Capital mistakes The analysis of mistakes in the written production of advanced Swedish ESL learners at university level with focus on the use of capital letters

(1)

School of Humanities/School of Education Bachelor thesis GI1313 15 points Date: 05.06.08 Department of English

Supervisor: Anna Wärnsby Examiner: Ibolya Maricic

Capital mistakes

The analysis of mistakes in the written production of advanced

Swedish ESL learners at university level with focus on the use of

capital letters

(2)

Abstract

Title: Capital mistakes - The analysis of mistakes in the written production of advanced Swedish ESL learners at university level with focus on the use of capital letters

Author: Erik Söderlind

Abstract: The purpose of this essay was to see what mistakes Swedish university students make when writing English essays and to focus on one specific mistake, namely the use of capital letters. Fifty essays written by students of Swedish origin were taken from the SWICLE (the Swedish Component of the International Corpus of Learner English), corrected and the mistakes were subsequently analysed. Furthermore, six Swedish school book series were examined in order to find examples of how Swedish pupils are taught how to use capital letters in English. The results showed that mistakes concerning capital letters were very common compared to other mistakes and that the students in the study seemed to have grasped some grammatical rules on the matter, but not all. Other common mistakes made in the essays were spelling mistakes and choice of words. Most common among the mistakes concerning capital letters were overuse of capital lettering and the lack of initial capital letters when writing weekdays, months and holidays. Only one out of the six school book series went deeper into the matter of when to use capital letters in English, which suggests that one of the reasons why capital letter mistakes are so common might be that the students are not extensively taught on the matter.

(3)

4.2 Misuse of capital letters 21 4.2.1 Overuse of capital letters 22 4.2.2 Underuse of capital letters in weekdays, months and holidays 22 4.2.3 Underuse of capital letters in nationals, languages and nationalities 23 4.2.4 Capital I in first person singular, subject form 23 4.2.5 Religions, political parties and their members 24 4.2.6 Words unaccounted for 24 4.3 Capital letters in school books 24 5. SUMMARY OF THE RESULTS 26 6. DISCUSSION 27 LIST OF REFERENCES 31 APPENDIX A - TABLES OF MISTAKES

(5)

1. Introduction

More than 2000 years have passed since Seneca the younger coined his famous quote “errare humanum est”. It is still, however, used today, even though it has lost its original meaning of pointing out the difference between divinity and humanity. This quote is the springboard for this essay. It is important to bear in mind when reading this essay that all humans make mistakes. If they did not, there would be nothing to write about, no basis to start from.

School pupils today are told that making mistakes is a crucial feature of learning, not least when learning languages. The red blob on the side of a paper handed in focuses the attention of the pupil and starts a process of thought which hopefully leads to a new level of understanding of the problem. In essence, pupils learn by their mistakes. In addition, what mistakes a person makes can reveal many things to a researcher, for example where the ‘perpetrator’ is from and what level of English she or he has attained. There are, depending on how you count, about five thousand languages in the world today, each and every one of them different from the others. It would be natural to think that when learning a second language, the mistakes a learner makes are related to his or her first language. This has, however, been much debated and the different standpoints are further discussed in Section 2.

For all teachers of English everywhere, not just in Sweden, it is essential to know what mistakes are actually made by the students in order to help them improve. This study, however, is primarily focused on Swedish schools and Swedish ESL (English as a Second Language) learners. In order to evaluate the results of a minimum of nine years of schooling in the subject of English (which is compulsory in Swedish schools today), the writer of this essay chose to look at the ‘finished product’, i.e. the mistakes made by students when they have left upper secondary school which was conducted by looking at essays written by Swedish university ESL learners. Note that in order to be able to analyse more easily how the Swedish language affects Swedish ESL learners, only native speakers of Swedish were chosen as subjects for this study.

To relate my findings to theories of acquisition, the empirical study of mistakes in university essays was combined with research on what might be the reason for the mistakes. It was not, however, possible to look at all types of mistakes made by students and so, one of the more common mistakes; the misuse of capital letters, was focused on and the writer also investigated how Swedish school books present or deal with the problem. The full version of the quote mentioned at the start of this introduction is Errare humanum est, perseverare

diabolicum. To err is human, to persist is diabolical. We would not like students to become

(6)

