See discussions, st ats, and author pr ofiles f or this public ation at : https:www .researchgate.ne tpublic ation241734483 [631458]

See discussions, st ats, and author pr ofiles f or this public ation at : https://www .researchgate.ne t/public ation/241734483
English Articles and Modals in the Writing of Some Batswana Students
Article    in  Languag e Culture and C urriculum · Dec ember 2007
DOI: 10.2167/lc c336.0
CITATIONS
3READS
2,225
1 author:
Some o f the author s of this public ation ar e also w orking on these r elat ed pr ojects:
Translanguaging View pr oject
Languag e Planning View pr oject
Modupe Alimi
Univ ersity of Botsw ana
13 PUBLICA TIONS    32 CITATIONS    
SEE PROFILE
All c ontent f ollo wing this p age was uplo aded b y Modupe Alimi on 13 Januar y 2015.
The user has r equest ed enhanc ement of the do wnlo aded file.

0790-8318/07/03  209-14  $20.00/0   2007 M.M.  Alimi
LAngu Age, CuL ture And CurriCuLuM   Vol. 20, no. 3, 2007
209English Articles and Modals in the
Writing of Some Batswana Students*
Modupe M. Alimi
Department of English, University of Botswana, Gaborone, Botswana
What syntactic patterns emerge in students’ use of articles and modals? What are the
reasons for these patterns? What implications do the findings of the study have for
English language instruction in Botswana? Exactly 1556 essays comprising class assign –
ments, written seminar presentations, test papers and examination scripts from 514
randomly selected students of the University of Botswana were analysed. The findings
indicate that there were systematic omissions, substitutions and insertion of the
definite and indefinite articles as well as recurrent use of the expression can be able.
Students’ indication of different forms of epistemic modality was confined to the use
of could while complex verb phrases involving negation had their constituents
reordered such that the negative operator not consistently succeeded the perfective
auxiliary. The study shows that the reasons for these errors are both intra- and inter –
lingual, namely the complexities of the two grammatical structures, articles and modals,
and the influence of L1. Since the Botswana Senior Secondary Assessment Syllabus
specifies ‘accuracy’ in the use of grammatical forms, by demanding high language
achievement standards, this paper argues that a return to the teaching of basic grammar
in the high school is important for the students’ mastery of articles and modals.
doi: 10.2167/lcc336.0
Keywords: second language learning, english articles and modals
Introduction
Learning  and  using  modals  and  articles  appropriately  are challenging  tasks 
for learners  of english  as a second  language. the  tasks  are made  more  difficult 
where  the learners’  first  language  lacks  equivalent  lexical  or grammatical  items 
of the L2. duskova  (1983),  for instance,  observes  that  the difficulties  with english 
articles  encountered  by Czech  speakers  learning english  are partly  the result  of 
the absence  of a similar  system  in the first  language.  Platt  et al.  (1984)  assert  that 
in many  eSL contexts,  the use of articles  deviates  from  standard  usage  largely 
because  of mother  tongue  interference.  Cook  (1994:  27) notes  that ‘the most 
persistent  recurrent  problems  in terms  of slips  and minor  errors  amongst  all 
groups  of the most  sophisticated  users  of english  as a second  language  involve 
confusion  with  the words  the, a (an)  and their  proper  necessary  non-use’.  Pica 
(2002:  2) also identifies  certain  sentence  modality  and modification  of nouns 
with  articles  as grammatical  structures  that are very  difficult  for second  lan-
guage  learners. in  her view,  L2 learners’  difficulty  with  these  structures  invari –
ably  results  in ‘incomplete  or incorrect  representations  in their  interlanguage 
*  this is a revised  version  of the paper  presented  at the 37th  international  Annual  iAteFL Conference 
at the Brighton  Conference  Centre  and the Quality  Hotel,  Brighton,  22–26  April,  2003.

