaults 
and allow the internal completion of phrases such 
as "dual disks" to mean "dual ported disks". This 
process is also responsible for noticing when infor- 
mation is either missing or ambiguously determined, 
thereby initiating a focused clarificational dialogue 
(Hayes 1981). 
• The representation of case frames is inherently 
non-uniform. Case fillers, case markers, and case 
headers are all represented separately, and this dis- 
tinction can be used by the parser interpretively 
instantiating the case frame. For instance, if a case 
frame accounts for the non-spurious part of an 
input containing spurious constituents, a recovery 
strategy can skip over the unrecognizable words by 
scanning for case markers as opposed to case fillers 
which typically are much harder to find and parse. 
This ability to exploit non-uniformity goes a long 
way to overcoming the problems with uniform pars- 
ing methods outlined in the previous section on 
pattern matching. 
4. Dialogue Level Extragrammaticality 
The underlying causes of many extragrammaticalities 
detected at the sentential level are rooted in dialogue 
phenomena. For instance, ellipses and other fragmen- 
tary inputs are patently ungrammatical at the senten- 
tial level, but can be understood in the context of a 
dialogue. Viewed at this more global level, ellipsis is 
not an "ungrammaticality". Nevertheless, the same 
computational mechanisms required to recover from 
lexical and (especially) sentential problems are neces- 
sary to detect ellipsis and parse the fragments correct- 
ly for incorporation into a larger structure. In the 
same way, many dialogue phenomena are classified 
pragmatically as extragrammaticalities. 
In addition to addressing dialogue level extragram- 
maticalities, any robust parsing system must engage 
the user in dialogue for cooperative resolution of pars- 
ing problems too difficult for automatic recovery. In- 
teraction with the user is also necessary for a coopera- 
tive parser to confirm any assumptions it makes in 
interpreting extragrammatical input and to resolve any 
ambiguities it cannot overcome on its own. We have 
referred several times in our discussions to the princi- 
ple of focused interaction, and stated that practical 
recovery dialogues should be focused as tightly as 
possible on the specific problem at hand. Section 4.2 
discusses some considerations for structuring focused 
interaction dialogues - in particular, why they need to 
be so tightly focused, and what mechanisms are need- 
ed to achieve tight focusing in a natural manner. 
4.1. Ellipsis 
Ellipsis is a many-faceted phenomenon. Its manifesta- 
tions are varied and wide ranging, and recovery strate- 
gies for many types of ellipsis remain to be discovered. 
Nevertheless, it is also a very common phenomenon 
and must be addressed by any interface intended for 
serious use by real users. Empirical observations have 
shown that users of natural language interfaces employ 
132 American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
ellipsis and other abbreviating devices (for example, 
anaphora, short definite noun phrases, cryptic lan- 
guage omitting semantically superfluous words, and 
lexical abbreviations) with alarming frequency 
(Carbonell 1983). The results of our empirical obser- 
vations can be summarized as follows: 
Terseness principle: Users of natural language 
interfaces insist on being as terse as possible, 
independent of task, communication media, typ- 
ing ability, or instructions to the contrary, with- 
out sacrificing the flexibility of expression inher- 
ent in natural language communication. 
Broadly speaking, one can classify ellipsis into in- 
trasentential and intersentential ellipsis, with the latter 
category being far more prevalent in practical natural 
language interfaces. Intrasentential ellipsis occurs 
most frequently in coordinate clauses such as: 
John likes oranges and Mary apples. 
Often, this type of ellipsis is detectable only on se- 
mantic grounds (there is no meaningful noun-noun 
unit called "Mary apples"). The following sentence 
with identical syntax has a preferred reading that con- 
tains no ellipsis: 
John likes oranges and Macintosh apples. 
We know of no proven general strategies for interpret- 
ing this class of intrasentential ellipsis. An interesting, 
but untried, approach might be an application of the 
strategies described below with each coordinate clause 
in an intrasentential ellipsis being considered as a sep- 
arate utterance and with extensions to exploit the syn- 
tactic and semantic parallelism between corresponding 
constituents of coordinate clauses. 
There are several forms of intersentential ellipsis: 
• Elaboration - An ellipsed fragment by either speak- 
er can be an elaboration of a previous utterance. 
Either speaker can make the elaboration, but the 
second speaker usually does so, as in the following 
example: 
User: Give me a large capacity disk. 
System: With dual ports? 
User: Yes, and a universal frequency adap- 
ter. 
• Echo - A fragment of the first speaker's utterance is 
echoed by the second speaker. As described more 
fully in Hayes and Reddy (1983), this allows the 
second speaker to confirm his understanding of the 
first speaker's utterance without requiring an ex- 
plicit confirmation. 
User: Add a dual disk to the order. 
System: A dual ported disk. What storage ca- 
pacity? 
If, on the other hand, the system had explicitly 
asked "Do you mean a dual ported disk?", the user 
would have been conversationally obliged to reply. 
However, in either case, the user is free to correct 
any misapprehension the system displays. Some- 
times, as in the example in the next bullet below, 
an echo may also be an expression of bewilder- 
ment. In general, this form of ellipsis is far more 
prevalent in spoken interactions than in typed com- 
munication, but the need for a robust parsing sys- 
tem to confirm assumptions it is making without 
being too disruptive of the flow of conversation 
makes it very useful for natural language interfaces 
in general (see Section 4.2). 
• Correction - An ellipsed fragment substitutes for a 
portion of an earlier utterance that was in error. 
The correction occurs in three typical modes: 
• The first speaker can correct himself immediate- 
ly (much like the repeated segment problem dis- 
cussed in Section 3.2). 
• The second speaker can offer a correction 
(marked as such, or simply an ellipsed fragment 
in the interrogative). 
• Or, the first speaker can correct himself in re- 
sponse to a clarificational query from the second 
speaker. The form of the clarificational query 
can be a direct question, a statement of confu- 
sion, or echoing the troublesome fragment of 
the input, thereby combining two forms of ellip- 
sis as illustrated below. 
User: Give me a dual port tape drive. 
System: A dual port tape drive? 
User: Sorry, a dual port disk drive. 
• Reformulation - Part of an old utterance is reformu- 
lated and meant to be interpreted in place Of the 
corresponding old constituent. This is perhaps the 
most common form of ellipsis and the only one for 
which tractable computational strategies have been 
implemented. All the examples below are of this 
type. 
The LIFER/LADDER system (Hendrix 1977, Sacer- 
doti 1977) handled a restricted form of reformulation 
ellipsis. LIFER's ellipsis algorithm accepted a frag- 
mentary input if it matched a partial parse tree derived 
from the previous complete parse tree by (a) selecting 
a subtree that accounted for a contiguous segment of 
the previous input, and (b) possibly pruning back one 
or more of its branches. If a fragmentary input 
matched such a partial parse tree, it was assumed to be 
a reformulation ellipsis and the missing parts of the 
partial parse tree were filled out from the previous 
complete parse tree. In particular, if a single grammar 
category accounted for the entire fragment, and this 
category was present in the last query parsed by the 
system, the ellipsis algorithm substituted the fragment 
American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 133 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
directly for whatever filled the category in the last 
parse. An example of this is: 
User: What is the length of the Kennedy? 
System: 200 meters 
User: The fastest aircraft carrier? 
Since both "the Kennedy" and the "the fastest aircraft 
carrier" match the semantic category <ship>, the lat- 
ter phrase is allowed to substitute for the former. Note 
that a purely syntactic parse would not be sufficiently 
selective to make the proper substitution. "The fastest 
aircraft carrier" is a noun phrase, and there are three 
noun phrases in the original sentence: "the length", 
"the length of the Kennedy" and "the Kennedy". 
However, the rigid structure of semantic grammars 
proves insufficient to handle some common forms of 
reformulation ellipsis. The semantic grammar formal- 
ism is too restrictive for a simple substitution strategy 
to apply effectively if there is more than one fragment, 
if there is a bridging fragment (such as "the smallest 
with two ports" in the example below that bridges 
over "disk"), or if the fragment does not preserve 
linear ordering. In contrast, case frame substitution 
provides the freedom to handle such ellipsed frag- 
ments. 