1.1 Aim and scope

The aim of this essay is threefold. Firstly, the essay is focused on what mistakes of possible native language interference are made by Swedish ESL learners that have left the Swedish school system. This is done by looking at 50 essays produced by Swedish ESL learners at the second year of university level. The corpus used in the study (the SWICLE, see Section 3.1) only contains essays written by third and fourth term university students of English, which is why no first year students are represented in the study. The second aim is to see if one type of mistake, the misuse of capital letters, is a common mistake that Swedish learners of English make and the study will also try to determine whether this problem might be caused by native language interference through examining each occurrence of such a mistake in more detail. Hypothetically, capital letter mistakes are easily spotted and therefore chosen as the focus of this study. Thirdly, the writer of this essay aims to look at how the specific ‘mistake’, i.e. the misuse of capital letters, is presented in Swedish school books in order to see if there is a link between the mistakes and the way this issue is presented to the students. The writer is aware that teachers use other teaching material than school books, but the presentation of this issue in the books will, hopefully, give indications on how Swedish schools deal with the problem. With all of this in mind, the research questions of this essay are as follows:

1. What mistakes are made in the selected university essays written by Swedish ESL learners?

2. Is the misuse of capital letters a common mistake made by Swedish ESL learners in the present data?

3. In what contexts does the misuse of capital letters occur?

4. How is the use of capital letters in English presented in Swedish school books?

In order to allow for a better understanding of the field of linguistics dealing with errors and mistakes in language learning, theories and concepts within the field are discussed in Section 2.

The study was conducted on the basis of 50 essays written by advanced Swedish ESL learners at Swedish universities. These were taken from the SWICLE, the Swedish

Component of the International Corpus of Learner English (see Section 3.1). Thus, the essay

(7)

This allowed the writer to get a result as close as possible to the actual level of knowledge of the writers.

As both Contrastive Analysis and Error Analysis (presented in Sections 2.1 and 2.2) are in essence standpoints on what mistakes really are and how they should be treated by the teacher and the learner, it would be natural to think that this essay aims to prove or disprove either theory. However, even though this is a didactics essay, it does not attempt to set guidelines for how teachers should treat occurrences of mistakes made by their students. What the essay tries to do is simply to show what mistakes are made and in what contexts learners make them, what the teacher then chooses to do with the newfound knowledge is up to her or him.

2. Theoretical background

A teacher may seem malicious to his or her students if he or she focuses only on what mistakes are made in texts written by the language learners, instead of focusing on what the students have done well, but this is not necessarily the case. Mistakes made by students may reveal to the teacher what level of proficiency the students have and thus, what needs to be practiced further. However, the analysis of such mistakes should not be conducted without reflection. Ellis (1997: 17), for example, advocates a distinction between a mistake and an error. Mistakes, he claims, are simply the result of slips of the tongue, and must not be seen as the result of a faulty interpretation of the grammatical rules of the second language. Moreover, the speaker or writer who has made a mistake may self-correct since the rules applicable are actually known to her or him. On the other hand, an error, according to Ellis (1997: 17), is the result of a gap in the knowledge of the learner; she or he is basically unaware of the correct form. So what then of instances when, in a text, a learner uses both the correct and the incorrect form? Ellis considers this a mistake, since the learner clearly knows the correct form. He does not, however, explore the possibility that the learner knows both forms but is unsure which one to use and therefore alternates between them in order to get at least some answers correct. Ellis continues by saying that one way to test whether a deviating form is a mistake or an error is to ask the learner to correct him-/herself.1

The study of mistakes performed by second language learner (henceforth referred to as ‘L2 learners’) of English can also be said to be the study of how the first language (hereafter ‘L1’) influences the language learning process. Throughout the last century, many theories on how

1

Note that due to the small amount of text each writer has produced in the essays used in this paper, it is almost impossible to distinguish between an error and a mistake and the term ‘mistake’ will therefore be used

(8)

to evaluate and analyse such mistakes or errors have been suggested, two of them being

Contrastive Analysis and Error Analysis. The latter was developed when the validity of

Contrastive Analysis was questioned (Skoog, 2006: 3 - 4). In order to better understand the analysis of mistakes and errors in the field of linguistics, the following section also contains explanations of concepts that are often used in the research in this field. One important term in this respect is interlanguage. The Interlanguage Hypothesis states that an L2 learner constructs sentences in which are found forms neither fully explainable by L1 nor by L2 rules (Ellis 1997: 33). Positive and negative transfer are two other important terms, describing how an L2 learner takes structures from her or his L1 and applies them to the L2 (Ellis 1997: 19). Sometimes, this leads to correct phrases being formed (positive transfer) but not always. An incorrect phrase being formed this way is an example of negative transfer. All of these terms are further discussed in this section, along with a list of rules on how to use capital letters in English and in Swedish and also what mistakes are commonly made by Swedish learners of English.

