Considering the Effects of 
Second Language Learning on Generation 
Kathleen F. McCoy, Christopher A. Penpington 
Computer and Information Sciences Departlnent 
Applied Science and Engineering Laboratories 
University of Delaware / A.I.duPont Institute 
Newark, DE 19716 
mccoy@cis.udel.edu, penningt@ asel.udel.edu 
Abstract 
In this paper we discuss how generation issues 
affect the design of a computer-assisted language 
learning tool designed to teach written English as 
a second language to deaf users of American Sign 
Language. We discuss a dual-component linguis- 
tic model that attempts to reflect the generation 
process of the learners. The first model compo- 
nent captures the influence of the first language on 
the acquisition of the second. The second model 
component captures the process of second lan- 
guage acquisition itself. 
The linguistic model helps the system identify 
errors along with their probable source(s). This 
information is crucial for effective correction. It is 
also useful in the response phase of the system to 
focus tutoring on the errors that are most benefi- 
cial to correct. In addition, the linguistic model 
can be used to tailor the system's realization of its 
response. In this way, the syntactic constructions 
generated by the system will provide understand- 
able and positive exemplars of the language fea- 
tures currently being acquired by the leamer. 
Keywords 
tailoring response generation, user modeling, 
computer-assisted language learning 
1 Introduction 
Our long-term goal is to develop a computer- 
assisted language learning (CALL) tool to help 
deaf students leam written English. The targeted 
students are users of American Sign Language 
(ASL), a language that is very different from 
English in its structure and discourse strategies. 
The approach we take is to view the student's 
leaming of written English as a task in second 
Linda Z. Suri 
Central Institute for the Deaf 
818 S. Euclid Avenue 
St. Louis, MO 63110 
linda_suri@ cidmac.wustl.edu 
language acquisition. In this respect, our effort is 
similar to other projects geared toward learning 
(English as) a second language. 
We envision a system that would be used by a par- 
Ocular student over an extended period of time. A 
student would use the system as a tutor, entering 
texts (perhaps of several paragraphs in length) 
that he/she has written. The system would analyze 
these texts for errors, engage the student in a cor- 
rective tutorial dialogue, and offer possible cor- 
rected versions for some of the original input 
sentences. 
To accomplish these goals, the proposed system 
must have several components. First, it must have 
the ability to analyze texts that are input by the 
student and determine what/where errors occur. 
Once the errors have been identified, the system 
must decide which of these errors it should con- 
centrate on in its response to the student. Finally, 
the system must have the ability to generate 
appropriate corrective tutorial messages. 
While at first glance, one might think that prob- 
lems of Natural Language Generation only occur 
in the last phase of the system processing, we 
believe that a "generation" perspective on the 
entire process is extremely beneficial. For exam- 
ple, the problem of identifying errors in the origi- 
nal input text may seem like an issue of straight 
analysis, but it is not completely so. While an 
error might be recognized doing a syntactic parse 
of the sentence, in order for beneficial correction, 
both the specific error and its probable source 
should be identified. 
Consider the following simple example: "My 
brother like to go...". It is clear that there is a 
problem in subject-verb agreement; however, 
does it occur because (1) the noun should be in 
the plural form, (2) the verb should be in singular 
71 
form, or (3) the student doesn't know that such 
agreement exists in the language? Depending on 
the reason for the.mistake, different kinds of tuto- 
rial correction will likely be more helpful. Our 
belief is that in order to identify probable sources 
of errors, the developer must take into account the 
student's generation process. In other words, the 
eventual system must possess an understanding of 
what is causing the student to generate sentences 
that contain these mistakes. 
In this paper we present a portion of our work that 
describes the student's generation process as it is 
affected by second language acquisition (SLA). 
Our linguistic model of the student's generation 
process essentially reflects those aspects of the 
second language that are currently being leamed. 
This also has implications on the system's genera- 
tion process. In particular we discuss two ways in 
which the system's responses can be tailored to 
the user. In deciding "what to say", the system's 
generation can be tailored to focus on those errors 
that involve language features that the student is 
in the process of acquiring. In deciding "how to 
say it", the system can attempt to use the con- 
structions that are currently being leamed (as well 
as those that have been mastered) and so provide 
the student with correct exemplars of the second 
language. This is particularly important for the 
tutor we are developing since a lack of under- 
standable input / feedback is a serious problem for 
the deaf community. 
After giving an overview of our project, we con- 
centrate on how the student's generation process 
is represented in our system. We have developed a 
model of how the effects of the first language (in 
our case, ASL) can be accounted for in the analy- 
sis phase of our system, and are currently devel- 
oping a model which captures the effects of 
language acquisition itself. We discuss how these 
models affect the system's decisions of both what 
to say (i.e., what errors to tutor the student about) 
and how to say it (i.e., what syntactic construc- 
tions to use in the realization of the system's mes- 
sage). 