The following examples are illustrative of the kind 
of sentence fragments the case frame method handles. 
We assume that each sentence fragment occurs imme- 
diately following the initial query below. Note also 
that we are using case frame here to refer to nominal 
as well as sentential case frames - the case frame be- 
ing instantiated in these examples is the one for a disk 
with cases such as storage capacity, number of ports, 
etc.. 
INITIAL QUERY: 
"What is the price of the three largest single 
port fixed media disks?" 
SUBSEQUENT QUERIES: 
"Speed?" 
"Two smallest?" 
"How about the price of the two smallest?" 
"Also the smallest with dual ports?" 
"Speed with two ports?" 
"Disk with two ports?" 
In these representative examples, punctuation is of no 
help, and pure syntax is of very limited utility. For 
instance, the last three phrases are syntactically similar 
(indeed, the last two are indistinguishable), but each 
requires that a different substitution be made on the 
preceding query. 
The DYPAR-II system (discussed in Section 5.2) 
handles ellipsis at the case frame level. Here we pre- 
sent the basic case frame ellipsis resolution method it 
employs. Its coverage appears to be a superset of the 
LIFER/LADDER system (Hendrix 1977, Sacerdoti 
1977) and the PLANES ellipsis module (Waltz and 
Goodman 1977). Although it handles most of the 
reformulation ellipsis we encountered, it is not meant 
to be a general linguistic solution to the ellipsis phe- 
nomenon. 
Consider the following example: 
>What is the size of the 3 largest single port fixed 
media disks? 
>disks with two ports? 
Note that it is impossible to resolve this kind of ellipsis 
in a general manner if the previous query is stored 
verbatim or as a semantic grammar parse tree. "Disks 
with two ports" would at best correspond to some 
<disk-descriptor> non-terminal, and hence, according 
to the LIFER algorithm, would replace the entire 
phrase "single port fixed media disks" that corre- 
sponded to <disk-descriptor> in the parse of the origi- 
nal query. However, an informal poll of potential 
users suggests that the preferred interpretation of the 
ellipsis retains the MEDIA specifier of the original 
query. The ellipsis resolution process, therefore, re- 
quires a finer grain substitution method than simply 
inserting the highest level non-terminals in the ellipsed 
input in place of the matching non-terminals in the 
parse tree of the previous utterance. 
Taking advantage of the fact that a case frame 
analysis of a sentence or object description captures 
the relevant semantic relations among its constituents 
in a canonical manner, a partially instantiated nominal 
case frame can be merged with the previous case 
frame as follows: 
• If a case is instantiated both in the original query 
and in the ellipsis, use the filler from the ellipsis. 
For instance "with two ports" overrides "single 
port" in our example, as both entail different val- 
ues of the same case descriptor, regardless of their 
different syntactic roles. ("Single port" in the 
original query is an adjectival construction, whereas 
"with two ports" is a post-nominal modifier in the 
ellipsed fragment.) 
• Retain any cases in the original parse that are not 
explicitly contradicted by new information in the 
ellipsed fragment. For instance, "fixed media" is 
retained as part of the disk description, as are all 
the sentential-level cases in the original query, such 
as the quantity specifier and the projection attri- 
bute of the query ("size"). 
If a case is specified in the ellipsed fragment, but 
not in the original query, use the filler from the 
ellipsis. For instance, the "fixed head" descriptor is 
added as the media case of the disk nominal case 
frame in resolving the ellipsed fragment in the fol- 
lowing example: 
>Which disks are configurable on a VAX 
11-7807 
134 American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
>Any configurable fixed head disks? 4.2. Focused interaction 
• In the event that a new case frame is mentioned in 
the ellipsed fragment, wholesale substitution occurs, 
much as in the semantic grammar approach. For 
instance, if after the last example one were to ask 
"How about tape drives?", the substitution would 
replace "fixed head disks" with "tape drives", rath- 
er than replacing only "disks" and producing the 
phrase "fixed head tape drives", which is semanti- 
cally anomalous. In these instances, the semantic 
relations captured in a case frame representation 
and not in a semantic grammar parse tree prove 
critical. 
The key advantage case frame instantiation pro- 
vides for ellipsis resolution is the ability to match cor- 
responding cases, rather than surface strings, syntactic 
structures, or non-canonical representations. Imple- 
menting an ellipsis resolution mechanism of equal 
power for a semantic grammar approach would, there- 
fore, be very difficult. The essential problem is that 
semantic grammars inextricably combine syntax with 
semantics in a manner that requires multiple represen- 
tations for the same semantic entity. For instance, the 
ordering of marked cases in the input does not reflect 
any difference in meaning, 2 while the surface ordering 
of unmarked cases does. With a semantic grammar, 
the parse trees produced by different marked case 
orderings can differ, so the knowledge that surface 
positioning of unmarked cases is meaningful, but posi- 
tioning of marked ones is not, must be contained with- 
in the ellipsis resolution process. This is a very unnat- 
ural repository for such basic information. Moreover, 
in order to attain the functionality described above for 
case frames, an ellipsis resolution based on semantic 
grammar parse trees would also have to keep track of 
semantically equivalent adjectival and post nominal 
forms (corresponding to different non-terminals and 
different relative positions in the parse trees). This is 
necessary to allow ellipsed structures such as "a disk 
with 1 port" to replace the "dual-port" part of the 
phrase "...dual-port fixed-media disk ..." in an earlier 
utterance. One way to achieve this effect would be to 
collect together specific nonterminals that can substi- 
tute for each other in certain contexts, in essence 
grouping non-canonical representations into context- 
sensitive semantic equivalence classes. However, this 
process would require hand-crafting large associative 
tables or similar data structures, a high price to pay for 
each domain-specific semantic grammar. In brief, the 
encoding of domain semantics and canonical structure 
for multiple surface manifestations makes case frame 
instantiation a much better basis for robust ellipsis 
resolution than semantic grammars. 
2 leaving aside the differential emphasis and other pragmatic 
considerations reflected in surface ordering 
In addition to dealing with ellipsis and other extra- 
grammatical phenomena that arise naturally for an 
interactive interface, a truly robust parsing system 
must initiate subdialogues of its own. Such dialogues 
are needed 
• when a robust parser makes assumptions that may 
not be justified and needs confirmation from the 
user that it has guessed correctly; 
• when a parser comes up against ambiguities that it 
cannot resolve on its own, either because of extra- 
grammaticality on the part of the user or because 
of some essential ambiguity in perfectly grammati- 
cal input; 
• when the more automated strategies may prove too 
costly or uncertain (e.g., when recovering from 
compound lexical errors); 
• or when the required information is simply not pres- 
ent. 
When an interactive system moves from the passive 
role of answering questions or awaiting individual user 
commands to a more active information-seeking role in 
clarificational dialogues, it must address the question 
of how to organize its communication so that it will 
behave in a way that fits with the conversational ex- 
pectations and conventions of its human user. Issues 
of when explicit replies are required, how to convey 
information in such a way as to require the minimal 
response from the user, how to keep the conversation 
within the domain of discourse of the system, etc., 
must all be addressed by a natural language interface 
capable of mixed-initiative dialogue. Examining all 
these topics here would take us too far afield from the 
issue of robust parsing, so we will confine ourselves to 
issues specific to the kind of recovery interaction de- 
scribed above. See Carbonell (1982) and Hayes and 
Reddy (1983) for a fuller discussion of the issues in- 
volved in organizing the dialogue of an interactive 
natural language system. 
We offer four guidelines for organizing recovery 
dialogues: 
• the interaction should be as focused as possible; 
• the required user response should be as terse as 
possible; 
• the interaction should be in terms of the system's 
domain of discourse rather than the linguistic con- 
cepts it uses internally; 
• there should be as few such interactions as possible. 