2.1 Contrastive Analysis and its problems

The essential focus of Contrastive Analysis is the idea that the L1 of the learner greatly influences the learning of the L2 (Odlin 1989: 15). To a contrastivist (i.e. a person dealing with Contrastive Analysis (James, 1980: 1)) comparing the mother tongue of a learner with his or her L2 will allow the teacher or researcher not only to explain why learners may experience difficulties when acquiring an L2, but also to predict future mistakes (Skoog 2006: 2). However, this also means that focus is put on the negative way in which L1 influences L2, disregarding the positive effect it might have. Furthermore, early focus in Contrastive Analysis, during the 1950s and 60s, was put on pronunciation and grammar and was aimed at eradicating influences from the L1 (Odlin 1989: 16). Even though most linguists eventually abandoned Contrastive Analysis in favour of a more positive approach to mistakes, it has exercised a major influence in the field and might therefore be used to better understand recent research (Skoog 2006: 3).

(9)

word know (conocer and saber) was in fact not a problem for Spanish learners of English.2 This meant that the differences between the two languages were not enough to predict or explain where difficulties might arise. In the end, the criticism of Contrastive Analysis lead to the development of other forms of mistake and error evaluation (Skoog 2006: 3).

2.2 Error Analysis

According to James (1998: 1), an error can be defined as “an unsuccessful bit of language” making the field of Error Analysis a study of “the process of determining the incidence, nature, causes and consequences of unsuccessful language” (James 1998: 1). As mentioned earlier, Error Analysis was one of the disciplines that followed the decreased application of Contrastive Analysis in linguistics. No longer solely focusing on the influence of the L1, error analysts also tried to come up with other, alternative, explanations as to why mistakes, or errors, were made (Odlin, 1989: 18). One example of this is the transfer of training, where the teacher’s questions are followed by a direct and sometimes faulty response as can be seen in this example from Odlin (1989:18):

1) Teacher: Am I your teacher? Learner: Yes, I am your teacher!

Another important difference between error analysists and contrastivists is that the former do not attempt to predict mistakes or errors, but rather explain why they happened and thereby learn more about how languages are actually learnt3 (Skoog 2006: 3-4). Furthermore, Skoog points out that Error Analysis is based on the idea that learning an L2 creates an independent language system in the learner, which lays the foundation for another important term in this field of linguistics.

2.3 The Interlanguage Hypothesis

The American linguist Larry Selinker coined the term ‘interlanguage’ when he realised that many mistakes L2 learners made could not be explained by solely looking at L1 or L2 structures and that the learners thus create unique language systems (Ellis 1997: 33).

2

In this specific example, however, this can easily be deduced from the fact that there is only one word to choose from in English, i.e. know.

3

(10)

Essentially, advocates of the Interlanguage Hypothesis argue that a completely new system of mental grammar is not created when learning a new language. Rather, it is initially based on the L1 system and, subsequently, changes into a system of its own. Furthermore, it is under the influence of input of new pieces of language from the L2 and thus constantly evolves and develops (or, as some research has claimed, new interlanguages are formed all the time). This creates the so-called language continuum where new rules for language are created, tweaked and deleted. However, interlanguage may also stop evolving, hindering the learner from progressing in the development of her or his understanding of the L2. This phenomenon is called language fossilisation and, once again according to Ellis (1997: 34) this happens, at some stage, to about 95 percent of all L2 learners. The remaining five percent are said to move on to having virtually the same mental grammar as native speakers.

The notion of overgeneralization, also a characteristic of young children’s L1 acquisition, is part of the interlanguage theory and basically means that a language learner uses a known L2 rule but uses it in a context where a native speaker would not (Skoog, 2006: 7). It may occur on the phonetic level (with the pronunciation of sounds at the end of words native speakers would keep silent), on the grammatical level (such as the overuse of the –ing form in English) and also on the lexical level, where a known word may be used to denominate items not usually included in the same category (a goose may be called a chicken and a teaspoon can be said to be a little spoon). Overgeneralization may also occur on a more pragmatic level. A native speaker of English might, for example, have problems with the distinction between tu and vous in French and similarly with du and Sie in German, all four examples being a less or more formal way of saying you (Skoog 2006: 7).

2.4 Positive/negative transfer

Transfer can be seen as an attempt by the learner to use the knowledge of previously learnt

(11)

influencing L2 learning, the others not yet being fully explored and defined. Transfer is, however, no longer seen as being a concept that can explain mistakes in L2 learning.

So how does transfer really help learners acquire or learn a new language? The answer depends on what kind of transfer is meant. Similarities in vocabulary help learners understand L2 texts faster, similarities in pronunciation help learners grasp the native speech more quickly and similarities in syntactic structures allow learners to master the grammar of the L2 more quickly. These are just a few random examples of how similarities aid L2 learners (Odlin, 1989: 36).