2 A Writing Tool for Deaf 
Students: ICICLE 
The problem of low literacy skills among deaf 
people has been well-documented and affects ev- 
ery aspect of deaf students' education. Since data 
on writing skills is not well documented, we note 
that the reading comprehension level of deaf stu- 
dents is considerably lower than that of their hear- 
ing counterparts, "... with about half of the 
population of deaf 18-year-olds reading at or be- 
low a fourth grade level and only about 10% read- 
ing above the eighth grade level..." \[Str88\]. We 
have undertaken a project designed to act as a 
"writing tutor" for deaf ASL signers learning writ- 
ten English. The eventual system will analyze a 
text written by a student, identify errors in the text, 
and engage the student in a tutorial dialogue aimed 
at some subset of the errors identified. 
  r....._ I ERROR 
I MAL-RULES \[ 
/ Y 
MODEL MODEL \[ 
~ USERMODEL 
RESPONSE GENERATOR 
Figure 1 ICICLE Overall System Design 
Figure 1 contains a block diagram of the overall 
system under development. The system, called 
ICICLE (Interactive Computer Identification and 
Correction of Language Errors), is designed to be 
a general purpose language learning tutor, how- 
ever, we have focused on its application to deaf 
users of ASL acquiring written English, essen- 
tially as a second language. 
In the ICICLE system, the input/feedback cycle 
begins when the student enters a portion of text 
into the computer. The student's text is first pro- 
cessed by the Error Identification component 
which is responsible for tagging all errors found in 
a given input. The possible effects of the student's 
first language on generated sentences is represent- 
ed by the Language Model in Figure 1. Our meth- 
odology for developing this knowledge source is 
described in the next section. At present, Error 
Identification analyzes only one input sentence at 
a time. It first does a syntactic parse of the sentence 
using an English grammar augmented with error 
production rules called mal-rules \[S1e82\], 
\[WVJ78\]. The mal-rules expand the coverage of 
the grammar to include the errors that might be ex- 
pected from people acquiring written English as a 
72 
second language. Thus these mal-rules capture ex- 
pected language generation pattems from this pop- 
ulation. Support for mal-rules will be developed 
within the framework of a probabilistic context- 
free grammar mechanism \[Cha93\], \[A1195\]. 1 
After using the expanded grammar to produce one 
or more syntactic parses of the input, the system 
selects a single parse using a scoring mechanism 
\[MPS96\] that takes into account a model of the 
acquisition process (labeled Acquisition Model in 
Figure 1 and further described in Section 4). Once 
a single parse is chosen, the errors are identified 
based on annotations associated with the real- 
rules. If syntactic mal-rules were used in the 
parse, the sentence and any relevant annotations 
will be passed to the Response Generator. The 
Error Identification component may also contain 
semantic rules and discourse information that 
could add annotations for the Response Genera- 
tor, though these are beyond the scope of this 
paper. 
Now the Response Generator will take this infor- 
mation, along with data from the User Model 
(only a portion of which is described in this 
paper), and decide which errors to correct in 
detail and how each should be corrected. 2 This 
process includes determining which syntactic 
constructions should be preferred in the actual 
realization of the response. The Response Gener- 
ator must also select an appropriate instructional 
strategy from the Tutoring Module. 
Finally, the system's responses are presented to 
the student who then has an opportunity to enter 
corrections to the text and have it re-checked. 
During the system processing, information about 
the student's language usage over time (as well as 
user/system interactions) can be tracked and 
updated through the History Module. 
While the overall system and its design encom- 
pass many important generation questions, we 
will focus on issues involving the Language 
Model and the Acquisition Model. We argue that 
these two models can be used to "explain" the stu- 
dent's sentence generation capabilities and should 
affect the system's response generation as well. 
1. The current implementation of this phase does not yet 
include probabilistic information. 
2. While in theory, instruction in ASL would be useful, the 
generation of ASL is well beyond the scope of this 
work. 
3 Capturing the Effects of the 
First Language 
A major proposal of our work is that a model in- 
corporating possible effects of the first language 
should be included in the component that is re- 
sponsible for identifying errors in the production 
of a second language. Such a model should indi- 
cate situations where the first language can have 
either a positive or negative influence. This claim 
is made based on an analysis of writing samples 3 
collected from a number of schools and organiza- 
tions for the deaf, concentrating on proficient ASL 
signers. 4 See \[Sur93\] for a complete discussion of 
this analysis as well as a detailed taxonomy of 
common language errors that were found. 
During our error analysis, we were constantly 
searching for why the errors we found were occur- 
ring. Knowing the underlying reason for a mistake 
is crucial to the goal of providing effective tutor- 
ing. Our analysis and intuitions led us to the notion 
of language transfer to explain many of the errors 
we were finding. The term "language transfer" 
generally refers to the influence that knowledge of 
one language (LI) has on the production and/or 
comprehension of a second language (L2). Trans- 
fer can be positive (in the sense that it may speed 
the acquisition of the L2); however, it may also re- 
sult in deviations in L2 production in places where 
the L1 and the L2 differ. While the existence of 
language transfer has been a rather controversial 
subject over the years (see \[McL87\], \[GS83\], 
\[Sur91\]), much recent research has provided con- 
vincing evidence that LT indeed occurs (see 
\[McL87\], \[Gas84\], \[GS83\]). 