To see the need for focused interaction, consider 
the input: 
Add two fixed head ported disks to my order 
The problem is that the user has omitted "dual" be- 
tween "head" and "ported". Assuming that disks can 
only be single or dual ported, and using the sentential 
level recovery strategies described earlier, a parser 
American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 135 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
should be able to come up with an interpretation of 
the input that is two ways ambiguous. Interaction 
with the user is required to resolve this ambiguity, but 
the degree to which the system's initial question is 
focused on the problem can make a big difference in 
how easy it is for the user to respond, and how much 
work is required of the system to interpret the user 
response. An unfocused way of asking the question is: 
Do yon mean: 
Add two fixed head single ported disks to my or- 
der, or 
Add two fixed head dual ported disks to my order 
Here the user is forced to compare two very similar 
looking possibilities to ascertain the system's interpre- 
tation problem. Comparisons of this kind to isolate 
possible interpretation problems place an unnecessary 
cognitive load on the user. Furthermore, it is unclear 
how the user should reply. Other than saying "the 
second one", he has little option but to repeat the 
whole input. Since the system's query is not focused 
on the source of the ambiguity, it is conversationally 
awkward for the user to give the single word reply, 
"dual". This response is highly elliptical, but from the 
point of view of required information, it is complete. 
It also satisfies our second guideline that the required 
response be as terse as possible. 
A much better way of asking the user to resolve the 
ambiguity is: 
Do you mean 'single' or 'dual' ported disks? 
This question focuses precisely on the ambiguity, and 
therefore requires no effort from the user besides that 
of giving the information the system desires. Moreo- 
ver, it invites the highly desirable reply "dual". Since 
the system is focused on the precise ambiguity, it can 
also generate a discourse expectation for this and oth- 
er appropriate elliptical fragments in the user's re- 
sponse, and thereby recognize them with little difficul- 
ty. 
The ability to generate focused queries to resolve 
ambiguities places certain requirements on how a par- 
ser represents the ambiguous structure internally. Un- 
less the ambiguity is represented as locally as possible, 
it will be very hard to generate focused queries. If a 
parser finds the ambiguity in the above example by 
discovering it has two independent parse structures at 
the end of the parsing process, then generating a fo- 
cused query involves a computationally taxing intracta- 
ble comparison process. However, if the ambiguity is 
represented as locally as possible, for instance as two 
alternative fillers for a single instantiation of a disk 
frame nested within the "add to order" frame, then 
generating the focused query is easy - just output a 
paraphrase of the case frame (the one for disk) at the 
level immediately above the ambiguity with a disjunc- 
tion taking the place of the single filler of the ambigu- 
ous case (the portedness case). Moreover, such a 
representation forms an excellent basis for interpreting 
the natural elliptical reply. As Hayes and Carbonell 
(1981) show, parsers based on case frame instantia- 
tion are particularly well suited to generating ambigui- 
ty representations of this kind. 
Another tactic related to focused interaction that 
parsing systems can employ to smooth recovery dia- 
logues is to couch their questions in terms that make it 
more likely that the user's reply will be something they 
can understand. Thus in: 
Please add two 300 megabyte rotating mass storage 
devices to my order. 
if "rotating mass storage device" is not in the system's 
vocabulary, it is unwise for it to reply "what is a rotat- 
ing mass storage device?", since the terms the user 
chooses to clarify his input may be equally unintelligi- 
ble to the system. It is much better to give the user a 
choice between the things that the system could recog- 
nize in the place where the unrecognizable phrase 
occurred. In this example, this would mean giving the 
user a choice between all the computer components 
that can admit 300 megabytes as a possible case filler. 
If this list was unmanageably long, the system should 
at least confirm explicitly that the unknown phrase 
refers to a computer component by something like: 
By 'rotating mass storage device' are you referring 
to a computer component? 
This at least establishes whether the user is trying to 
do something that the system can help him with or 
whether the user has misconceptions about the abilities 
of the system. 
Upon confirmation that the user meant 'disk', the 
system could add the new phrase as a synonym for 
disk, perhaps after engaging the user in further clarifi- 
cational dialogue to ascertain that 'disk' is not merely 
one kind of 'rotating mass storage device', or vice 
versa. If it were the case that one was more general 
than the other, the new entry could be placed in a 
semantic hierarchy and used in future recognition 
(perhaps after determining on what key features the 
two differ). 
Our third guideline stated that the interaction 
should be in terms of the domain of discourse rather 
than the internal linguistic conventions of the system. 
Breaking this rule might involve requiring the user to, 
for instance, compare the parse trees representing two 
ambiguous interpretations of his input or telling him 
the name of the internal state where the parse failed in 
an ATN parser. Such interaction requires a linguisti- 
cally and computationally sophisticated user. Moreo- 
ver, it is highly non-focused from the user's point of 
view since it requires him to translate the parser's view 
of the problem into one that has meaning within the 
task domain, thereby switching contexts from perform- 
136 American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
ance of the task to linguistic issues. This enforced 
digression places an undue cognitive load on the user 
and should be avoided. 
The final guideline is to minimize the amount of 
corrective interactions that occur. It is very tedious 
for a user to be confronted with questions about what 
he meant after almost every input, or as Codd (1974) 
has suggested, to approve a paraphrase of each input 
before the system does anything. Clearly, there are 
situations when the user must be asked a direct ques- 
tion, to wit, when information is missing or in the 
presence of real ambiguity. However, a technique not 
requiring a reply is preferable when the system makes 
assumptions that are very likely to be correct, or when 
there are strong preferences for one alternative among 
several in ambiguity, anaphora, or ellipsis resolution. 
The echoing technique mentioned in Section 4.1 is 
very useful in keeping required user replies to a mini- 
mum while still allowing the user to overrule any un- 
warranted assumptions on the part of the system. The 
trick is for the system to incorporate any assumptions 
it makes into its next output, so the user can see what 
it has understood, correct it if it is wrong, and ignore 
it if it is correct: 
User: Add two dual ported rotating mass storage 
devices to my order 
System: What storage capacity should the two dual 
ported disks have? 
Here the system informs the user of its assumption 
about the meaning of "rotating mass storage device" 
(possible because only disks have dual ports) without 
asking him directly if he means "disk". 
This section has given a brief glimpse of some of 
the dialogue issues that arise in a robust parsing sys- 
tem. The overriding point here is that robust parsing 
techniques do not stop at the single sentence level. 
Instead, they must be integrated with dialogue tech- 
niques that allow for active user cooperation as a re- 
covery strategy of last resort. 
5. Experiments in Robust Parsing 
Having examined various kinds of extragrammaticality 
and the kinds of recovery strategies required to handle 
them, we turn finally to a series of experiments we 
have conducted or plan to conduct in robust parsing. 
Before describing some of the parsers involved in 
these experiments, we summarize some of the broad 
lessons that can be drawn from our earlier discussion. 
These observations have had a major role in guiding 
the design of our experimental systems. 
• The parsing process should be as interpretive as 
possible. We have seen several times the need for 
a parsing process to "stand back" and look at a 
broad picture of the set of expectations (or gram- 
mar) it is applying to the input when an ungram- 
maticality arises. The more interpretive a parser is, 
the better able it is to do this. A highly interpre- 
tive parser is also better able to apply its expecta- 
tions to the input in more than one way, which may 
be crucial if the standard way does not work in the 
face of an ungrammaticality. 
• The parsing process should make it easy to apply 
semantic information. As we have seen, semantic 
information is often very important in resolving 
ungrammaticality. 
1, The parsing process should be able to take advan- 
tage of non-uniformity in language like that identi- 
fied in Section 3.5.2. As we have seen, recovery 
can be much more efficient and reliable if a parser 
is able to make use of variations in ease of recogni- 
tion or discriminating power between different con- 
stituents. This kind of "opportunism" can be built 
into recovery strategies. 
I, The parsing process should be capable of operating 
top-down as well as bottom-up. We have seen 
examples where both of these modes are essential. 