210 Language, Culture and Curriculum
development  and  is manifested  in either  omission  or substitution  of incorrect 
versions  for correct  ones’. thompson  (2002),  identifying  with  these  assertions, 
reiterates  that  learning  modal  auxiliaries  is an intricate  process  ‘involving 
 learning  how  to form  these  modals  correctly,  recognising  categories  of modalities 
and choosing the appropriate modal for expressing modality’.
in line with  the foregoing,  this paper  examines  students’  use of articles  and 
modals  at the university  of Botswana  by providing  answers  to the following 
questions.  What  syntactic  patterns  emerge  in students’  use of articles  and 
modals?  What  are some  of the reasons  for these  patterns  and what  implications 
do the findings  of the study  have  for english  language  instruction  in Botswana?
Articles  which  are classified  as determiners  modify  nouns  in relation  to their 
referential  properties  in standard  usage.  they  are used  to indicate  generally 
whether  or not the speaker  believes  his addressee(s)  can identify  the referent  of 
the nP. this  identification  of the referent  may  be definite/specific,  indefinite/
specific  or generic  (Quirk  et al. , 1972:  149–150).  According  to Master  (1994:  230), 
the misuse  of articles  rarely  impedes  communication  in the spoken  form.  in 
written  language,  however,  writers  are required  to be precise  with  them  because 
of the absence  of extra  linguistic  cues  such  as gestures  that  could  facilitate  com –
munication.  Platt  et al.  (1984:  54) observe  that  the rules  for using  articles,  defi-
nite  and  indefinite,  in the new  englishes  are not  the same  as in the native 
environment. in  many  of the new  varieties,  the emphasis  seems  to be on ‘the 
particular/non-particular  distinction  rather  than  the  definite/indefinite  or  
the known/unknown  division  of the more  established englishes’. this  distinc –
tion  is quite  useful  in appreciating  the conflict  that  learners  face  in learning 
some L2 grammatical forms.
Precision  in the use of modals,  whether  in the spoken  or written  form,  is 
important  because  of the shade  of meanings  (obligation,  volition,  ability,  predic –
tion,  permission  and probability)  they  signal.  Papafragou  (1997)  commenting  on 
the intricacies  of modal  verbs,  in  english  observes  that ‘complex  factors  will 
interact  in the acquisition  of a system  as complicated  as the english  modal  set’. 
Modal  verbs  are generally  classified  as conveying  epistemic  or non-epistemic 
(deontic  or root)  meaning.   in their  epistemic  function,  ‘speakers  use them  to 
comment  on the content  of the clause’  while  in their  non epistemic  function,  they 
are used  to intervene  in the speech  event  (downing  & Locke,  2002:  383).
As to the acquisition  of the two uses  of modal  verbs,  the consensus  in the 
literature  (see Perkins,  1983;  Shepherd,  1982;  Wellman,  1990;  Wells,  1979, 
1985)  is that first language  learners  acquire  the non-epistemic  use of modals 
much  earlier  than  their  epistemic  functions.   this difference,  according  to 
Papafragou  (1997),  has to do with  the ‘stronger  demands  on the human  meta-
representational  device’  that epistemic  modality  makes  which  also automati –
cally  delays  the appearance  of ‘genuine  epistemic  instances’  till the later  stages 
of language  development.  the experience  of second  language  learners  with 
proficient  epistemic  use of modal  verbs  is unlikely  to differ  significantly 
because  of these  ‘higher-order  meta  representational  cognitive  abilities’.
The role of instruction
there  are different  views  on the role  of instruction  in general  in second  lan-
guage  acquisition. ellis  (1994:  633)  is of the view  that  ‘there  is some  evidence  to

English Articles and Modals in Batswana Students’ Writing 211
suggest  that  formal  instruction  directed  at features  that  are not subject  to devel –
opmental  constraints  can succeed’.  For example,  Pienemann  (1984)  in a study 
involving  learners  of  german  reports  that  the learners  who  were  taught  the 
german  copula  were  able  to use this structure  accurately.  Based  on this finding, 
ellis  (1994:  633)  suggests  that  accuracy  in the use of simple  structures  improves 
with  instruction. in  their  study  on pronoun  acquisition,  Seow  and tay  (2004) 
report  that  classroom  instruction,  formal  and  informal,  produced  varied  out-
comes  for the two  groups  of learners.  For  example,  the learners  who  were 
instructed  formally  were  generally  more  competent  in pronoun  use. they  there –
fore  recommend  that  ‘complementing  implicit  learning  with  occasional  strong 
doses of explicit teaching of grammar’ would be beneficial to eSL learners.
thus,  different  factors  determine  the success  or otherwise  of instruction  in 
second  language  acquisition.  As already  indicated,  articles  and  modals  are 
complex  grammatical  structures.  Pica’s  (1983,  1985:  214)  observation  that  the 
acquisition  of articles  does  not seem  to be influenced  by instruction  because  of 
the difficulty  of harmonising  linguistic  and  extra  linguistic  factors  essential  for 
understanding  articles  and  gaining  competence  in their  use,  reiterates  this 
point.  She,  however,  suggests  that  ‘a more  systematic  presentation’  could  pro-
duce  a significant  result.  Master  (1994:  244)  on the other  hand  correlates  signifi –
cant  ‘increase  in the mean  post-test  scores  of his subjects  with  the systematic 
presentation  of the articles’.  the  view  held  in this  paper  is that  individual 
 learners  have  differential  learning  rates  and  that  systematic  instruction  in the 
second  language  environment  should  accelerate  learners’  acquisition  of modals 
and  articles. this  is the view  also  indirectly  expressed  by Cook  (1994:  28) when 
he attributes  the misuse  of the articles,  especially  in Africa,  to poor  instruction, 
which  itself  is the result  of inadequate  resources  (books)  and  lack  of well-trained 
eSL teachers.
ESL implications
it is important,  at this  point,  to briefly  examine  the different  approaches  to 
language  teaching  in view  of the implications  that  these  approaches  have  for 
learners’  proficiency  particularly  in the  eSL  context.  Larsen-Freeman  (2001a: 
251–266,  2001b:  34–41)  indicates  that  language  teaching  has alternated  between 
two  major  approaches,  formal  and  functional. early  formal  grammarians  pro-
posed  the establishment  of grammatical  categories  based  on the distribution  of 
structures  in sentences  with  Chomsky’s  transformational  generative  grammar 
as a major  influence.  His  distinction  between  competence  and  performance 
influenced  language  learning  which  was  conceived  ‘as a process  where  learners 
abstract  the rules  of the target  language  through  hypothesis  formation  and 
 testing’  (Larsen-Freeman,  2001b:  35).  thus,  the  formal  approach  explains 
 syntactic  facts  on the basis  of formal  grammatical  properties.  the  functional 
approach,  which  provided  the theoretical  base  for Communicative  Language 
teaching  (CLt),  on the other  hand,  emphasises  learners’  development  of com-
municative  ability  in the language  and  the subsequent  acquisition  of its forms 
because  it is assumed  that  rules  do not just generate  grammatical  sentences,  but 
that sentences are means of expressing meaning.
the  Botswana general  Certificate  of Secondary education teaching  Syllabus, 
which  stipulates  CLt  as the  teaching  methodology,  requires  that  students