When it comes to negative transfer (or, as it is also called, interference), that is, transfer of language units from L1 which are not correct in the L2, Odlin (1989: 36-38) identifies four main categories of the transfer mentioned. Underproduction (also known as avoidance

strategy) means that the learner, being so distant from the target language, avoids or refrains

from using certain structures in the L2 so as to not make mistakes. This results in language production divergent from the language norm. However, the speech or text produced has comparatively few mistakes, mainly because the learner has avoided them. One example mentioned by Odlin is Schachter’s 1974 study, where Korean and Japanese students produced very few, or no, relative clauses compared to learners with native tongues closer to English. The opposite of underproduction is, of course, overproduction, where structures appear more often than in native speech. Drawing once again on the Japanese speakers, Odlin claims that they produce far too many simple sentences instead of relative clauses. Another example mentioned is English speakers of Hebrew, who apologise as often in their L2 as in their L1, thus deviating from Hebrew norms.

The most common case of negative transfer is the production error. Odlin identifies three main types of this transfer, namely substitutions, calques and alterations of structures. When a learner removes an L2 word or structure from a sentence and replaces it with a word from his or her L1, a substitution has been made. Odlin exemplifies this by presenting a sentence from a Swedish learner of English as a Second Language (ESL):

2) But sometimes I must go bort. (Odlin 1989: 37)

The learner has replaced the English word away with the Swedish equivalent bort.

(12)

grammatical but rather in the idiomatic sense. An example of this is the Spanish learner of ESL who preferred to use the form the porch of Carmen instead of the more idiomatically correct English form Carmen’s porch. The last type of production errors is alterations of structures which do not include a direct influence of the L1 but rather the L2 learners’ overcompensation in their effort to produce correct language. One example of this is

hypercorrection, where the language learner is aware of a phonetical transfer problem and

tries to compensate for it, but in doing so tries so hard that a new problem arises. This can be seen in Arabic speakers of English who sometimes confuse the phoneme /b/ with that of /p/ allowing for spelling mistakes such as *blaying4 and *hapit.

The final category of negative transfer is misinterpretation, where the learner due to cultural, syntactic or phonetic differences between the two languages misinterprets a message or constructs a piece of language that means something different than what was intended.

2.5 Common mistakes made by Swedish learners of English

As the English and Swedish languages both have developed from the Germanic branch of the Indo-European language tree they are very similar in construction (Swan and Smith 2001: 21) and close contacts between the languages over the past centuries have resulted in them being tied even closer to each other. However, there are a few differences which result in common mistakes for Swedish learners of English. This section will include a few of these occurrences where the two languages differ from each other. The categorizations are taken from Swan and Smith (2001: 21-36).

2.5.1 Spelling

Swedish uses the letter k in many instances where English uses c which results in mistakes such as *kannibal (Swan and Smith 2001: 26). Also, Swedish learners, once having discovered the phonological distinction between w and v, tend to overuse w since this sounds more ‘English’, resulting in mistakes such as *wolley ball.

2.5.2 Punctuation

Most Swedish compounds are written as one word which has as a result that learners have trouble with spacing in between words and overuse the hyphen such as in *fire-alarm and *frontdoor (Swan and Smith 2001: 26). Also belonging to this category is the use of

4

(13)

apostrophes which is far more common in English than in Swedish, resulting in mistakes when using the genitive such as *That is Susans cat.

2.5.3 Order of constituents

Swedish word order differs in some respects from that of English. One example of this is that other clause elements than the subject of a sentence can be fronted in Swedish (Swan and Smith 2001: 26-27). Normally, only adverbials are fronted in English in this way, something which seldom causes subject-verb inversion (with the exception of when negative or restrictive adverbials are present). An example of this is *That have I not seen. Naturally, Swedish ESL learners have trouble fronting adverbials as well.

2.5.4 Nouns – count/mass distinction and number

Differences occur between Swedish and English on several occasions when it comes to the distinction between countables and uncountables, singular and plural (Swan and Smith 2001: 27). This leads to mistakes such as: *informations, *many money, and *public expenditures.

2.5.5 Articles

There are several examples of where the English use of the definite article differs from the Swedish use, one of them being the use of the definite article before an uncountable or plural noun (Swan and Smith 2001: 28). The use of the indefinite article differs as well in some aspects, especially in idiomatic expressions. Both may be seen in the following examples: *a

man with hat and *People blame the society for everything.