Our earliest experiments in robust parsing were 
conducted through the FlexP parsing system (Hayes 
and Mouradian 1981). This system was based on 
partial pattern matching, and while it had the first and 
last of the characteristics listed above, it did not meas- 
ure up well to the other two. Indeed, many of our 
ideas on the importance of those characteristics were 
developed though observation of FlexP's shortcomings 
as described in 3.5.2, and more fully in Hayes and 
Carbonell (1981). With these lessons in mind, we 
constructed two additional experimental parsers: 
CASPAR to explore the utility of case frame instantia- 
tion in robust parsing, and DYPAR to explore the no- 
tion of combining several different parsing strategies 
in a single parser. Both experiments proved fruitful, 
as the next two sections show, and DYPAR has now 
been developed into a complete parsing system, the 
DYPAR-II parser, as part of the XCALIBUR expert 
system interface (Carbonell et al. 1983). After that, 
we describe an approach to parsing we are currently 
developing that we believe to be based on the best 
features of both systems. A final section discusses 
other methods and approaches that we consider prom- 
ising avenues for future research. 
5.1. The CASPAR parser 
As our earlier discussion on sentential-level ungram- 
maticality pointed out, case frame instantiation ap- 
pears to have many advantages as a framework for 
robust parsing. Our initial experiments in realizing 
these advantages were conducted through the CASPAR 
parser (Hayes and Carbonell 1981). CASPAR was 
restricted in coverage, but could deal with simple im- 
perative verb phrases (that is, imperative verbs fol- 
lowed by a sequence of noun phrases possibly marked 
by prepositions) in a very robust way. 
American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 137 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
Examples of grammatical input for CASPAR (drawn 
from an interface to a data base keeping track of 
course registration at a university) include: 
Cancel math 247 
Enrol Jim Campbell in English 324 
Transfer student 5518 from Economics 101 to 
Business Administration 111 
Such constructions are classic examples of case con- 
structions; the verb or command is the central con- 
cept, and the noun phrases or arguments are its cases. 
Considered as surface cases, the command arguments 
are either marked by prepositions, or unmarked and 
identified by position, such as the position of direct 
object in the examples above. 
The types of grammatical deviation that CASPAR 
could deal with include: 
• Unexpected and unrecognizable (to the system) 
interjections as in: 
÷S÷Q÷ S 3 Enrol if you don't mind student 
2476 I think in Economics 247. 
• missing case markers: 
Enrol Jim Campbell Economics 247. 
• out of order cases: 
In Economics 247 Jim Campbell enrol. 
• ambiguous cases: 
Transfer Jim Campbell Economics 247 English 
332. 
Combinations of these ungrammaticalities could also 
be dealt with. 
CASPAR used a parsing strategy specifically de- 
signed to exploit the recognition characteristics of 
imperative case frames, viz. that the prepositions used 
to mark cases are much easier to recognize than their 
corresponding case fillers. Below the clause level, 
CASPAR used linear pattern matching to recognize 
lower level constituents, which were defined in seman- 
tic terms appropriate to the restricted domain in which 
CASPAR was used. The algorithm used by CASPAR 
was as follows: 
1. Starting from the left of the input string, apply 
the linear pattern matcher in scanning mode 4 us- 
ing all the patterns which correspond to impera- 
tive verbs (commands). If this succeeds, the 
3 The reason for including these particular extraneous charac- 
ters will be easily guessed by users of certain computers. 
4 The linear pattern matcher may be operated in anchored 
mode, where it tries to match one of a number of linear patterns 
starting at a fixed word in the input, or in scanning mode, where it 
tries to match the patterns it is given at successive points in the 
input string until one of the patterns matches, or it reaches the end 
of the string. 
command corresponding to the pattern that 
matched becomes the current command, and the 
remainder of the input string is parsed relative to 
its domain-specific case frame. If it fails, 
CASPAR cannot parse the input. 
2. If the current command has an unmarked direct 
object case, apply the linear pattern matcher in 
anchored mode at the next 5 word using the set of 
patterns appropriate to the type of object that 
should fill the case. If this succeeds, record the 
filler thus obtained as the filler for the case. 
3. Starting from the next word, apply the pattern 
matcher in scanning mode using the patterns cor- 
responding to the surface markers of all the mark- 
ed cases that have not yet been filled. If this 
fails, terminate. 
4. If the last step succeeds, CASPAR selects a mark- 
ed case - the one from which the successful pat- 
tern came. Apply the matcher in anchored mode 
at the next word using the set of patterns appro- 
priate to the type of object that should fill the 
case selected. If this succeeds record the filler 
thus obtained as the filler for the case. 
5. Go to step 3. 
Unless the input turns out to be completely unparsa- 
ble, this algorithm will produce a command and a 
(possibly incomplete) set of arguments. It is also in- 
sensitive to spurious input immediately preceding a 
case marker. However, it is not able to deal with any 
of the other ungrammaticalities mentioned above. 
Dealing with them involves going back over any parts 
of the input that were skipped by the pattern matcher 
in scanning mode. If, after the above algorithm has 
terminated, there are any such skipped substrings, and 
there are also arguments to the command that have 
not been filled, the pattern matcher is applied in scan- 
ning mode to each of the skipped substrings using the 
patterns corresponding to the filler types of the un- 
filled arguments. This will pick up any arguments 
which were misplaced, or had garbled or missing case 
markers. 
This algorithm would deal with, for instance, the 
convoluted example: 
To Economics 247 Jim Campbell transfer please 
from Mathematics 121 
as follows: 
• The initial scan for a command verb would find 
"transfer", and thus cause all further parsing to be 
in terms of the case frame for that command. 
5 The word after the last one the pattern matcher matched 
the last time it was applied. If some input was skipped in finding 
the verb in step l, this is tacked onto the end of the sequence used 
by the next operation. 
138 American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
• The direct object required by "transfer" would not 
be found its expected place, after the verb, so 
CASPAR would skip to look for a case marker. 
• The case marker "from" would be found, and 
CASPAR would subsequently recognize the case 
marked by "from" and put it in the source course 
slot of the transfer case frame. 
• The end of the input is then reached, but some cas- 
es remain unfilled, so CASPAR goes into skipping 
mode looking for case markers on the missed initial 
segment and finds the destination course case. 
• Now only the 'Jim Campbell' and 'please' segments 
are left and the student case is left unfilled, so 
CASPAR can fill the student case correctly, and has 
'please' left over as spurious input. 
While limited in its scope of coverage, CASPAR 
provides a practical demonstration of how well case 
frame instantiation fulfills our list of desiderata for 
robust parsing. 
• CASPAR uses its case frames in a highly interpretive 
manner. It can, for instance, search directly after 
the verb for the filler of a case which is expected to 
be the direct object, but if that does not work, it is 
prepared to recognize the same case elsewhere in 
the input. Also, when it deals with out of order 
input, it "steps back" and takes a broad view by 
only considering unparsed input segments as poten- 
tial fillers of cases that have not yet been filled. 
• The case frame representation makes it easy to 
bring semantic information to bear, e.g. restrictions 
on what can fill each case, considerations of which 
cases are optional or mandatory, and whether any 
cases can have fillers that impose pragmatic const- 
raints. 
• CASPAR also shows the ability of case frame instan- 
tiation to exploit variations in importance and ease 
of recognition among different constituents. The 
power of exploiting such variations is shown both 
by the range of grammatical deviations CASPAR 
can handle, and by the efficiency it displays in 
straightforward parsing of grammatical input. This 
efficiency is derived from the limited number of 
patterns that the pattern matcher has to deal with 
at any one time. On its first application, the 
matcher only deals with command patterns; on sub- 
sequent applications, it alternates between the pat- 
terns for the markers of the unfilled cases of the 
current command, and the patterns for a specific 
object type. Also, except in post-processing of 
skipped input, only case marker and command pat- 
terns are employed when the pattern matcher is in 
its less efficient scanning mode. The constituents 
that are more difficult to recognize (e.g., object 
descriptions) are processed in the more efficient 
anchored mode. 
Only in its predominance of top-down versus 
bottom-up processing does CASPAR fail to meet 
our desiderata. The only bottom-up component to 
CASPAR is the initial verb recognition phrase. If 
the verb were not there, it would be completely 
unable to parse. An extension to CASPAR to ame- 
liorate this problem would be to start parsing case 
fillers bottom-up, and hypothesize the existence of 
the verb whose case frame most closely matched 
the set of case fillers found (or ask the user if there 
was no clear choice). This is obviously a much less 
efficient mode of operation than the one presented 
above, but it illustrates a way in which the basic 
case frame information could be interpreted to deal 
with the lack of a recognizable case header. 