212 Language, Culture and Curriculum
develop  knowledge  of the structure  and  form  of english  as used  in a range  of 
situations.  in  addition,  the Botswana  Senior  Secondary  Assessment  Syllabus 
specifies  a very  high  standard  by requiring  students  to write  without  gross 
errors,  including  errors  of tense,  spelling,  punctuation,  article  and  preposition. 
if the overall  goal  of english  language  learning,  as specified  by the teaching  and 
assessment  syllabi  is grammatical  accuracy,  then  it is logical  to teach  the stu-
dents  grammatical  forms  such  as articles  and  modals  in spite  of the teaching 
methodology  prescribed  in the teaching  syllabus. this  study  is important  and 
relevant,  therefore,  because  it delineates  the syntactic  patterns  characterising 
students’  usage  of english  articles  and  modal  verbs  (central  modals),  explores 
some  of the reasons  for the recurrent  patterns  of usage,  and  makes  suggestions 
for effectively teaching these grammatical forms to the learners.
Methodology
the data  for the study  comprise  essays  obtained  from  514 graduate  and 
 undergraduate  students  in the  department  of  english  at the  university  of 
Botswana.  there  were  100 students  each  from  the first,  second  and fourth  year 
 programmes,  and 200 and 14 students,  respectively,  from  the third  year  and 
graduate  class.  the graduate  class  had the fewest  number  of students  because   
of the generally  low enrolment  in the programme.  the third  year  had the largest 
because,  over  the period  of the study,  the researcher  had taught  at least  one 
course  per semester  at this level  and therefore  had more  access  to students  within 
this group.  except  for the graduate  group,  which  constituted  a very  small  class, 
all the students  were  randomly  selected.  For each  undergraduate  student,  three 
essays:  two continuous  assessment  test papers  and one examination  paper  were 
examined.   in the case of each  post  graduate  student,  four essays  comprising 
three  seminar  presentations  and one examination  paper  were  analysed.
the  use  of graduate  and  undergraduate  students  provides  information  on 
the students’  competencies.  For instance,  the initial  competence  of the students 
in the use of articles  and  modals  is revealed  by their  writing  in the first  year 
while  the samples  from  the second,  third  and  fourth  year  reveal  apparent 
progress  and/or  fossilisation  tendencies. the  essays  from  the graduate  students 
who are predominantly teachers tend to affirm fossilisation.
All the essays  were  content  analysed  and  the recurrent  patterns  of usage  of 
articles  and  modals  were  identified  and  classified.  All the errors  reported,  except 
the incorrect  placement  of the negative  marker  not, which  is confined  to essays 
from  the first  year,  are widespread  in the essays.  there  were  other  types  of 
 deviant  usage  in the data  but these  were  ignored  as they  do not fall within  the 
scope  of the study. in  the next  section  of the paper,  the findings  regarding  each 
question for the study are presented beginning with the patterns of usage.
Findings
Analysis  of the student  essays  showed  that  there  are three  major  patterns  of 
error  in the use  of articles,  and  three  others  in the use  of modals.  these  are 
 presented,  with  examples,  in tables  1–5. thereafter,  some  of the reasons  for the 
students’ usage are provided and the implications of the findings discussed.

English Articles and Modals in Batswana Students’ Writing 213
Articles
the  data  show  three  different  error  types  in students’  use  of the articles: 
 missing  articles  a/an and  the (table  1), substitution  of one  article  for another 
(table 2), and redundant use of articles (table 3).
in table  1, excerpts  from  the students’  essays  across  all the levels  involving 
omission  of the articles  are presented. there  are indications  from  the data  that 
the learners  are unable  to distinguish  between  definite/specific,  indefinite/
specific or even generic referents.
table  2 shows  some  examples  of sentences  involving  the substitution  of one 
article  for another. the  substitution  of the indefinite  article  for the definite  was 
prevalent in all the students’ essays irrespective of their level. this implies that 
the students  do not distinguish  between  using  the indefinite  article  a(n) to show 
that  a language  user  makes  no assumption  that  his addressee(s)  can  identify   
the referent  of the  nP  and  using  the definite  article  to indicate  that  his/her 
Table 1   Missing articles
S/N Sentence Article(s) omitted
1 (*) name can be used as a referring expression  a
2 (*) Hurricane destroyed all the buildings in (*)   
 Philipinesthe
3 Macbeth instead of waiting to attain his destiny   
 in (*) normal waya
4 According to (*) bible, it says thou shall not kill  the
*indicates that an article is missing.
Table 2   Substitution of articles
S/N Sentence Article(s) substituted
1 this method is good because it gives the 
students freedom to work and feel really 
involved. Student get motivated when they feel involved in (a) learning process ‘a’ substituted for ‘the’
2 in english, (a) word ‘dog’ would mean a dog 
as an animal but contradictory to the French meaning of a dog. ‘a’ substituted for ‘the’
3 (A) word insult as a verb can not be divided 
into syllables unless stress marks are used ‘a’ substituted for ‘the’
4 Also when we look at Okonkwo, he ruled his 
family with (a) hand of a lion‘a’ substituted for ‘the’
() indicates article incorrectly used.