Given that transfer has been documented between 
spoken languages, it is reasonable to ask whether 
or not language transfer could occur between 
ASL (a visual-gestural language) and written 
English. At first glance, transfer may seem sur- 
3. 
4. 
Other researchers (e.g., \[PQ731, \[QWM761, \[RQP76\], 
\[QPS77\], \[KK78\], \[QP841) studied errors in deaf writ- 
ing. Our work differs in that we attribute many errors to 
language transfer (LT) between ASL and written 
English as is explained below. 
We would like to thank John Albertini of the National 
Technical Institute for the Deaf (NTID), Bob McDonald 
of Gallaudet University, Lore Rosenthal of the Pennsyl- 
vania School for the Deaf, George Schellum (formerly) 
of the Margaret S. Sterck School, and MJ Bienvenu of 
the Bicultural Center for helping us gather writing sam- 
ples. 
73 
prising since the components of ASL grammar 
and written English grammar are very different 
\[Sto60\], \[BP78\], \[Pad82\], \[HS83\], \[KB79\], 
\[BPB83\]. ASL grammar components include sign 
order, morphological modulations of signs, and 
non-manual behavior that occurs simultaneously 
with the manual signs \[BC80\], \[Lid80\], \[Pad81 \], 
\[KG83\], ling78\], \[Bak80\]. Written English gram- 
mar components include word order, morphologi- 
cal modulations of words, and punctuation, but 
nothing that clearly corresponds to the simulta- 
neous manual/non-manual behavior found in 
ASL. On the surface, the fact that ASL and writ- 
ten English occur in different modalities seems 
problematic as well. However, there is some evi- 
dence that ASL is processed similarly to spoken 
languages (e.g., \[Sac90\]). 
3.1 Characterizing Language 
Transfer 
Because of the differences in grammar and modal- 
ity between ASL and English, we have attempted 
to abstractly characterize how languages could 
differ in a way that is independent of the grammar 
components. By looking at language on a feature 
by feature basis, we have identified several lan- 
guage mis-matches that may lead to (negative) 
transfer. 
First, languages may differ in when they mark a 
particular feature. As a result the marking of that 
feature in the L2 may seem redundant in the first 
language. For example, in ASL it is usual to 
establish tense at the beginning of a discourse 
segment or time frame, and then not to mark it 
again until the time frame changes. Of course, in 
English, tense is marked (on the verb) in every 
finite clause; so, some tense markings in English 
may seem redundant to an ASL signer. Transfer 
of such a feature (i.e., when to mark tense) might 
explain omission errors (in this case, of tense 
markings) in the L2. In fact, sentences containing 
these types of errors were common in our sam- 
ples. Consider the following: 
• "We went to see Senator Biden's office... Then 
we go to see the Vietnam memorial...." 
This example is a particularly good illustration of 
a difficulty due to a question of when to mark 
tense, since the writer clearly knows how to form 
the past tense of "go" (because the appropriate 
past tense form appears in the sample). 
Second, languages may differ in how they mark a 
feature. That is, two languages may express a 
concept in radically different ways and thus the 
mapping between the languages may be unclear. 
To illustrate this, consider the realization of the 
verb "to be" in ASL and English. Of course, "to 
be" is a standard verb in English. In ASL, how- 
ever. "to be" is not lexicalized using a standard 
sign. Instead it is conveyed implicitly in a topic- 
comment structure .5 For example, according to an 
ASL informant, to say "The shirt is red", the 
signer would typically sign SHIRT and mark it as 
a topic by raising his/her eyebrows, tilting his/her 
head and maintaining fairly constant eye gaze on 
the addressee, and then sign RED, with a different 
head position, brow position and gaze. 
Because of this vastly different method of realiza- 
tion, we might expect and often do find problems 
in sentences involving a main verb of "to be". 
Difficulties include both dropping the verb and 
confusing "have" and "be" as main verbs. 6 These 
errors are exemplified by the following: 
• "Once the situation changes they _ different 
people." 
• "... some birth controls are side-effect." (Possi- 
ble Correction: "... have side-effects...") 
A third way languages may differ is in regard to 
requiring morphological changes or additional 
lexical items for strictly syntactic reasons. For 
example, English requires a subject-verb agree- 
ment marking ("+s") on most verbs in the present 
tense when the subject is third-person-singular. 
This morphological marking often conveys no 
extra information. The situation of subject-verb 
agreement is more complex in ASL. When sub- 
ject-verb agreement is marked in ASL, it involves 
5. In a topic-comment structure, the topic is signed first, 
and then the comment is signed, grammatical signals 
marking the topic and comment. The grammatical mark- 
ings for topichood involve raising the eyebrows, tilting 
the head, and maintaining fairly constant eye gaze on the 
addressee (unless directional gaze is needed for other 
grammatical purposes). The final sign of the topic is also 
held slightly longer than usual. When the comment is 
signed, the head position, brows and gaze change. "How 
they change depends on the type of comment that fol- 
lows (e.g. \[sic\] statement, 'yes-no' question, com- 
mand)." (p. 157) \[BC801 
6. Because both having and being are expressed through 
the same grammatical structure in ASL, language trans- 
fer could explain why some ASL singers sometimes 
con~se the use of the verbs "be" and "have" in English. 