5.2. The DYPAR parser 
DYPAR originated as an experimental vehicle to test 
the feasibility and potential benefits of combining 
multiple parsing strategies into a uniform framework. 
Initially, three parsing strategies (pattern matching, 
semantic grammar interpretation, and syntactic trans- 
formations) were combined. Transformations were 
used to reduce variant sentential structures to canoni- 
cal form. In addition to a large set of operators, 6 the 
patterns could contain recursive non-terminal sub- 
constituents corresponding to semantic grammar cate- 
gories or other subconstituents. Each grammar non- 
terminal could expand to a full pattern containing 
additional non-terminals. 
The experiment proved successful in that DYPAR 
allowed one to write grammars at least an order of 
magnitude more concise than pure semantic grammars 
of equivalent coverage. This version of the system is 
called DYPAR-I (Boggs, Carbonell, and Monarch 
1983) and has been made available for general use. 
Subsequently, case frame instantiation was introduced 
as the new dominant strategy, and the new system, 
DYPAR-II, is currently used as the experimental parser 
for XCALIBUR, a natural language interface to expert 
systems (Carbonell et al. 1983). 
The multi-strategy approach to parsing grammatical 
input in DYPAR-II facilitated the introduction of sever- 
al additional strategies to recover from different kinds 
of extragrammaticality: 
• Spelling correction combined with morphological 
decomposition of inflected words. 
• Bridging garbled or spurious phrases in otherwise 
comprehensible input. 
• Recognizing constituents when they occur in unex- 
pected order in the input. 
• Generalized case frame ellipsis resolution, exploit- 
ing strong domain semantics. 
6 Operators in DYPAR-I include: matching arbitrary subcon- 
stituent repetition, optional constituents, free permutation matches, 
register assignment and reference, forbidden constituents, and 
anchored and scanning modes. 
American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 139 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
The two sentential-level recovery strategies (second 
and third on the list above) were inspired by, and 
largely patterned after, the corresponding strategies in 
CASPAR, therefore little additional commentary is 
required. However, an additional complication in 
DYPAR-II is that the case frame instantiation process 
recognizes recursively embedded case frames, and in 
the presence of ill-formed input must deal with multi- 
ple levels of expectations. Were it not for strong do- 
main semantics, this additional source of complexity 
would have introduced some ambiguity in the correc- 
tion process requiring additional interaction with the 
user. 
5.2.1. Spelling correction and morphology in 
DYPAR 
DYPAR combines expectation-based spelling correction 
and morphological decomposition of inflected words. 
Since the DYPAR grammars are compiled into a cross- 
referenced form that indexes dictionary entries from 
patterns, it proved simple to generate lists of expected 
words when encountering an unrecognizable term. 
Although often the lists were short (highly constrained 
by expectations), on occasion a substantial fraction of 
the dictionary was generated. 
Since spelling correction interacts with morphologi- 
cal decomposition, the two were combined into a sin- 
gle recovery algorithm. Here we present a somewhat 
simplified form of the algorithm in which the only 
morphological operations allowed are on word endings 
(e.g., singularization and other suffix stripping opera- 
tions). 
1. Define the reduced dictionary to be the set of ex- 
pected words at the point in the parse where the 
unrecognized word was found. This set may con- 
tain expected or allowed morphological inflexions 
and variants, as well as root forms of words. 
2. Morphological decomposition phase - If the word 
(plus any accompanying morphological informa- 
tion) is a member of the reduced dictionary, return 
it and exit. 
3. Attempt to perform a one level morphological oper- 
ation on the current word (e.g., stripping a legal 
suffix) 
a. If successful, set the word to the decomposed 
form (e.g. root and suffix), save the potential 
decomposition on a list, and go to step 2. 
b. If no morphological operation is possible, go to 
the spelling correction phase (step 4). Only legal 
sequences of suffixes are allowed. 
4. Spelling correction phase - For each element in the 
list of possible decompositions (starting with the 
original unrecognized word), apply the spelling 
correction algorithm to the root word using the 
reduced dictionary as the candidate correction set. 
a. If successful, return the corrected word (along 
with any morphological information) and exit. 
b. If no spelling correction is possible, go on to the 
next proposed decomposition. 
5. If no proposed morphological decomposition yields 
a recognizable root, either by direct match or spell- 
ing correction, exit the algorithm with a failure 
condition. 
Clearly this strategy incorporates a best-match or 
minimal-correction criterion, rather than generating 
the set of all possible corrections. Moreover, words 
are only looked up in the reduced dictionary. This 
means that misspellings into words that are in the full 
dictionary but violate expectations (and are therefore 
not members of the reduced dictionary) are handled in 
the same manner as ordinary misspellings. 
Let us trace this correction strategy on the word 
"intrestingness'. Since that word is not recognized, we 
enter the algorithm above and generate a reduced dic- 
tionary. Assume that the reduced dictionary contains 
the word "interest", but none of its morphological 
variants. First we strip the "ness" suffix, but the re- 
suiting character string remains unrecognizable. Then 
we strip the "ing" suffix with similar results. Finally 
we strip off the coincidental "est" as a suffix and still 
find no recognizable root. At this point, morphology 
can do no more and the algorithm enters the spelling 
correction phase with the following candidate 
((root: (intrestingness) suffixes: 0) 
(root: (intresting) suffixes: (ness)) 
(root: (intrest) suffixes: (ing ness)) 
(root: (intr) suffixes: (est ing ness)) 
Next, we attempt to spelling correct "intrestingness" 
using the reduced dictionary and fail. We also fail with 
"intresting", but succeed with "intrest" and exit the 
algorithm with the value 
(root: (interest) suffixes: (ing ness)) 
and without considering the spurious "est" stripping. 
Had the word been correctly spelt, or had any of the 
compound morphological forms been inserted into the 
dictionary explicitly, the algorithm would have suc- 
ceeded and exited sooner. 
5.2.2. Ellipsis resolution 
DYPAR-II utilizes a variant of the case frame ellipsis 
resolution method discussed in Section 4.1. In addi- 
tion to the general algorithm, it incorporates a method 
for dealing with ellipsis when another component of 
the XCALIBUR system has generated strong discourse 
expectations. The ellipsed fragment is parsed in the 
context of these expectations, as illustrated by the 
recovery strategy below: 
Exemplary discourse expectation rule: 
IF: The system generated a query for confirmation 
or disconfirmation of a proposed value of a 
140 American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
filler of a case in a case frame in focus, 
THEN: EXPECT one or more of the following: 
1) A confirmation or disconfirmation pattern 
appropriate to the query in focus. 
2) A different but semantically permissible 
filler of the case frame in question 
(optionally naming the attribute or provid- 
ing the case marker). 
3) A comparative or evaluative pattern appro- 
priate to the proposed value of the case in 
focus. 
4) A query about possible fillers or constraints 
on possible fillers of the case in question. 
\[If this expectation is confirmed, a sub- 
dialogue is entered, where previously fo- 
cused entities remain in focus.\] 
The following dialogue fragment illustrates how 
these expectations come into play in a focused dia- 
logue: 
>Add a line printer with graphics capabilities. 
Is 150 lines per minute acceptable? 
>No, 320 is better Expectations 1, 2 & 3 
(or) other options for the speed? Expectation 4 
(or) Too slow, try 300 or faster Expectations 2 & 3 
The utterance "try 300 or faster" is syntactically a 
complete sentence, but semantically it is just as frag- 
mentary as the previous utterances. The strong dis- 
course expectations suggest that it be processed in the 
same manner as syntactically incomplete utterances, 
since it satisfies the dialogue expectations listed above. 
Thus, the terseness principle operates at all levels: 
syntactic, semantic and pragmatic. 
Additionally, DYPAR-II contains rules to ensure 
semantic completeness of utterances even in the ab- 
sence of specific discourse expectations. As we have 
just seen, not all sentence fragments are fragmentary 
in the syntactic sense. But not all such purely seman- 
tic ellipsis can be predicted through dialogue generated 
expectations. 