214 Language, Culture and Curriculum
addressee(s)  can  actually  identify  the referent.  the  data  suggest  that  in the 
 students’  view,  there  is apparently  no difference  between  the definite/specific,
indefinite/specific  and  generic  referents.
table 3 illustrates  students’  redundant  use of articles.  the analysis  indicates 
that redundant  use of articles  results  from  using  the indefinite  article  with  non-
countable  nouns  as in  examples  2 and 3 (hot tea, good morality .)  it was also 
observed  that the students  had difficulty  indicating  indefinite/specific  refer –
ence  with  plural  countable  nouns  which  require  zero  article  as in example  4 
(speech sounds ). in example  1, the use of ‘a’ is classified  as redundant  because 
the word  meaning  indicates  generic  reference.  if an article  is to be used,  then 
‘the’  would  be more  appropriate.   the findings  in  table 3 therefore  support 
those  of tables  1 and 2. Students  are unable  to determine  when  a referent  is 
definite/specific,  indefinite/specific  or even  generic,  which  may,  itself,  result 
from  inadequate  understanding  of count,  non-count,  mass  and abstract,  nouns 
in english.
Modals
there  are three  observations  regarding  the use of modals. the  first  is that  stu-
dents  incorrectly  order  the elements  of complex  verb  phrases,  consisting  of 
modal  and  perfective  auxiliaries,  by placing  the negative  marker  not after  the 
perfective  auxiliary. the  second  is that  students  depend  excessively  on can/could  
for the expression  of different  types  of epistemic  modalities. these  two  patterns 
of usage  are confined  to the essays  from  first  year  students. the  third  observa –
tion  is that  students  systematically  use the modal  can with  be able . this  pattern 
is widespread.
in table  4, some  examples  relating  to the incorrect  placement  of the negative 
marker  as well  as the excessive  use of can/could  are presented.  Students  prefer 
to place  the negative  marker  not after  the perfective  auxiliary. there  is ample 
evidence to show that the  incorrect placement of  the negative marker is  closely 
linked  with  its fixed  position  within  the verb  phrase  in the learners’  L1. the 
findings  also  show  that  the modal,  can/could  is used  almost  always  to express Table 3  redundant use of article
S/N Sentence Article inserted
1 Free variation are sounds that contrast with one 
another without changing (a) meaning a
2 John drank ( a) very hot tea  a
3 A society with women will have (a) good morality 
because women have qualities that provide a 
comfortable environment to live in a
4 the tongue is made is made up of muscles found in 
the oral tract . . . it moves in different positions . . . this makes it possible for (the) speech sounds to be produced the
() indicates redundant use of articles.

English Articles and Modals in Batswana Students’ Writing 215
different  types  of commitment  to propositions.  For instance, examples  1 and  3, 
should  actually  express  prediction  based  on inference  instead  of possibility. 
Similarly,  can/could  are used  in conditional  clauses  where  lexical  verbs  would 
have been more appropriate as in 2, 3 and 4.
in  table  5, examples  involving  the use  of can be able  for can or is able  are 
 presented. the  expression  can be able  which  is used  to denote  ability  featured 
prominently  in all the samples,  from  the first  year  to the graduate  students,  and 
in all the cases,  it may  be substituted  with  be able . in the next  section,  the second 
concern  of this  paper,  the reasons  for the patterns  observed  with  respect  to the 
use of articles and modals, is addressed.Table 4  incorrect placement of the negative marker with can/could
S/N Sentences Remark
1 if they had told Oedipus that they adopted 
him when he asked them, Oedipus  could
have not decided  to run away from the oracle use of  could  for would   
   and incorrect place –
ment of  not
2 if only he *( could have left)  Oedipus to die, 
Oedipus  would have no t encountered  this 
problemverb phrase word order
3 if they *( could have)  told him that then 
Oedipus wouldn’t have fled away so if he 
did any of these, Oedipus  could have not  ran 
away to thebes where he was destined for incest and patricide use of 
could  for would   
   or might , and incorrect 
placement of  not
4 if it *( would have not  been) for the kgotla, there 
would have not been  a herdman who worked 
hand in hand with the kgosi incorrect placement of   
 not
*() indicates that items enclosed may be replaced with lexical verbs.
Table 5  the use of  can be able  for is able
S/N Sentence
1 After testing, feedback is given so that the learner  can be able  to gauge her 
ability and be able to make well informed decisions on whether to continue or look for a job
2 Polysemy is whereby the meaning of a word is included in another word 
and then that word which its meaning is included  can be able  to also refer 
to the initial word
3 Open classes do allow the formation of new meanings as we acquire new 
technology so that we  can be able  to name things that were not existing 
before
4 thank’s god to have a genius friend like myself, because i already had a 
plan to sneak out during the night so that he  can be able  to attend the party