2.5.6 Pronouns and determiners

Swedish lacks a distinction between who and which and also between any and some (Swan and Smith 2001: 30). Moreover, Swedish has a completely unspecific personal pronoun (‘man’) which corresponds to the English general use of you, one and they. Finally, English uses a possessive determiner in some cases where Swedish does not, resulting in mistakes such as the following: *He put the hand in the pocket.

2.5.7 Verbs

(14)

form is very limited in use, resulting in continuous mistakes regarding this (Swan and Smith 2001: 31).

2.5.8 Prepositions

The use of prepositions differs somewhat between Swedish and English; prepositions are for example sometimes followed by the equivalent of a that-clause (att-clause) in Swedish, resulting in the following type of mistake: *He convinced me of that he was innocent. Prepositions may also be followed by an infinitive and in passive sentences, the agent is introduced by the preposition av (of) which leads to mistakes like the following: *The bone

was eaten of the dog. Moreover, the very common preposition i is often exchanged for other

prepositions than in (an example of this is: *go in school).

As some common mistakes for Swedish ESL learners have now been presented, the study moves on to describing one mistake in more detail, namely that of where English and Swedish uses capital letters.

2.6 The use of capital letters in English

The English language uses capital letters in many instances where Swedish does not. In order to better understand the mistakes made by Swedish ESL learners, it is important to first know where the two languages differ in this respect. The following rules for the use of capital letters in English, and the examples, were taken from Ljung and Ohlander (1992: 322). The examples in English and their Swedish equivalents are shown in italics.

1. In English, the first personal pronoun I is spelled with a capital letter whereas the Swedish equivalent jag is not.

2. In English, all week days, months and holidays are capitalized, something which is not done in Swedish. It is also important to note that in English, all words in semantically related compounds are capitalized. Examples of this are Monday (måndag), January (januari) and

Christmas Day (juldagen).

(15)

Swedish does not (swahili). The same goes for a Frenchman (en fransos/fransman) and the

Swedes (svenskarna).

4. All words derived from proper nouns are capitalized in English but not in Swedish. This means that the word Londoner translates as londonbo without a capital letter and the same goes for Marxism (marxism).

5. Religions, political parties and their supporters are all spelled with an initial capital letter in English but not in Swedish, unless the word is a proper noun. This means that the Liberal

Party is spelled with capital letters in English, but not in Swedish (liberalerna). However, the

Swedish Liberal Party (Folkpartiet) is a proper noun and thus capitalised. Similarly, all words for religious groups and their followers begin with a capital letter in English (e.g. Islam) but not in Swedish (islam).

6. All words or abbreviations connoting family relations put before a name, such as Mr and

Aunt are capitalised in English but not in Swedish (herr, faster). This also applies to titles

such as General and Doctor.

7. When naming major buildings, institutions and organisations, English chooses to capitalise all words in such compounds whereas Swedish capitalises the first letter in the first word and, more often than not, creates a single word from the various subunits. This means that the

Ministry of Defence translates as Försvarsministeriet and the World Health Organisation

turns into Världshälsoorganisationen.

8. Some, but not all, established concepts in history, culture and society are spelled with initial capital letters in English, something which never happens in Swedish unless the concept is an initialism, for example fotbolls-VM (World Cup of Football). The Stone Age however, being an established concept in history, takes the capital letter whereas its Swedish translation (stenåldern) does not.

(16)

Overall, smaller words like articles, monosyllabic prepositions and the word and usually have lower case letters in English and this applies to compounds as well.

10. In English compound words, unlike Swedish ones, all the elements have initial capital letters. The United Nations is for example spelled Förenta nationerna in Swedish.

These differences between the use of capital letters in English and Swedish are henceforth referred to as rules number 1-10. Now having established the theoretical background of the study, I move on to describing how the study was performed.

3. Material and method

This section describes the data used in this study. It also addresses what problems arose when dealing with the analysis of the data and how they were dealt with. Finally, a discussion is made on how the material was analysed.

3.1 The SWICLE

(17)

communicated with the parent in Swedish, I considered this to have a very small impact on the language of the child (and thus on the essay written by him or her), making it unimportant for the purpose of this study. Furthermore, all essays in the study were written without the use of a dictionary of any sort. This (together with the fact that the chosen essays were timed) ensures the spontaneity of the mistakes produced. Thus, 50 essays were chosen for the study (for a list of what essays were used, see Appendix B). These included 26728 words in total, with an average of 534.5 words per essay. The highest number of words in an essay was 895 and the lowest was 371. With these numbers at hand, it was time to see what mistakes had been made by the writers.