>Which fixed media disks are configurable on a 
VAX780? 
The RP07-aa, the RP07-ab, ... 
>Add the largest 
In this example, there is no good basis for predicting 
what the user will do in response to the information in 
the answer to his question. His response turns out to 
be semantically elliptical - we need to answer the 
question "largest what?" before proceeding. One can 
call this problem a special case of definite noun phrase 
resolution, rather than semantic ellipsis, but terminol- 
ogy is immaterial. Such phrases occur with regularity 
in our corpus of examples and must be resolved by a 
fairly general process. The following rule answers the 
question from context, regardless of the syntactic com- 
pleteness of the new utterance. 
Contextual substitution rule 
IF: A command or query case frame lacks one or 
more required case fillers, and the last case 
frame in focus has an instantiated case that 
meets all the semantic tests for the case miss- 
ing the filler, 
THEN" I) Copy the filler onto the new case frame, 
and 
2) Attempt to copy uninstantiated case fillers 
as well (if they meet semantic tests). 
3) Echo the action being performed for im- 
plicit confirmation by the user. 
For the example above, the case frame with a missing 
component is the selection case frame introduced by 
"largest" that requires a set of components from 
which to select. The previous (and therefore still fo- 
cused) input has a set of disks in its only case slot and 
this meets the semantic criteria for the selection slot; 
hence it is copied over and used. 
Rules such as the one above are fairly general in 
coverage, and the statement of the rule is independent 
of any specific case grammar or domain semantics. 
The rules, however, rely on the presence of the same 
specific case frames and the semantic constraints as 
used in the normal parsing of isolated grammatical 
constructions. 
5.3. Multi-strategy parsing 
In addition to underscoring the importance of our four 
desiderata for robust parsers listed at the beginning of 
this section, our experiments with CASPAR and 
DYPAR demonstrated that robustness can be achieved 
by the use of several different parsing strategies on the 
same input. These strategies operate both on gram- 
matical input and as a means of recovery from un- 
grammatical input. The notion of multiple strategies 
fits very well with the four desiderata. In particular: 
• The required high degree of interpretiveness can be 
obtained by having several different strategies ap- 
ply the same grammatical information to the input 
in several different ways. 
• Different strategies can be written to take advan- 
tage of different aspects of non-uniformity for dif- 
ferent construction types. 
• Some strategies can operate top-down and others 
bottom up. 
Nor, as we have seen in DYPAR, is a multiple strat- 
egy approach inconsistent with our previous emphasis 
on case frame instantiation as a suitable vehicle for 
robust parsing. Indeed, many of the strategies re- 
quired by a robust parser will be based on case frame 
instantiation with all the flexibility that that entails. 
However, case frame instantiation cannot carry the 
American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 141 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
entire burden of robustness alone, and so must be 
supplemented by other strategies such as the ones 
present in DYPAR. In fact, even the method of case 
frame instantiation presented for CASPAR can be seen 
as two strategies: one an initial pass using standard 
expectations, and the other a recovery strategy for 
when the first fails. The bottom-up strategy discussed 
at the end of the section on CASPAR would make a 
third. 
5.3.1. Coordinating multiple strategies through 
an entity-oriented approach 
A major problem that arises in using multiple parsing 
strategies is coordination between the strategies. 
Questions of interaction and order of application are 
involved. In CASPAR and DYPAR, the problem was 
solved simply by "hard-wiring" the interactions, but 
this is not satisfactory in general, especially if we wish 
to extend the set of strategies available in a smooth 
way. One alternative we have begun to explore in- 
volves the idea of entity-oriented parsing (Hayes 
1984). 
The central notion behind entity-oriented parsing is 
that the primary task of a natural language interface is 
to recognize entities - objects, actions, states, com- 
mands, etc. - from the domain of discourse of the 
interface. This recognition may be recursive in the 
sense that descriptions of entities may contain descrip- 
tions of subsidiary entities (for example, commands 
refer to objects). 
In entity-oriented parsing, all the entities that a 
particular interface system needs to recognize are de- 
fined separately. These definitions contain informa- 
tion both about the way the entities will be manifested 
in the natural language input (this information can also 
be used to generate output), and about the internal 
semantic structure of the entities. This arrangement 
has the following advantages for parsing robustness: 
• The individual entity definitions form an ideal 
framework around which to organize multiple pars- 
ing strategies. In particular, each definition can 
specify which strategies are applicable to recogniz- 
ing it. Of course, this only provides a framework 
for robust recognition, the robustness achieved will 
still depend on the quality of the various recogni- 
tion strategies used. 
• The individual definition of all recognizable domain 
entities allows them to be recognized independent- 
ly. Assuming there is appropriate indexing of enti- 
ties through lexical items that might appear in a 
surface description of them, this recognition can be 
done bottom-up, thus allowing for recognition of 
elliptical, fragmentary, or partially incomprehensi- 
ble input. The same definitions can also be used in 
a more efficient top-down manner when the input 
conforms to the system's expectations. 
• This style of organization is particularly well suited 
to case frame instantiation. The appropriate case 
frames can be associated with each entity definition 
for use by case-oriented strategies. Of course, this 
does not prevent other strategies from being used 
to recognize the entity, so long as suitable informa- 
tion for the other strategies to interpret is provided 
in the entity definition. 
These arguments can be made more concrete by exam- 
ple. 
5.3.2. Example entity definitions 
First we examine some example entity and language 
definitions suitable for use in entity-oriented parsing. 
The examples are drawn from the domain of an inter- 
face to a data base of college courses. Here is the 
(partial) definition of a course. Square brackets de- 
note attribute/value lists, and round brackets ordinary 
lists. 
\[ 
EntityName: CollegeCourse 
Type: Structured 
Components: ( 
\[ComponentName: CourseNumber 
Type: Integer 
GreaterThan: 99 
LessThan: 1000 
\] 
\[ComponentName: CourseDepartment 
Type: CollegeDepartment 
\] 
\[ComponentName: CourseClass 
Type: CollegeClass 
\] 
\[ComponentName: Courselnstructor 
Type: CollegeProfessor 
\] 
) 
SurfaceRepresentation: ( 
\[SyntaxType: Pattern 
Pattern: ($CourseDepartment $CourseNumber) 
\] 
\[SyntaxType: NounPhrase 
Head: (course I seminar I ...) 
AdjectivalComponents: (CourseDepartment ...) 
Adjectives: ( 
\[AdjectivalPhrase: (new I most recent) 
Component: CourseSemester 
Value: CurrentSemester 
\] 
) 
PostNominalCases: ( 
\[Case-marker: (?intended for I directed to I ...) 
Component: CourseClass 
142 American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 
Jaime G. Carbonel| and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
\[ 
) 
\] 
\] 
\[Case-marker: (?taught by I ...) 
Component: Courselnstructor 
\] 
) 
Precise details of this language are not relevant here. 
Important features to note include the definition of a 
course as a structured object with components: num- 
ber, department, instructor, etc.. This definition is 
separate from the surface representation of a course 
which is defined to take one of two forms: a simple 
word pattern of the course department followed by the 
course number (dollar signs refer back to the compo- 
nents), or a full noun phrase with adjectives, post- 
nominal cases, etc. Since we are assuming a multi- 
strategy approach to parsing, the two quite different 
kinds of surface language definition do not cause any 
problem - they can both be applied to the input inde- 
pendently by different construction-specific strategies, 
and the one which accounts for the input best will be 
used. 
Subsidiary objects like CollegeDepartment are de- 
fined in similar fashion. 
\[ 
EntityName: CollegeDepartment 
Type: Enumeration 
EnumeratedValues: ( 
ComputerScienceDepartment 
MathematicsDepartment 
HistoryDepartment 
) 
SurfaceRepresentation: ( 
\[SyntaxType: Pattern 
Pattern: (CS I Computer Science I CompSci I ...) 