216 Language, Culture and Curriculum
Reasons for the Patterns of Usage of Articles and Modals
the  patterns  of articles  and  modals  usage  by the learners  in this study,  which 
are at variance  with  standard  usage,  may  be attributed  to two  major  interre –
lated  reasons:  intra-lingual,  particularly  the difficulties  associated  with  the L2 
structures, and interlingual transfer.
English articles and modals
As the  literature  indicates,  articles  are extremely  difficult  for non-native 
speakers  or speakers  whose  first  language  is not a romance  language  for various 
reasons.  First,  many  of such  languages  do not have  articles  and  therefore  use 
entirely  different  systems  of marking  noun  phrases.  Second,  the rules  for using 
articles,  which  are complicated,  have  too many  exceptions  and  these  ultimately 
confuse  the learners.  For example,  the generic/specific  distinction  respectively 
in boys and  a boy  is straight  forward,  but the definite/specific  and  indefinite/
specific  distinctions  are quite  challenging  tasks  for the students. the  omission 
of the articles  in the data  may  be attributed  to this  dichotomy.  For instance,  in 
example  2, table  1, (The) hurricane destroyed all the buildings in the Philippines , the 
writer  obviously  intended  to be definite  and  specific.  On the other  hand,  the 
writer  of the sentence,  (A) name can be used as a referring expression  (example  1 
table  1) intended  to be indefinite  but  specific.  Yet in the two  sentences,  the 
 articles  the and  a are omitted. the  redundant  use of articles  in table  3, especially 
with  plural  countable  and  mass  nouns,  which  is attributed  to the students’  lack 
of distinction  between  definite/specific  and  indefinite/specific  (that  requires 
zero  use  of article)  also  underscores  the complexities  involved  with  learning 
these L2 structures.
Similarly,  generic  reference  appears  simple  with  count  nouns,  but the rule 
becomes  complicated  with  concrete,  abstract,  mass  and  plural  nouns  which  are 
used  with  zero  article  when  the intended  reference  is generic. the  modification 
of some  of the non-countable  nouns  in the data  using  the indefinite  article,  for 
example,  semantic knowledge  and good morality  in table 3, supports  this position.
Interlingual effects
While  the inability  to distinguish  between  the definite  and  indefinite  forms  of 
determining  the noun  phrase  is largely  a product  of the numerous  rules  and 
exceptions  in  english,  there  are obviously  interlingual  forces  at play.  this  is 
because  the students  are used  to a totally  different  system  (predominantly 
Setswana)  in which  nouns  are determined  not by the use of articles  but by the 
use of demonstratives and some quantifiers (university of Botswana, 2000: 6). 
the examples below illustrate this point:
(1)  the lions roar   
ditau  di-a-duma   
Lions  subject agr -pres.-tense-roar
(2)  A lion roars   
tau  e-a-duma   
Lion  subject.agr -pres.tense-roar
(3)  Lions roar   
ditau  di-a-duma   
Lions  subject agr -pres.-tense-roar

English Articles and Modals in Batswana Students’ Writing 217
(4)  the lion is roaring   
tau  e-a-duma   
Lion  subject.agr -pres.-tense-roar
(5)  A lion is roaring   
tau  e-a-duma   
Lion  subject.agr -pres.-tense-roar.
(6)  Lions are roaring   
ditau  di-a-duma   
Lions  subject agr -pres.-tense-roar
(7)  the lions are roaring   
ditau  di-a-duma   
Lions  subject agr -pres.-tense-roar
(8)  A boy ran   
Mosimane  o-ne-a-taboga.   
Boy  subj.agr. past- run
(9)  the boy ran   
Mosimane  o-ne-a-taboga.   
Boy  subj.agr. past- run
(10)  that boy ran   
Mosimane yole  o-ne-a-taboga.   
Boy  that  subj.agr. past- run
the  excerpts  above  show  that  specification  of the noun  in Setswana,  except  for 
(10),  is essentially  the same  irrespective  of the meaning  intended,  definite/
specific,  indefinite/specific  or generic,  and  irrespective  of the  nouns  type, 
(countable, non-countable or mass).
in order  to use epistemic  modals  successfully,  speakers  must  necessarily  employ 
their  ‘meta-logical  and meta-cognitive  abilities  to perform  deductive  operations 
on abstract  propositions  (i.e. on the content  of their  beliefs)  and to arrive  at logical 
conclusions’  (Papafragou,  1997).  the examples  in table 4 which  are from  the first 
year  samples  show  that the students  are more  inclined  to use could  (possibility) 
where  would  (prediction)  is more  appropriate.  the nature  of the task,  to make 
 logical  deductions  from  the events  which  already  took  place  for example  in King
Oedipus  and Macbeth  demands  that the students  demonstrate  some  knowledge   
of epistemic  modality,  that is ability  to make  predictions.  their choice  of could   
in place  of would  seems  to be a reflection  of not correctly  determining  and stating 
their  commitment  or attitude  to propositions.  this situation  is further  compli –
cated  by the fact that Setswana,  like many  other  Bantu  languages,  has no modal 
verbs.  in the examples  of english  sentences  translated  into Setswana  below  could,
should might and would  are represented  by just one lexical  item  kabo:
(1)  i could not have eaten   
gagona jaaka  ke-kabo ke jele   
not that  subj.agr - could  subj.agr -eat
(2)  i should not have eaten   
Ke kabo  ke-sa-ja   
subj.agr  could  subj.ag - not-eat
(3)  i might not have eaten   
Ke kabo  ke-sa-ja   
Subj.agr  could  subj-agr - not-eat