74 
a much different marking than what English 
requires. Also, the ASL marking is (generally) not 
empty of informational content. This could 
explain omissions of the morphological marking 
("+s") on these verbs in the written English of 
proficient ASL signers. Consider the sentence: 
• "My brother like to go..." 
A final area of language transfer occurs when one 
language (say L1) has two or more words or 
phrases which co~espond to a single word/phrase 
in the other language, and vice versa. For exam- 
ple, ASL uses the same sign (i.e., lexical item) for 
"other" and "another". Thus, language transfer 
might explain why an ASL learner of written 
English may have difficulty leaming which word 
("other" or "another") is appropriate to use in 
English. 
3.2 Developing the Language Model 
To summarize, part of our model of the learner's 
generation process includes a Language Model 
which captures the influence of the first language 
on the production of the second language. Our 
work in this area has included an analysis of writ- 
ing samples from deaf writers who are proficient 
in ASL. The analysis supports the hypothesis that 
these people are using the natural and beneficial 
strategy of building on their ASL knowledge 
when acquiring English. Our findings also reveal 
that many of the error classes (perhaps as many as 
76% of those found in our initial sample analysis) 
could be attributed to language transfer from ASL 
to English (if language transfer is defined in the 
way that we sugges0. However, we do not claim 
that every instance of an error class that could be 
explained by language transfer must be. There are 
other factors at work as well. 
What we do propose is that a Language Model 
that accounts for possible effects of the LI on the 
L2 should be developed by comparing the two 
languages on a feature by feature basis. When any 
of the four mismatches we have described occur, 
mal-rules that encode the conflicting realizations 
should be included in the Language Model. Thus, 
the resulting model will contain annotated mal- 
rules that capture the errors we expect from sec- 
ond language learners. Presumably this model 
should capture the generation process of a learner 
who is using a model of their first language as a 
basis for generating in the second. 
4 Taking the L2 Acquisition 
Process into Account 
So far our description of the learner's generation 
process takes only the first language into account. 
If the system contained only this model, then it 
would predict that the student would make all mis- 
takes invited by language transfer at every possi- 
bility. In fact, the set of errors a given student 
makes may be influenced by the L 1, but that set 
changes over time as the L2 is acquired. A student 
stops making some errors and begins making oth- 
ers. At the same time the set of errors a student 
makes changes, so too does the set of construc- 
tions a student uses (appropriately). For example, 
generally a beginning student does not attempt to 
use a sentence that contains a complex sentential 
compLment or a relative clause, until after he/she 
has mastered the use of simple subject-verb-object 
sentences. In this section we introduce a compo- 
nent which attempts to capture aspects of the sec- 
ond language acquisition process that affect the 
text the student is generating. 
In acquiring English as a second language, there 
is considerable linguistic evidence that the acqui- 
sition order of English features is relatively con- 
sistent and fixed regardless of the first language 
ling89\], \[DB74\], \[BMK74\]. In fact, a stronger 
version of this statement (i.e., that the language 
acquisition order is always fixed), is one of the 
central tenets of universal grammar theory (see 
for example, \[Haw91\] and \[KH87\]). While our 
belief is that for specific individuals this ordering 
may be influenced by some factors (such as the 
instructional situation or significant transfer from 
L1), these basic findings should play an important 
role in a model of second language acquisition. 
A second area of research which may also shed 
some light on the acquisition process encom- 
passes research in language assessment and edu- 
cational grade expectations (e.g., \[Ber88\], 
\[Lee74\], \[Cry82\]). This body of research outlines 
sets of syntactic constructions (language features) 
that students are generally expected to master by a 
certain point in time. This work can be interpreted 
as specifying groups of features that should be 
acquired at roughly the same time. For example, 
one would expect that the group of features that 
differ between a first and second grade reading 
level should be acquired together (i.e., between 
first and second grade). 
75 
We have attempted to account for the preceding 
results in a language assessment model called 
SLALOM ("Steps of Language Acquisition in a 
Layered Organization Model"). The basic idea 
behind SLALOM is to divide the English lan- 
guage (the L2 in our case) into a set of feature 
hierarchies (e.g., morphology, types of noun 
phrases, types of relative clauses). Within any sin- 
gle hierarchy, the features are ordered according 
to their "difficulty" of acquisition, reflecting their 
relative linguistic complexity. The ordering 
within feature hierarchies has been the subject of 
investigation in work such as \[Ing89\], \[DB74\], 
and \[BMK74\]. 
SLALOM 
Complex 
A B C D 
Feature Hierarchy 
Figure 2 Language Complexity in SLALOM 7 
Figure 2 contains an illustration of a piece of 
SLALOM. We have depicted parts of four hierar- 
chies in the figure: morphological syntactic fea- 
tures, noun phrases, verb complements, and 
various relative clauses. Within each hierarchy, 
the intention is to capture an ordering on the fea- 
ture acquisition. So, for example, the model 
reflects the fact that the +ing progressive form of 
verbs is generally acquired before the +s plural 
form of nouns, which is generally acquired before 
the +s form of possessives, etc. 