Value: ComputerScienceDepartment 
\] 
) 
\] 
CollegeCourse itself will be a subsidiary entity in 
other higher-level entities of our restricted domain, 
such as a command to the data base system to enrol a 
student in a course. 
\[ 
EntityName: EnrolCommand 
Type: Structured 
Components: ( 
\[ComponentName: Enrollee 
Type: CollegeStudent 
\] 
\[ComponentName: EnrolIn 
Type: CollegeCourse 
\] 
) 
SurfaceRepresentation: ( 
\[SyntaxType: ImperativeCaseFrame 
Head: (enroll register I include I ...) 
DirectObject: ($Enrollee) 
Cases: ( 
\[Case-marker: (in I into I ...) 
Component: EnrolIn 
\[ 
) 
\] 
) 
\[ 
5.3.3. Parsing with an entity-oriented approach 
Now we turn to the question of how language defini- 
tions like those in the examples just given can be used 
to drive a parser. Let us examine first how a simple 
data base command like 
Enrol Susan Smith in CS 101 
might be parsed using the above language definitions. 
The first job is to recognize that we are parsing an 
EnrolCommand. In a purely top-down system, we 
would establish this by having a list of all the entities 
that we are prepared to recognize as complete inputs 
and trying each one of these to see if they could be 
recognized, a rather inefficient process. A more natu- 
ral strategy in an entity-oriented approach is to try to 
index bottom-up from words in the input to those 
entities that they might appear in. In this case, the 
best indexer for EnrolCommand is the first word, 
'enrol'. In general, the best indexer need not be the 
first word of the input and we need to consider all 
words, thus raising the potential of indexing more than 
one entity. Hence we might also index CollegeStu- 
dent, CollegeCourse, and CollegeDepartment. A sim- 
ple method of cutting down the number of index- 
generated possibilities to investigate top-down is to 
eliminate all those that are subsidiary to others that 
have been indexed. For our example, this would elim- 
inate everything except EnrolCommand, the desired 
result. One final point about indexing: it is clearly 
undesirable to index from every word that could ap- 
pear in the surface representation of an entity; only 
highly discriminating words like 'enrol' or 'CS' should 
be used. Whether a word is sufficiently discriminating 
can be determined either manually, which is unreliable, 
or automatically by keeping a count of the number of 
entities indexed by a given word and removing it from 
the index if it indexes more than a certain threshold 
number. 
American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 143 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
Once EnrolCommand has been established as the 
entity to recognize in the above example, the remain- 
der of the recognition can be accomplished straightfor- 
wardly in a top-down manner. The definition of the 
surface representation of EnrolCommand is an impera- 
tive case frame with a CollegeStudent as direct object 
and with a CollegeCourse as a second case indicated 
by 'in'. This information can be used directly by a 
simple case frame recognition strategy of the type used 
in CASPAR. No translation into a structurally differ- 
ent representation is necessary. The most natural way 
to represent the resulting parse would be: 
\[InstanceOf: EnrolCommand 
Enrollee: \[InstanceOf: CollegeStudent 
FirstNames: (Susan) 
Surname: Smith 
\] 
Enrolln: \[InstanceOf: CollegeCourse 
CourseDepartment: ComputerScienceDepartment 
CourseNumber: 101 
\] 
\] 
Note how this parse result is expressed in terms of the 
underlying structural representation used in the entity 
definitions without the need for a separate semantic 
interpretation step. 
To see the possibilities for robustness with the 
entity-oriented approach, consider the input: 
Place Susan Smith in computer science for fresh- 
men 
There are two problems here: we assume that the user 
intended 'place' as a synonym for 'enrol', but that it 
happens not to be in the system's vocabulary; the user 
has also shortened the grammatically acceptable 
phrase, 'the computer science course for freshmen', to 
an equivalent phrase not covered by the surface repre- 
sentation for CollegeCourse as defined above. Since 
'place' is not a synonym for 'enrol' in the language as 
presently defined, we cannot index EnrolCommand 
from it and hence cannot get the same kind of top- 
down recognition as before. So we are forced to rec- 
ognize smaller fragments bottom-up. Let's assume we 
have a complete listing of students and so can recog- 
nize 'Susan Smith' as a student. That leaves 'computer 
science for freshmen'. We can recognize 'computer 
science' as a CollegeDepartment and 'freshmen' as a 
CollegeClass, so since they are both components of 
CollegeCourse, we can attempt to unify our currently 
fragmentary recognition by trying to recognize a 
course description from the segment of the input that 
they span, viz. 'computer science for freshmen'. 
There are two possible surface representations giv- 
en for CollegeCourse. The first, a pattern, is partially 
matched by 'computer science', but does not unify the 
two fragments. The second, a noun phrase accounts 
for both of the fragments (one is adjectival, the other 
part of a post-nominal case), but would not normally 
match them because the head noun is missing. In 
fragment recognition mode, however, this kind of gap 
is acceptable, and the phrase can be accepted as a 
description of a CollegeCourse with ComputerScien- 
ceDepartment as CourseDepartment, and Freshman- 
Class as CourseClass. 
The input still consists of two fragments, however, 
a CollegeStudent and a CollegeCourse, and since we 
do not have any information about the word 'place', 
we are forced to consider all the entities that have 
those two sub-entities as components. We will sup- 
pose there are three: EnrolCommand, WithdrawCom- 
mand, and TransferCommand (with the obvious inter- 
pretations). Trying to recognize each of these, we can 
rule out TransferCommand in favour of the first two 
because it requires two courses and we only have one. 
Also, EnrolCommand is preferred to WithdrawCom- 
mand since the preposition 'in' indicates the Enrolln 
case of EnrolCommand, but does not indicate With- 
drawFrom, the course-containing case of Withdraw- 
Command. Thus we can conclude that the user in- 
tended an EnrolCommand. 
In following this bottom-up fragment combination 
procedure, we have ignored other combination possi- 
bilities that did not lead to the correct answer - for 
instance, taking 'Computer Science' as the StudentDe- 
partment case of the CollegeStudent, 'Susan Smith'. 
In practice, an algorithm for bottom-up fragment com- 
bination would have to consider all such possibilities. 
However, if, as in this case, the combination did not 
turn out to fit into a higher-level combination that 
accounted for all of the input, it could be discarded in 
favour of combinations that did lead to a complete 
parse. More than one complete parse would be han- 
dled, just like any other ambiguity, through focused 
interaction. 
Even assuming that the above example had a uni- 
que result, since it involved several significant assump- 
tions, we would need to use focused interaction tech- 
niques (Hayes 1981) to present a paraphrase of our 
interpretation to the user for approval before acting on 
it. Note that if the user does approve it, we should be 
able (perhaps with further approval) to add 'place' to 
the vocabulary as a synonym for 'enrol' since 'place' 
was an unrecognized word in the surface position 
where 'enrol' should have been. 
A pilot implementation of a parser constructed 
according to the entity-oriented principles outlined 
above has been completed and preliminary evaluation 
is promising. We are hoping to build a more complete 
parser along these lines. 
6. Concluding Remarks 
Any practical natural language interface must be capa- 
ble of dealing with a wide range of extragrammatical 
144 American Journal of Computational Linguistics, Volume 9, Numbers 3-4, July-December 1983 
Jaime G. Carbonell and Philip J. Hayes Recovery Strategies for Parsing Extrammatical Language 
input. This paper has proposed a taxonomy of the 
prevalent forms of extragrammaticality in real lan- 
guage use and presented recovery strategies for many 
of them. We also discussed how well various ap- 
proaches to parsing could support the recovery strate- 
gies, and concluded that case frame instantiation pro- 
vided the best framework among the commonly used 
parsing methodologies. 
At a more general level, we argued that the superi- 
ority of case frame instantiation over other parsing 
methodologies for robust parsing is due to how well it 
satisfies four parsing characteristics that are important 
for many of the recovery strategies that we described: 
• The parsing process should be as interpretive as 
possible. 
• The parsing process should make it easy to apply 
semantic information. 
• The parsing process should be able to take advan- 
tage of non-uniformity in language. 
• The parsing process should be capable of operating 
top-down as well as bottom-up. 