218 Language, Culture and Curriculum
(4)  i would not have eaten   
Ke kabo  ke-sa-ja   
Subj.agr could  subj-agr - not-eat
(5)  i can not eat any time   
ga  ke-kake  kaja  ka  nako  epe  
not  subj.agr -not  pres.eat  at  time  any
Furthermore,  the fact that  the substitution  of would  with  could  is more  confined 
to the first  year  samples  lends  credence  to the difficulty  of the subject  itself  and 
suggests  that  this  group  of students  is still  extremely  dependent  on can/could , 
perhaps  the first  modal  auxiliaries  that  they  acquired  to signal  invariably  vari-
ous types of commitment to propositions.
the  second  observation  is that  the incorrect  word  order  of negative  verb 
phrases  consisting  of the modal  and  perfective  auxiliaries  where  the negative 
operator  not succeeds,  rather  than  precedes,  the perfective  auxiliary  seems  to be 
closely  associated  with  the structure  of the verb  phrase  in Setswana. it  is impor –
tant  to observe  that  negation  in Setswana  is much  more  complex  because  differ –
ent items,  lexical  verbs  and  particles,  [e.g.  gagona  and  sa in (1) and  (2)] may  be 
used  to express  negation  and  the particles  may  be interrupted  by other  verbal 
elements  as in example  (5) above. in  addition,  the perfective  auxiliary  have does 
not have  a direct  equivalent  in Setswana  and  the position  of the negative  opera –
tor not is fixed. it  comes  just  before  the main  verb  as shown  in the translated 
versions  of (2), (3) and  (4) above. these  three  factors  seem  to influence  the con-
sistent  post  perfective  auxiliary  placement  of the negative  operator  in the first 
year samples.
the  third  pattern  of usage  which  features  in all the samples  is the contraption 
can be able.   this  is a type  of usage  which  is induced  by interlingual  factors. the 
usage  is a product  of the conflation  of two  verb  forms  in Setswana:  ka, the verb 
that  represents  can and  kgona  which  means  be able . it is not unusual  in Setswana 
for these  two  to co-occur  as a discourse  marker  for emphasis,  as demonstrated 
by the following examples:
(6)  O ka kgona go tsamaya   
She can be able to walk
(7)  O ka tsamaya   
She can be able to walk
even  if kgona  is omitted,  ka will  still  denote  can.  the  possibility  of the  
co-occurrence  of these  two  verbs  denoting  ability  in a single  verb  phrase  in the 
students’  L1 seems  to be the main  reason  for the prevalence  of this expression  in 
the samples. notably,  Arua  (2004)  had  reckoned  this usage  as a syntactic  feature 
of Botswana english. this  expression  is so common  and  perhaps  so ingrained 
in some  students.  they  can  actually  identify  it as an instance  of interlingual 
error and yet within the same context in their own writing use it.
Implications for English Language Instruction
the  discussions  so far indicate  that  students’  use of articles  and  modals  is 
largely  conditioned  by two  factors:  the complexity  of some  grammatical  struc –
tures  in the english  language  itself  and  the pressure  exerted  on the students  by

English Articles and Modals in Batswana Students’ Writing 219
the grammatical  structure  of their  first  language. in  this section  of the paper,  the 
major  implications  of the findings  are highlighted  and suggestions  for improving 
the use of the grammatical structures discussed.
Communicative  Language teaching  (CLt)  is the approach  used  in teaching 
english  language  in the secondary  schools  in Botswana.  While  this approach  is 
lauded  for its versatility,  it obviously  has  its short  comings  especially  in the 
second  language  environment  where  teaching  of grammatical  structures  is 
essential  for the learners  to build  their  linguistic  competence  in the target  lan-
guage.  As Seow  and  tay  (2004:  11) argue,  ‘the  informal  classroom  learning 
environment  is highly  desirable’,  but ‘explicitly  taught  grammatical  knowledge 
seems  to play  a part  to help  learners  notice  and  retain  the organised  linguistic 
systems  that,  in turn,  facilitate  the acquisition  of grammatical  accuracy’.  Without 
formal  instruction  in the rudiments  of grammar,  the learner’s  task  is made 
extremely  difficult  since  he is being  required  to apply  the rules  of a language 
that  are completely  inaccessible  to him. there  is, therefore,  the need  at the high 
school  to modify  the implementation  of the CLt  to ensure  that  students  are 
exposed to the fundamentals of english grammar.
Cook  (1994:  32), earlier  referred  to, asserts  that the rejection  of traditional 
grammar  at some  point  is one of the reasons  for ‘eSL learners’  poor  under –
standing  of  english  determiners’.   in spite  of the criticisms  of traditional 
grammar,  it is still the one form  of grammar  that provides  second  language 
learners  the solid  foundation  required  for a good  understanding  of the ‘mys –
tique  of english’.  it is therefore  important  for policy  decisions  in Botswana  to 
incorporate  the teaching  of the rudiments  of grammar  in the high  school.  in 
addition,  the grammar  instruction  would  be more  beneficial  to these  learners 
if it proceeds  from  what  the learners  know  and are familiar  with,  the forms 
of specification  and modality  in Setswana,  to the new,  what  they  are expected 
to learn,  in this case,  english  articles  and modals.  Furthermore,  the instruc –
tion should  draw  attention  systematically  to the structural  differences 
between  the learners’  first language  and the target  language,  so that the 
learners  may  become  aware  of the problems  that could  arise  when  they  sub-
stitute  L1 with  L2 structures.
the second  implication  which  derives  from  the first  is the education  language 
policy  in Botswana  where  english  as the  official  language  coexists  with 
Setswana,  the national  language. in  a number  of domains  including  the english 
language  classroom,  both  languages  interact  and perhaps  compete. even  though 
the education  language  policy  stipulates  that  the medium  of instruction  from 
standard  three  is  english,  this  is rarely  the case  in practice,  as ‘children  are 
taught  in Setswana  throughout  their  elementary  school’  (Bagwasi,  2003:  214). 
the  ancillary  to this  scenario  is the  standard  set by the  Botswana  Senior 
Secondary  Assessment  Syllabus  in english  language  which  requires  students  to 
‘demonstrate  their  ability  to handle  spelling,  punctuation  and  grammar  accu –
rately’  (Ministry  of education,  1999:  3). in  fact,  the details  of the assessment  for 
writing  stipulate  the ‘use  of appropriate  grammatical  structures  in writing’  with 
the specific  objective  of using  ‘articles  correctly’  (Ministry  of education,  1999: 
11).  Obviously  the type  of correctness  prescribed  by the assessment  syllabus 
requires  that  the learners  have  many  more  and  varied  opportunities  to practise 
the language.