Notice that there are also connections among the 
hierarchies. This is intended to capture sets of fea- 
tures which are acquired at approximately the 
same time. These connections may be derived 
from work in language assessment and grade 
expectations such as found in \[Ber88\], \[Lee74\], 
and \[Cry82\]. So, for example, the figure indicates 
that while the +s plural ending is being acquired, 
7. We intend this figure as an illustration only. In particular 
our current research is focusing on identifying the pre- 
cise hierarchies, orderings and syntactic features in the 
hierarchies, as well as relationships among the hierar- 
chies. 
so too are both proper and regular nouns, and one 
and two word sentences. During this time, we do 
not expect to see any relative clauses. 
We anticipate that SLALOM, when fully devel- 
oped, will initially outline the typical steps of sec- 
ond language acquisition. This model will then be 
tailored to the needs of individual students via a 
series of "filters", one for each user characteristic 
that might alter the initial generic model. For in- 
stance, it is possible that the specific features of 
the student's L 1 will affect the rate or order of ac- 
quisition of the L2. In particular, one would expect 
features shared in the L1 and L2 to be acquired 
more quickly than those which ale not (due to pos- 
itive language transfer). Another possible filter 
could reflect how much and what kind of formal 
instruction the student has had in written English. 
For example, if the student's educational program 
stressed subject-verb agreement, this feature could 
have already been learned, even though others 
"before" it in the original SLA model may remain 
problematic. 
In developing the language learning model and its 
filters, we plan to compare our initial model 
(derived from acquisition literature) with the writ- 
ing samples that we have already collected. 8 We 
also expect to seek input from English teachers of 
deaf students. Additionally, we hope to collect 
samples of teachers' corrections and compare 
them to the models that will have been hypothe- 
sized. 
Once the SLALOM model is complete for the 
population under study, presumably we will have 
a model of the order in which we expect our ASL 
users to acquire written English. Essentially, we 
will need to "place" a particular user in the model. 
With this placement we will have a model of (1) 
what features we expect the student has mastered 
and is using consistently -- these are features 
below the user's level in the model, (2) what fea- 
tures we expect the user to be using or attempting 
to use, but with limited success -- these are fea- 
tures at the user's level, and (3) what features we 
8. Note that we have collected writing samples with some 
user information for the authors of each sample. While 
our analysis so far has been restricted to proficient ASL 
signers, samples from other deaf writers might help us 
determine what the ASL "influence" filter (for example) 
might look like since it would apply to one group of 
samples but not to another. 
76 
do not expect to see used (correctly) -- these are 
features above the user's level. 
Essentially we cfin view the placement in SLA- 
LOM as highlighting language features (and cor- 
responding mal-rules) that we expect the user to 
be using at a given point in time. Thus, it gives us 
a glimpse of the user's generation process by 
zeroing in on the mal-rules we expect him/her to 
be using at this point in their acquisition of 
English. 
The initial placement of the student on SLALOM 
will most likely be based on an analysis of the 
first input sample. Once this initial determination 
is made, further input from the student, as well as 
feedback given during the correction and tutorial 
phases, could cause the system to update the 
user's profile in the model. It is important to note 
that although the default levels (i.e., cross-hierar- 
chical connections) for the process of second lan- 
guage acquisition will be somewhat predeffined, 
the model is flexible enough to allow and account 
for individual variations beyond those represented 
by the initial model and its filters. In other words, 
additional information about each student's lan- 
guage usage gathered over time should provide a 
better and more accurate reflection of the current 
set of language features they are acquiring. 
5 Response Strategies 
There are a large number of issues that must be 
dealt with in determining an appropriate response 
for a student. These include choosing which errors 
to respond to, selecting an overall type of re- 
sponse, and generating the actual English response 
itself. In this paper we focus on just two aspects of 
response generation that rely on our model of the 
learner's generation process. The first is determin- 
ing which errors to respond to in detail, and the 
second deals with the kind of English syntactic 
constructions to use in the realization of the re- 
sponse. 
Our decisions concerning both of these aspects 
stem from work in second language acquisition 
and educational research. This work indicates that 
as a learner is mastering a subject, there is a certain 
subset of the material that is currently "within their 
grasp". This has been called the Zone of Proximal 
Development (ZPD) by Vygotsky \[Vyg86\]. This 
general idea has been applied to assessment and 
writing instruction by \[Rue90\]. There is a similar 
principle outlined in \[Kra81\] with respect to sec- 
ond language acquisition. Intuitively the knowl- 
edge or concepts within the ZPD are "ready to be 
learned" by the learner. It is what he/she is current- 
ly in the process of acquiring. 
We see our model of the user (including his/her 
placement in SLALOM) as capturing the ZPD 
with respect to second language acquisition. This 
has several implications on the responses given by 
the system. 