We claimed that while case frame instantiation satis- 
fies these desiderata better than any other commonly 
used parsing methodology, it was possible to do even 
better by using a multi-strategy approach in which 
case frame instantiation was just one member (albeit a 
very important one) of a whole array of parsing and 
recovery strategies. We described some experiments 
that led us to this view and outlined a parsing metho- 
dology, entity-oriented parsing, that we believe will 
support a multi-strategy approach. 
It is our hope that by pursuing lines of research 
leading to parsers that maximize the characteristics 
listed above, we can approach, in semantically limited 
domains, the extraordinary degree of robustness in 
language recognition exhibited by human beings, and 
gain some insights into how robustness might be 
achieved in more general language settings. 

References 
Bobrow, D.G. and Winograd, T. 1977 An Overview of KRL, a 
Knowledge Representation Language. Cognitive Science 1 (1): 
3-46. 
Bobrow, R.J. 1978 The RUS System. BBN Report 3878. Bolt 
Beranek and Newman, Cambridge, Massachuseets. 
Bobrow, R.J. and Webber, B. 1980 Knowledge Representation for 
Syntactic/Semantic Processing. Proc. National Conference of the 
American Association for Artificial Intelligence. Stanford Univer- 
sity, Stanford, California (August). 
Bobrow, D.G.; Kaplan, R.M.; Kay, M.; Norman D.A.; Thompson, 
H.; and Winograd, T. 1977 GUS: a Frame-Driven Dialogue 
System. Artificial lntelligence 8: 155-173. 
Boggs, W.M. and Carbonell, J.G. and Monarch, I. 1983 The 
DYPAR-I Tutorial and Reference Manual. Technical report. 
Computer Science Department, Carnegie-Mellon University, 
Pittsburgh, Pennsylvania. 
Brown, J.S. and Burton, R.R. 1975 Multiple Representations of 
Knowledge for Tutorial Reasoning. In Bobrow, D.G. and Col- 
lins, A., eds. Representation and Understanding. Academic Press, 
New York, New York: 311-349. 
Carbonell, J.G. 1979 Towards a Self-Extending Parser. Proceed- 
ings of the 17th Meeting of the Association for Computational 
Linguistics: 3-7. 
Carbonell, J.G. and Hayes, P.J. 1984 Robust Parsing Using Multi- 
ple Construction-Specific Strategies. In Bole, L., ed., Natural 
Language Parsing Systems. Springer-Verlag, New York, New 
York. 
Carbonell, J.G.; Boggs, W.M.; Mauldin, M.L.; and Anick, P.G. 
1983 The XCALIBUR Project, A Natural Language Interface 
to Expert Systems. Proceedings of the Eighth International Joint 
Conference on Artificial Intelligence. 
Carbonell, J.G.; Boggs, W.M.; Mauldin, M.L.; and Anick, P.G. 
1983 XCALIBUR Progress Report #l: First Steps Towards an 
Integrated Natural Language Interface. Technical report. 
Computer Science Department, Carnegie-Mellon University, 
Pittsburgh, Pennsylvania. 
Carbonell, J.G. 1982 Meta-Language Utterances in Purposive 
Discourse. Technical report. Computer Science Department, 
Carnegie-Mellon University, Pittsburgh, Pennsylvania. 
Carbonell, J.G. 1983 Discourse Pragmatics in Task-Oriented 
Natural Language Interfaces. Proceedings of the 21st Annual 
Meeting of the Association for Computational Linguistics. 
Codd, E.F. 1974 Seven Steps to RENDEZVOUS with the Casual 
User In Klimbie, J.W. and Koffeman, K.L., eds., Proceedings of 
the 1FIP TC-2 Working Conference on Data Base Management 
Systems. North Holland, Amsterdam: 179-200. 
Dejong, G. 1979 Skimming Stories in Real-Time. Ph.D. disserta- 
tion. Computer Science Department, Yale University, New 
Haven, Connecticut. 
Durham, i.; Lamb, D.D.; and Saxe, J.B. 1983 Spelling Correction 
in User Interfaces. Communications of the ACM 26. 
Haas, N. and Hendrix, G.G. 1983 Learning by Being Told: Ac- 
quiring Knowledge for Information Management. In Michalski, 
R.S.; Carbonell, J.G.; and Mitchell, T.M., eds., Machine Learn- 
ing, An Artificial Intelligence Approach. Tioga Press, Palo Alto, 
California. 
Hayes, P.J. 1981 A Construction Specific Approach to Focused 
Interaction in Flexible Parsing. Proceedings of 19th Annual 
Meeting of the Association for Computational Linguistics (June): 
149-152. 
Hayes, P.J. 1984 Entity-Oriented Parsing. COLING84, Stanford 
University, Stanford, California (July). 
Hayes, P.J. and Mouradian, G.V. 1981 Flexible Parsing. Ameri- 
can Journal of Computational Linguistics 7(4); 232-241. 
Hayes, P.J. and Carbonell, J.G. 1981 Multi-strategy Construction- 
Specific Parsing for Flexible Data Base Query and Update. 
Proceedings of the Seventh International Joint Conference on Arti- 
ficial Intelligence. Vancouver (August): 432-439. 
Hayes, P.J. and Reddy, D.R. 1983 Steps Toward Graceful Interac- 
tion. Spoken and Written Man-Machine Communication, Interna- 
tional Journal of Man-Machine Studies. 19(3): 211-284. 
Hayes, P.J. and Carbonell, J.G. 1981 Multi-Strategy Parsing and 
its Role in Robust Man-Machine Communication. Technical 
report CMU-CS-81-118. Computer Science Department, 
Carnegie-Mellon University, Pittsburgh, Pennsylvania (May). 
Hendrix, G.G. 1977 Human Engineering for Applied Natural 
Language Processing. Proceedings of the Fifth International Joint 
Conference on A rtificial Intelligence: 183-191. 
Kwasny, S.C. and Sondheimer, N.K. 1981 Relaxation Techniques 
for Parsing Grammatically Ill-Formed Input in Natural Lan- 
guage Understanding Systems. American Journal of Computa- 
tional Linguistics 7(2): 99-108. 
Sacerdoti, E.D. 1977 Language Access to Distributed Data with 
Error Recovery. Proceedings of the Fifth International Joint 
Conference on Artificial Intelligence: 196-202. 
Schank, R.C.; Lebowitz, M.; and Birnbaum, L. 1980 An Integrat- 
ed Understander. American Journal of Computational Linguistics 
6(1): 13-30. 
Waltz, D.L. 1978 An English Language Question Answering 
System for a Large Relational Data Base, Communications of the 
ACM 21(7): 526-539. 
Waltz, D.L. and Goodman, A.B. 1977 Writing a Natural Language 
Data Base System. Proceedings of the Fifth International Joint 
Conference on Artificial Intelligence: 144-150. 
Weischedel, R.M. and Sondheimer, N.K. 1983 Meta-Rules as a 
Basis for Processing Ill-formed Input. American Journal of 
Computational Linguistics 9(3-4): 161-177. 
Weischedel, R.M. and Black, J. 1980 Responding-to Potentially 
Unparseable Sentences. American Journal of Computational 
Linguistics 6: 97-109. 
Wilks, Y. A. 1975 Preference Semantics. In Keenan, ed., Formal 
Semantics of Natural Lanbuage. Cambridge University Press, 
Cambridge, England. 
Woods, W. A. 1980 Cascaded ATN Grammars. American Journal 
of Computational Linguistics 6: 1-12. 
Woods, W.A.; Kaplan, R.M.; and Nash-Webber, B. 1972 The 
Lunar Sciences Language System: Final Report. Technical 
report 2378. Bolt Beranek and Newman, Cambridge, Massa- 
chusetts. 
Woods, W.A.; Bates, M.; Brown, G.; Bruce, B.; Cook, C.; Klovst- 
ad, J.; Makhoul, J.; Nash-Webber, B.; Schwartz, R.; Wolf, J.; 
and Zue, V. 1976 Speech Understanding Systems - Final 
Technical Report. Technical report 3438. Bolt Beranek and 
Newman, Cambridge, Massachusetts. 
Wright, K. and Fox, M 1983 The SRL Users Manual. Technical 
report. Robotics Institute, Carnegie-Mellon University, Pitts- 
burgh, Pennsylvania. 