220 Language, Culture and Curriculum
Finally,  the fact that  the population  for the study  came  from  the department 
of english  in the university  has far reaching  implications  for the teaching  of 
english  in Botswana  mainly  because  a large  percentage  of the graduates  from 
the department  are employed  as teachers  in the high  schools. the  prescribed 
variety  in Botswana  secondary  schools  is Standard english.  For example  one of 
the specific  objectives  of the assessment  syllabus  is that  students  should  be able 
to ‘adapt  speech  for different  purposes  and  audiences’  and  therefore  ‘use 
Standard english  in official  or formal  situations’  (Ministry  of education,  1999: 
9). A recent  study  (Arua,  2005)  to establish  the variety  of  english  taught  in 
Botswana  schools,  revealed  that  the  teachers  did  not  demonstrate  enough 
awareness  of some  cross-linguistic  structures  (such  as can be able ) peculiar  to 
Botswana  though  they  claim  that  they  teach  Standard english. these  two  related 
observations  seem  to be indicative  of some  connection  between  students’  mas-
tery  of the grammatical  structures  and  teachers’  level  of awareness. this  subject 
would however be more appropriate for future research.
Conclusion
the aim of this paper  was to examine  the recurrent  syntactic  patterns  in 
 students’  usage  of articles  and modals,  discuss  the reasons  for the patterns 
and the implications  of the findings  for  english  language  teaching  in 
Botswana.   using  essays  from  undergraduate  and graduate  students  at the 
university  of Botswana,  the findings  reveal  that articles  are regularly  omit –
ted, mixed  up, or used  redundantly;  and that the students  are more  inclined 
to express  different  types  of modality  using  only  can/could . in addition,  the 
negative  marker  not, consistently  succeeds  the perfective  auxiliary  instead  of 
preceding  it within  the verb  phrase.  the study  also shows  that the complexi –
ties involved  with  articles  and modals  make  them  difficult  to explain  to the 
students,  especially  as the grammatical  system  of their  L1 differs  significantly 
from  that of the target  language.  in terms  of the implications  of the findings, 
the paper  argues  that the educational  language  policy,  which  stipulates 
Setswana  as the medium  of instruction  from  standard  1–3 and the require –
ments  of the Botswana  Senior  Secondary  Assessment  Syllabus  apparently 
conflict.  Furthermore,  the teaching  methodology,  CLt, in the secondary 
schools  does  not provide  the forum  for adequate  instruction  in grammar  and 
that the variety  of english  that is taught  in schools  does  not seem  to focus  on 
the accuracy  of the use of these  structures.
the paper  recommends  a reconsideration  of the implementation  of the teaching 
approach  used  in secondary  schools  such  that  the importance  of grammar 
instruction  for the learners  is recognised. it  also  proposes  that  teachers  raise 
their  level  of awareness  so that  they  can provide  the type  of highly  structured 
classroom  tasks  which  will  systematically  draw  learners’  attention  to the 
 differences  between  L1 and  L2 structures.  Finally,  streamlining  the educational 
language  policy  in order  to provide  more  opportunities  for the learners  to use 
english would be of great benefit to them.
Acknowledgement
i am grateful  for the useful  comments  and  suggestions  from  my colleagues: 
Sibonile ellece, Mompoloki Bagwasi, rose Lestsholo and Arua e. Arua.