3.2 Correcting the essays

(18)

4. Results

This section includes presentations of all the data gathered in the study. Firstly, all mistakes found in the essays that were not connected to the use of capital letters are presented since that is also part of the study. Secondly, a more in-depth presentation of the mistakes when using capital letters is undertaken. Finally, the issue of how Swedish school books have presented and dealt with the problem of capital letters is addressed. Note that there might be several mistakes in every example. However, only the ones with relevance to the paragraph in question have been underlined.

4.1 Mistakes in the text

A total of 324 mistakes, including those where the writer failed or forgot to use capital letters in a correct way, were found in the 50 essays, which means that an average of 6.48 mistakes were made per essay. The actual number of mistakes per essay ranged from nil to 22. In the following Sections (4.1.1 – 4.1.10), a brief summary of these mistakes will be presented, not including those mistakes that had to do with the use of capital letters. The division into various categories follows that of Swan and Smith (2001: 26-34, see Section 2.5) but another category, Choice of words and idioms, not included in the book has been added in order to account for the mistakes encountered in the data. A table of all mistakes excluding those of capital letters is seen in Appendix A.

4.1.1 Spelling

The most common mistakes made were by far various spelling mistakes which accounted for 98 instances, i.e. about a third of the total number of mistakes made at an average of 1.96 mistakes per essay (for details, view Table 1 in Appendix A). Some of these may have been due to language interference whereas others might simply be careless mistakes or slips. Almost all of the essays had at least one spelling mistake and two of them are featured below.

3) It seems, then, that one key to a solution is dependance.

4) Quality mobil phones are not sold to customers under a certain income.

(19)

4.1.2 Choice of words and idioms

The second largest category of mistakes, not counting capital letters which will be presented in more detail below, was that of choice of words and idioms. This category loosely includes all sentences where a word was used which did not properly fit in the context and should have been exchanged for another. Moreover, the category includes various sentences where set expressions and idioms were used which either do not exist in the English language (i.e. are directly translated from Swedish) or are phrased somewhat differently than in the actual sentence. A total of 55 of these mistakes were discovered which meant that on average, about 1.1 of this type of faults was made per essay (for details, see Table 1 in Appendix A). The mistakes accounted for about 17 per cent of the total number of mistakes. One example of each of these three types is shown below.

5) Zen-Buddhism, ordinary Buddhism, Hinduism, Taoism and Hare Krishna are examples of “old” religious movements that have increased its number of members.

6) To blame it all on money might perhaps be to simple, there are always more than one side on a coin.

7) The golden middle is what we should aim for and that means hard work for both Swedes and immigrants.

In example five, the sentence might be more correct if the word ordinary were replaced by the word common. The reason for the mistake the writer has made might be that both words have the same translation in Swedish: ‘vanlig’. (6) contains an English idiom usually written as ‘There is always more than one side to a coin’ where the preposition on is replaced by to. (7) includes a translation of the Swedish idiom ‘den gyllene medelvägen’ which directly translates as the golden middle road but would be more suitable if replaced by strike the

golden mean.

4.1.3 Verb forms and tenses

(20)

were made per essay accounting for 12 per cent of the total number of mistakes (for details, view Table 1 in Appendix A). Examples of these three categories are shown below.

8) Even though the number of immigrants being allowed into the country have been reduced, there are still many waiting for treatment of their applications.

9) It has, however, during the years since the fifties become considered, somewhat, of a problem, since the immigration has increased and the job situation has decreased.

10) It leads to immigrant children, with a chance to become bilingual, lose their first language and through that, also lose an important link to their culture.

(8) features a concord mistake where the word number (singular) is connected to the verb

have, which should be connected to a noun in the plural. (9) includes the verb decreased

which should perhaps be replaced by the verb worsened. Finally, (10) contains two occasions of the word lose and the sentence might look better if these were replaced by the present participle losing.

4.1.4 Articles

This category of mistakes includes cases where the wrong article has been used. In the data chosen for the study, 20 such mistakes were made and most of them were faulty uses of the definite article. The results indicate that the distinction between the indefinite articles a and

an is of little trouble for the Swedish university students, whereas the definite article is more

of a problem. An average of 0.4 mistakes were made per essay, i.e. six per cent of all mistakes were of this type (for details, view Table 1 in Appendix A). The mistakes included both situations where the definite article should have been used but was not and where it was used but should not have been. One example of each is shown below.

11) The inbalance in the powerstructure in the areas where the Europeans and Americans (people from the US) have been involved has produced many large conflicts like the civilwar in Somalia and Lebanon and the coup d’etat in Chile and Nicaragua.