5.1 Deciding the Errors to Focus On 
According to the above literature, instruction and 
corrective feedback dealing with aspects within 
the ZPD may be beneficial. On the other hand, in- 
struction or corrective feedback dealing with as- 
pects outside of the ZPD will likely have little 
effect and may even be harmful to the learning 
process (in the sense that the user may become 
bored or confused by information that they are un- 
able to comprehend or apply). Thus in our system 
we plan to concentrate on correcting errors that in- 
volve features that are at or slightly above the 
learner's placement in SLALOM. Thus, we will 
provide instruction on those aspects of the lan- 
guage that the user is ready to acquire. 
As an example of the way SLALOM can be used 
in correction, consider the apparent similarity in 
the following two mistakes taken from different 
sources. The first was typed by one of the authors 
in an early draft of this paper: "That is, learners do 
not seem to acquired exactly one construction at a 
time ..... " A similar mistake is found in one of our 
writing samples from a student who is learning 
English as a second language. The student's writ- 
ing is fairly good, but still contains many errors in- 
volving appropriate verb morphology: "It's very 
hard for me to tell you what I am think about XXX 
because..."9 
Notice that the errors look very similar to each oth- 
er. In the first sample the bare form of the verb 
should appear after the infinitive marker "to"; in- 
stead, the verb appears to be inflected with tense. 
In the second sample the +ing form of the verb 
should appear affter the helping verb "am", but the 
bare form is included instead. Thus, in some sense, 
both instances could occur because the writer does 
not understand what form of the verb is required in 
the given circumstances. At first glance it may 
9. We have replaced the name of the student's school with 
XXX to protect the identity of our sources. 
77 
seem approp~ate to include tutorial dialogue ex- 
pinning verb morphology in both cases. 
However, if the level of English acquisition is tak- 
en into account via a model like SLALOM, signif- 
icantly better correction can be given. For 
example, in the first case the user's placement in 
SLALOM would indicate that he/she has already 
mastered the appropriate verb morphology. Thus 
the error should be seen as a typo. It should be 
pointed out so that it can be corrected, but tutorial 
dialogue on appropriate verb morphology is cer- 
tainly not necessary and would be inappropriate. 
In contrast, the second writer would be placed at a 
very different level within SLALOM. In particu- 
lar, his/her placement would indicate that he/she is 
still in the process of acquiring verb morphology; 
mistakes of the kind given in this sentence are 
rather common for this writer. As a result, tutorial 
dialogue explaining appropriate verb morphology 
is quite appropriate and has a good chance of hav- 
ing a positive effect on the leamer's writing. 
5.2 Tailoring Responses to the 
Student's Language Ability 
One of the few areas of general agreement among 
most SLA researchers is that linguistic input at or 
near the user's current second language proficien- 
cy is beneficial for the acquisition/learning pro- 
cess \[Kra82\], \[Tar82\], \[Vyg86\], \[Hat831. This 
principle may be expressed in different ways, but 
the idea is essentially the same. 
For example, in Krashen's "monitor" model of 
SLA \[Kra81\], \[Kra82\] he describes the learning 
process as a series of language level attainments. 
The level of the user's current ability in the sec- 
ond language is designated i. According to 
Krashen, in order to facilitate quicker progress, 
the most helpful input will include features of the 
i+ l level. 
For Tarone and others, this phenomena is referred 
to as foreigner talk. It describes the almost uncon- 
scious manner in which native speakers of a lan- 
guage automatically simplify their speech to 
accommodate second language learners. Input 
simplification is also a key premise in pidginiza- 
tion/creolization accounts of acquisition \[Hat83\]. 
For the most part, these types of simplification 
have been shown to be very helpful to the learner. 
However, Chaudron points out that there are a 
couple of "simplification" strategies that may not 
always be beneficial \[Cha83\]. 1° 
Given this evidence, it is very important for the 
ICICLE system to have the ability to generate text 
at or near the user's current second language pro- 
ficiency. The syntactic complexity can no longer 
be left to chance or indirect constraints. Our 
intention is that the user's placement in SLALOM 
will help guide the kind of constructions that the 
system should use in its generated text. That is, 
the system should attempt to generate texts which 
use constructions that are at (or slightly above) 
the level he/she is placed in SLALOM. 
Our initial investigations lead us to believe that 
this may be done within FUE a functional unifica- 
tion-based text generation system \[Elh93\] through 
a dynamic ordering of altemative realizations 
(ALTs). l I In other words, at ALT branches in the 
grammar, the alternatives will be ordered depend- 
ing on the user's current placement within SLA- 
LOM. Those constructions that are at or slightly 
above the user's level will be preferred by order- 
ing them before other potential realizations. If the 
user model is updated and the user's placement on 
SLALOM changes, a new ordering of the alterna- 
tives would reflect this change. 
6 Conclusion 
In this paper we have introduced a section of an in- 
telligent computer-assisted language learning sys- 
tem that attempts to capture the user's current 
generation capabilities. This linguistic model is 
based on two components: a model of the user's 
first language (in our case, ASL) and a model cap- 
turing the user's progress in acquiring the second 
language. Having this dual model in the system 
helps us understand the user's input in terms of the 
errors that they make. In addition, the linguistic 
model can be used to provide corrective feedback 
that is most likely to be beneficial to the user. 