English Articles and Modals in Batswana Students’ Writing 221
Correspondence
Any  correspondence  should  be directed  to dr  Modupe  M. Alimi, department 
of  english,  university  of Botswana,  Private  Bag  0022,  gabarone,  Botswana 
(alimimm@mopipi.ub.bw).
References
Arua,  A.e.  (2004)  Botswana english:  Some  syntactic  and  lexical  features.  English World-
Wide  25 (2), 255–272.
Arua,  A.e.  (2005) using  two  varieties  in eSL  classes.  Paper  presented  at the 39th Annual
TESOL Convention , San Antonio, texas, uSA, 30 March–2 April.
Bagwasi,  M.M.  (2003) the  functional  distribution  of Setswana  and english  in Botswana. 
Language, Culture and Curriculum  16 (2), 212–217.
Cook, d.  (1994) determining  the determiners. in  W. Adegbite  and  C. Onukaogu  (eds) 
Language in Education in Nigeria: Some Critical Perspectives  (pp.  25–45). ile ife, nigeria: 
Centre for Language education and development.
downing,  A. and  Locke,  P. (2002)  A University Course in English Grammar.  London: 
routledge.
duskova,  L. (1983)  On sources  of errors  in foreign  language  learning. in  B. robinett  and 
J. Schachter  (eds)  Second Language Learning: Contrastive Analysis, Error Analysis and
Related Aspect  (pp. 215–233). Ann Arbor: university of Michigan Press.
ellis,  r.  (1994)  The Study of Second Language Acquisition . Oxford:  Oxford  university 
Press.
Larsen-Freeman, d.  (2001a) teaching  grammar. in  M. Cele-Murcia  (ed.)  Teaching English
as a Second or Foreign Language  (pp. 251–266). Boston: Heinle & Heinle.
Larsen-Freeman, d.  (2001b) grammar. in r.  Carter  and d. nunan  (eds)  The Cambridge
Guide to Teaching English to Speakers of Other Languages  (pp.  34–41).  Cambridge: 
Cambridge university Press.
Master,  P. (1994)  the  effect  of systematic  instruction  on learning  the  english  article 
system. in t.  Odlin  (ed.)  Perspectives on Pedagogical Grammar  (pp.  229–249).  Cambridge: 
Cambridge university Press.
Ministry  of education  (1997)  General Certificate of Secondary Education Teaching Syllabus:
English . gabarone: Ministry of education, Curriculum development division.
Ministry  of  education  (1999)  Senior Secondary Assessment Syllabus, English Language . 
gaborone: Ministry of education, examinations research and testing division.
Papafragou,  A. (1997)  Modality  in language  development:  A reconsideration  of the 
 evidence.  uCLWorking  Papers.  On WWW  at http://www.phon.ucl.ac.uk/home/
PuB/WPL/97papers/papafrag.pdf. Accessed december 2006.
Perkins, M. (1983)  Modal Expressions in English . London: Frances Pinter.
Pica,  t.  (1983)  Adult  acquisition  of a second  language  under  different  conditions  of 
exposure.  Language Learning  33 (4), 465–497.
Pica, t.  (1985) the  selective  impact  of classroom  instruction  on second  language  acquisi –
tion.  Applied Linguistics  6 (3), 214–222.
Pica, t.  (2002)  Subject-matter  content:  How  does  it assist  the interactional  and  linguistic 
needs of classroom language learners?  The Modern Language Journal  86 (1), 1–19.
Pienemann,  M. (1984)  Psychological  constraints  on the teachability  of languages.  Studies
in Second Language Acquisition  6, 186–214.
Platt,  J., Weber,  H. and  Ho,  M.L.  (1984)  The New Englishes . London: routledge  & Kegan 
Paul.
Quirk, r., greenbaum,  S., Leech, g.  and  Svartvik,  J. (1972)  A Grammar of Contemporary
English . London: Longman.
Seow,  A. and  tay,  g.  (2004)  the  acquisition  of  english  personal  and  possessive  pro-
nouns  in two  classroom  learning  environments.  Teaching English as a Second or Foreign
Language  8 (3). On WWW at http://tesl-ej.org/ej31/a1.html. Accessed June 2007.
Shepherd,  S. (1982)  From  deontic  to epistemic:  An analysis  of modals  in the history  of 
english,  Creoles  and  language  acquisition.  in  A. Ahlqvist  (ed.)  Papers from the 5th
International Conference on Historical Linguistics , (pp. 316–323).  Amsterdam:  Benjamins.

222 Language, Culture and Curriculum
thompson,  M. (2002)  Modals  in english  language  teaching.  Karen’s Linguistics Issues . On 
WWW at http://www3.telus.net/linguisticsissues. Accessed March 2002.
university  of Botswana  (2000)  The Structure of Setswana Sentences: An Introduction . 
department  of African  Languages  and  Literature,  Perspective  Series  2.  gaborone: 
Lightbooks, Lentswe La Lesedi (Pty) Ltd.
Wellman,  H. (1990)  The Child’s Theory of Mind . Cambridge,  MA  and  London: the  Mit 
Press.
Wells, g.  (1979)  Learning  and  using  the auxiliary  verb  in english. in  V. Lee (ed.)  Cognitive
Development: Language and Thinking from Birth to Adolescence  (pp.  250–270).  London: 
Croom Helm.
Wells,  g.  (1985)  Language Development in the Preschool Years . Cambridge:  Cambridge 
university Press.
View publication statsView publication stats