12) The new equality in the Western society

(21)

includes the definite article, the word society is in English an abstract uncountable noun used generically and therefore does not take the definite article.

4.1.5 Prepositions

This category includes all mistakes where the wrong preposition was used in a sentence. A total of 16 of these mistakes were discovered in the study which means that an average of 0.32 mistakes of this type were made in each essay accounting for five per cent of the total number of mistakes (for details, view Table 1 in Appendix A). One example of a preposition mistake is shown below.

13) Another example is the warning of too much sun exposure that has made people think behind the immediate consequence of cancer to the more overall use of chemicals that makes the ozon layer become thinner, which in its turn makes the sun more burning.

In this context, the preposition behind should perhaps be replaced by beyond.

4.1.6 Punctuation

Into the category Punctuation, Swan and Smith (2001: 26) not only put full stops and semicolons, but also the use of the hyphen and of spacing in words that are usually compounds in Swedish. Additionally, the use of apostrophes is also included in the category. A total of 15 mistakes can be found in this category. This means that about 4.6 percent of all mistakes where punctuation mistakes (for details, view Table 1 in Appendix A). One example of each of the three types can be seen below.

14) Eventhough my grandfather did not have the opportunity to go to school he later started his own business working with the timbre market.

15) As I see it, to-day’s young people and those of the next generation will not necessarily suffer from the prevailing state of high unemployment.

16) Undoubtedly as I saw what three generations of unemployment does to people, heared the teachers comments on their pupils in ways such as […].

(22)

forgotten the genitive s on the word teachers (teachers’). Alternatively, the writer of (16) might have put an extra s on the word comment, changing it from a verb to a noun.

4.1.7 Nouns, count/mass distinction and number

This category included all those occasions where mistakes were made between uncountable and countable nouns as Swedish and English differ somewhat between the two. Moreover, all instances where a faulty singular or plural construction was used were also included in this category. Moreover, one occasion of the faulty use of a numeral was included. A total of six of these mistakes were made and two examples are featured below.

17) This essay deals solely with possible reasons for connections in our modern society in the last decade of the 2000th century.

18) The meals are served at the same times, made by the same persons using the same ingrediences as last christmas.

In (17), the writer has both the expression 20th century and the year 2000 at once. In (18), the writer has taken the noun person, which in English exists only in the singular, and turned it into a plural. However, the English plural of person is people.

4.1.8 Order of constituents

Only five mistakes where the writer had made a mistake of word order were found in the essays, which reveal that this is not something Swedish university students have much trouble with. However, one example of a faulty order of constituents is shown below.

19) How can we make Christmas of 1997 the most memorable Christmas we ever have celebrated?

(23)

4.1.9 Pronouns

Five mistakes where the writer had made mistakes of pronouns were found in the essays, including three instances where the Swedish word som had been translated as which instead of

that or who. Two examples of faulty use of pronouns are shown below.

20) Zen-Buddhism, ordinary Buddhism, Hinduism, Taoism and Hare Krishna are examples of “old” religious movements that have increased its number of members.

21) Sects which strip their members of money are terrifyingly common nowadays.

In (20), the anaphora its should be replaced by their as the antecedent ‘old religious movements’ is a plural noun phrase. In (21), which should be replaced by that since no selection of sects has been presented.

4.1.10 Words missing and question marks

In this final category, seven mistakes can be found that cannot be placed in any other category. These include occurrences where words are simply missing, maybe because of slips by the writer, or where the writer has simply put a question mark instead of writing something.

22) […] but since capitalism is considering nothing but profit, environmental have been set aside.

After the word environmental there is a word missing. Suggested words to place here would be issues or concerns.

4.2 Misuse of capital letters

(24)

study since a writer might sometimes spell a word correctly and sometimes not in the same text. Details of all capital letter mistakes are available in Table 2 in Appendix A.

4.2.1 Overuse of capital letters

The largest category of capital letter mistakes was in fact mistakes where the writer overused capital lettering. 18 such mistakes were found which is about 30 per cent of the total number of capital letter mistakes (see Table 2 in Appendix A). Some cases of such overuse may be seen in the following examples.

23) It might sound harsh to blame the Industrial revolution and therefore also the progress of our society […]

24) Even if the Olympics would mean 3000 new jobs or more, this is only a temporary solution of the unemployment problem, and only a drop in the unemployment Ocean. 25) There is a proposal from the Government of introducing the 35-hour work week within the next ten years.