7 Acknowledgments 
This work is supported by NSF Grant # IRI- 
9416916 and by a Rehabilitation Engineering Re- 
search Center Grant from the National Institute on 
10. He admits, though, that this may be due in part to 
teacher talk, a related communication phenomenon. 
11. We have had contact with Michael Elhadad about the 
possibility of implementing machinery in FUF that will 
allow the kind of dynamic ordering we require. 
78 
Disability and Rehabilitation Research 
(#H 133E30010). Additional support has been pro- 
vided by the Nemours Foundation. We thank Xin- 
gong Chang for his work on implementing the 
mal-rule grammar for the ASL writing project. 
The implementation uses the bottom-up augment- 
ed context-free chart parser from \[A1195\]. Thanks 
also goes to Karen Hamilton for her implementa- 
tion of the database used for our error analysiL 

References 
\[A1195\] James Allen. Natural Language Un- 
derstanding, Second Edition. Benjamin/Cum- 
mings, CA, 1995. 
\[Bak80\] C. Baker. Sentences in American 
Sign Language. In C. Baker and R. Battison, edi- 
tors, Sign Language and the Deaf Community, 
pages 75-86. National Association of the Deaf, 
Silver Spring, MD, 1980. 
\[BC80\] C. Baker and D. Cokely. American 
Sign Language: A Teacher's Resource Text on 
Grammar and Culture. TJ Publishers, Silver 
Spring, MD, 1980. 
\[Ber88\] Gerald P. Berent. An assessment of 
syntactic capabilities. In Michael Strong, editor, 
Language Learning and Deafness, Cambridge 
Applied Linguistic Series, pages 133-16l. Cam- 
bridge University Press, Cambridge; New York, 
1988. 
\[BMK74\] Nathalie Bailey, Carolyn Madden, 
and Stephen D. Krashen. Is there a "natural se- 
quence" in adult second language processing'? 
Language Learning, 24(2):235-243, 1974. 
\[BP78\] C. Baker and C. Padden. Focusing on 
the non-manual components of American Sign 
Language. In P. Siple, editor, Understanding Lan- 
guage through Sign Language Research, pages 
27-58. AP, New York, 1978. 
\[BPB83\] K. Bellman, H. Poizner, and 
U. Bellugi. Invariant characteristics of some mor- 
phological processes in American Sign Language. 
Discourse Processes, 6:199-223, 1983. 
\[Cha83\] Craig Chaudron. Foreigner talk in the 
classroom - an aid to learning? In Herbert W. Se- 
liger and Michael H. Long, editors, Classroom- 
Oriented Research in SLA, pages 127-145. New- 
bury House, Rowley, MA, 1983. 
\[Cha93\] Eugene Chamiak. Statistical Lan- 
guage Learning. MIT Press, Cambridge, MA, 
1993. 
\[Cry82\] David Crystal. Profiling Linguistic 
Disability. Edward Arnold, London, 1982. 
\[DB74\] Heidi C. Dulay and Marina K. Burr. 
Natural sequences in child second language acqui- 
sition. Language Learning, 24(1):37-53, 1974. 
\[Elh93\] M. Elhadad. FUF: the Universal Uni- 
fier User Manual Version 5.2, Jun. 1993. 
Gas84\] S. Gass. A review of interlanguage 
syntax: Language transfer and language univer- 
sals. Language Learning, 34(2):115-132, 1984. 
\[GS83\] S. Gass and L. Selinker, editors. Lan- 
guage Transfer in Language Learning. Newbury 
House, Rowley, MA, 1983. 
\[Hat83\] Evelyn Hatch. Simplified input and 
second language acquisition. In Roger W. Ander- 
sen, editor, Pidginization and Creolization as 
Language Acquisition, pages 64-86. Newbury 
House, Rowley, MA, 1983. 
\[Haw91\] John A. Hawkins. Language univer- 
sals in relation to acquisition and change: A trib- 
ute to Roman Jakobson. In Linda R. Waugh and 
Stephen Rudy, editors, New Vistas in Grammar." 
lnvariance and Variation, volume 49 of Current 
Issues in Linguistic Theory, pages 473--493. John 
Benjamins, Amsterdam / Philadelphia, 1991. 
\[HS83\] R.J. Hoffmeister and C. Shettle. Ad- 
aptations in communication made by deaf signers 
to different audience types. Discourse Processes, 
6:259-274, 1983. 
\[Ing78\] R.M. Ingrain. Theme, rheme, topic 
and comment in the syntax of American Sign Lan- 
,,uaoe Sign Language Studies, 20:193-218, Fall 
t978. 
\[Ing89\] David Ingram. First Language Acqui- 
sition." Method, Description, and Explanation. 
Cambridge University Press, Cambridge; New 
York, 1989. 
\[KB79\] E.S. Klima and U. Bellugi. The Signs 
of Language. Harvard University Press, Cam- 
bridge, MA, 1979. 
\[KG83\] J. Kegl and P. Gee. Narrative/story 
structure, pausing and American Sign Language. 
