Technical Correspondence 
Questioning the Need for Parsing 
Ill-formed Inputs 
Parsing of ill-formed input may be an important issue 
in the design of robust natural language systems, as 
pointed out by Eastman and McLean 1981. The un- 
predictable range of user input in natural language 
requires a mechanism to deal with unacceptable syn- 
tactic constructs. However, human behavior is compli- 
cated, and the capacity to adapt our mode of expres- 
sion to the varying contexts in which we communicate 
should not be overlooked. Because of our adaptabili- 
ty, particularly in patterns of speech, the ability to 
parse ill-formed inputs may not be a critical require- 
ment for all natural language systems. 
For example, we conducted an experiment that 
simulated a voice-driven office automation system, 
where subjects were provided with a tutorial and asked 
to solve simple problems using voice commands. 1 The 
purpose of the experiment was to test the usefulness 
of our proposed system and to gather data for the 
designers of the natural language processor we planned 
to implement (Biermann et al. 1983). 
The subjects in the experiment were seated in front 
of a terminal display with a microphone and were in- 
structed to speak their commands using discrete 
speech (about one word per second) or slow connect- 
ed speech. The monitor sat in a separate room, listed 
to their requests on earphones, and typed the corre- 
sponding commands to the system. In this way, sub- 
jects experienced a system that responded to spoken 
English sentences. If a command was not judged to 
be recognizable by our voice recognition system or 
parsable by our implemented grammar, the subject 
received an error message and was forced to rephrase 
the request. 
The surprising result of this experiment was that 
the subjects consistently spoke well-formed sentences 
when forced into the regimen of slow methodical 
speech and had very few inputs rejected on the basis 
of improper syntax. We had expected to find frequent 
occurrences of the three classes of ill-formed input 
described by McLean 1981, yet none of our subjects 
exhibited such errors to a significant degree. 
Out of more than 1600 sentences spoken by our 15 
subjects, only 10 utterances were rejected as ungram- 
matical inputs. Five of the ill-formed sentences con- 
tained incomplete noun phrases not processed by our 
system (e.g., "Combine the first paragraph with the 
second \[paragraph\]."), and the remainder involved 
skipped articles (e.g., "Remove period.", rather than 
"Remove the period."). We also found that relatively 
simple syntax was used, with a total of only six sen- 
tences containing conjunctions and with a notable 
absence of relative clauses. 
The subjects were required to begin each utterance 
with an imperative verb and to use words from a list 
of recognized vocabulary (in this case, some 50 words 
from the office domain). These constraints may have 
contributed to the lack of syntactic complexity we 
noted, but it is still remarkable that during an hour- 
long session, under the pressures of problem-solving, 
subjects routinely managed to form grammatically 
acceptable inputs. 
As a consequence of these observations, we have 
practically discontinued our efforts to parse ill-formed 
sentences. For pragmatic reasons, our work in that 
area has become a low priority in the overall develop- 
ment of a voice-driven natural language system. 
Linda Fineman 
Department of Computer Science 
Duke University 
Durham, NC 27706 

References 
Biermann, A., Rodman, R., Ballard, B., Betaneourt, T., Bilbro, G., 
Deas, H., Fineman, L., Fink, P., Gilbert, K., Gregory, D., and 
Heidlage, F. 1983 Interactive natural language problem solv- 
ing: a pragmatic approach. Paper presented at the Conference 
on Applied Natural Language Processing, Santa Monica, CA 
(February). 
Eastman, C.M. and McLean, D.S. 1981 On the need for parsing 
ill-formed input. AJCL 7 4 (October-December), Technical 
Correspondence. 
McLean, D.S. 1981 METASZK: a natural language front end to 
System 2000. M.S. thesis. Department of Mathematics and 
Computer Science, Florida State University, Tallahassee, FL 
(March). 