Discourse Processes, 6:243-258, 1983. 
79 
\[KH87\] Edward L. Keenan and Sarah Hawk- 
ins. The psychological validity of the accessibility 
hierarchy. In Edward L. Keenan, editor, Universal 
Grammar: 15 Essays, pages 60-85. Croon Helm, 
London, 1987. 
\[KK78\] Richard R. Kretschrner Jr. and 
Laura W. Kretschmer. Language Development 
and Intervention with the Hearing Impaired. Uni- 
versity Park Press, Baltimore, MD, 1978. 
\[Kra81\] Stephen Krashen. Second Language 
Acquisition and Second Language Learning. Per- 
gamon Press, Oxford, 1981. 
\[Kra82\] Stephen D. Krashen. Principles and 
Practice in Second Language Acquisition. Perga- 
mon Press, Oxford, 1982. 
\[Lee74\] Laura Lee. Developmental Sentence 
Analysis: A Grammatical Assessment Procedure 
for Speech and Language Clinicians. Northwest- 
em University Press, Evanston, IL, 1974. 
\[LidS0\] Scott K. Liddell. American Sign Lan- 
guage Syntax. Mouton Publishers, 1980. 
\[McL87\] R. McLaughlin. Theories of Second- 
Language Acquisition. Edward Arnold, London, 
1987. 
\[MPS96\] K.F. McCoy, C. A. Pennington, and 
L.Z. Suri. English error correction: A syntactic 
user model based on principled "mal-rule" scor- 
ing. In Proceedings of User Modeling '96, Kailua- 
Kona, HI, 1996. 
\[Pad81\] C. Padden. Some arguments for syn- 
tactic patterning in American Sign Language. 
Sign Language Studies, 32:239-259, Fall 1981. 
\[Pad82\] C. Padden. Interaction of Morpholo- 
gy and Syntax in American Sign Language. PhD 
thesis, UCSD, 1982. 
\[PQ73\] D. Power and S. Quigley. Deaf chil- 
dren's acquisition of the passive voice. Journal of 
Speech and Hearing Research, 16:5-11, 1973. 
\[QP84\] S.P. Quigley and P.V. Paul. Lan- 
guage and Deafness. College-Hill Press, Inc., San 
Diego, 1984. 
\[QPS77\] S.P. Quigley, D.J. Power, and 
M. W. Steinkamp. The language structure of deaf 
children. The Volta Review, 79(80):72-84, Febru- 
ary-March 1977. 
\[QWM76\] S. Quigley, R. Wilbur, and 
D. Montanelli. Complement structures in the lan- 
guage of deaf students. Journal of Speech and 
Hearing Research, 19:448-457, 1976. 
\[RQP76\] W.K. Russell, S. P. Quigley, and D.J. 
Power. Linguistics and Deaf Children: Transfor- 
mational Syntax and Its Application. The Alex- 
ander Graham Bell Association for the Deal', Inc., 
Washington, D.C., 1976. 
\[Rue90\] Robert Rueda. Assisted performance 
in writing instruction with learning-disabled stu- 
dents. In Luis C. Moll, editor, Vygotsky and Edu- 
cation." Instructional Implications and 
Applications of Sociohistorical Psychology, pages 
403--426. Cambridge University Press, Cam- 
bridge, 1990. 
\[Sac90\] Oliver W. Sacks. Seeing Voices. Uni- 
versity of California Press, Berkeley and Los An- 
geles, CA, 1990. 
\[S1e82\] D. Sleeman. Inferring (mal) rules 
from pupil's protocols. In Proceedings of ECA1- 
82, pages 160-164, Orsay, France, 1982. ECAI- 
82. 
\[Sto60\] W.C. Stokoe, Jr. Sign Language 
Structure. Studies in Linguistics Occasional Pa- 
pers, (8), 1960. 
\[Str88\] Michael Strong. Language Learning 
and Deafness. Cambridge University Press, New 
York, 1988. 
\[Sur91\] LindaZ. Suri. Language transfer: A 
foundation for correcting the written English of 
ASL signers. Technical Report TR-91-19, Dept. 
of CIS, University of Delaware, 1991. 
\[Sur93\] LindaZ. Suri. Extending Focusing 
Frameworks to Process Complex Sentences and 
to Correct the Written English of Proficient Sign- 
ers of American Sign Language. PhD thesis, Uni- 
versity of Delaware, 1993. Available as Dept. of 
CIS Technical Report TR-94-21. 
\[Tar82\] Elaine E. Tarone. Systematicity and 
attention in interlanguage. Language Learning, 
32(1):69-84, 1982. 
\[Vyg86\] Lev Semenovich Vygotsky. Thought 
and Language. MIT Press, Cambridge, MA, 
1986. 
\[WVJ78\] RalPh M. Weischedel, Wilfried M. 
Voge, and Mark James. An artificial intelligence 
approach to language instruction. Artificial Intel- 
ligence, 10:225-240, 1978. 