(23) shows a misinterpretation of Grammar rule no. 8 (concepts in history, society and culture, see Section 2.6), the author understood the industrial revolution to be an established historical concept in need of capital lettering. (24) can be considered faulty in the context of writing an argumentative essay albeit it is ‘poetic’. (25) shows a misinterpretation of Grammar rule no. 7 as the word government is not treated as an established institution in English.

4.2.2 Underuse of capital letters in weekdays, months and holidays

The second largest group of mistakes of capital lettering was the overuse of capital lettering when it came to weekdays, holidays and months. Primarily, the essays displayed faults when it came to holidays. Of the 17 mistakes of this type found, only two pertained to something other than holidays.

26) That is what christmas is all about.

27) (In Sweden, this is to be done at three different occasions, 1:st of Advent, Lucia and Christmas eve).

(25)

Examples (26) – (28) display mistakes that can be related to the learners not fully being aware of the English grammar rule of capitalizing holidays. (26) shows a clean break of the rule whereas the writer of (27) probably treats 24 December as a Swedish compound thereby not capitalizing the word ‘eve’. The writer of (28) correctly uses a capital ‘C’ at the beginning of

Christmas earlier in the text but alternates between the two forms throughout the essay.

4.2.3 Underuse of capital letters in nationals, languages and nationalities

The third largest group of mistakes when it comes to capital lettering pertained to words connected to nationalities and languages. 16 such mistakes were found accounting for around 26 per cent of the total number of mistakes. Most common were mistakes where the writer used lower-case letters in nationalities (29-30).

29) Where only 10-15 years ago, the term “melting pot” was used, in an american society that still believed that assimilation was the right thing, they have now realized that integration works.

30) My experiences, as a teacher in Swedish as a foreign language, are that those who live in isolated groups with no swedish friends or colleagues never seem to learn Swedish very well […].

Interestingly, the writer of (30) has used a lower case letter when discussing a nationality (swedish) but capital letters when writing about the Swedish language. This may suggest that the writer has not internalised the rule completely, or it might just be a ‘slip of the pen’.

4.2.4 Capital I in first person singular, subject form

Only two cases appeared where the writer had missed capitalizing ‘I’, contrary to English grammar rules.

31) How stupid! i get it anyway.

32) One mistake which i often notice in political debates on all levels is that people tend to mix up the term “equal” with “identical” or “the same”.

(26)

hand occurs in the middle of a sentence and may be due to language interference, even though the writer correctly uses capital I in other sentences in the text.

4.2.5 Religions, political parties and their members

Only one example of misuse of capital letters with respect to the group ‘religions, political parties and their members’ appeared in the study:

33) Also, they would all have to become protestants and abandon their religions, which is an impossible thought, considering the fact that religion in many cases is the reason they left their own countries originally.

As Protestants are members of a religious group, the word takes a capital letter in English. The Swedish equivalent protestanter, however, is spelled with a lower case letter, something which might be a reason for the mistake.

4.2.6 Words unaccounted for

This last group included two recorded words that do not belong in any of the groups previously established. Two cases of the word ‘Internet’ spelled without a capital letter were found, as well as five instances where the word ‘Earth’, as referring to the planet, was spelled with lower case letters only. Both of these words are spelled with capital letters in Standard English but, most of the time, with lower case letters in Swedish.

34) One thing which is important to remember is that the consuming society in which we are living is not a threat to the earth.

35) […] and you need internet to be among the first to spot the job of your dream before someone else does.

As all occurrences of the various mistakes have now been presented, I will move on to describing how a few school books deal with the use of capital letters.

4.3 Capital letters in school books

(27)

school system works with the problem. A random sample of twelve school book series of English published in Sweden for lower and upper secondary school was selected and read through. Most book series had one or more exercises where nationalities were practised, naturally including capital lettering. However, only three of the books series dealt with capital letters to a higher degree than that. In Action in Progress (Studentlitteratur: 2007) Hedencrona et al. include a paragraph on capital letters in the grammar section at the back of the book. In four sentences the book explains that capital letters are used when writing months, holidays and weekdays, as well as when talking about nationalities, languages and countries. The book has a total of 248 pages and is part of a series of several books aimed at upper secondary school learners.

Tholin et al. also have a paragraph on the matter, presented in the middle of the book A

Piece of Cake Workbook 3 (Almqvist & Wiksell: 1996) aimed at grade nine pupils in lower

secondary school. Under the headline ‘When to use capital letters’, four types of capital letter differences are put forth: in the word I, in days of the week, in months and in nationalities and languages.

The most extensive presentation, in the study, of when to use capital letters is found in

Engelska 2 Övningsbok by Bergman et al. (Akademiförlaget: 1992) which is aimed at the

second course of English in upper secondary school. Moreover, an exercise on the matter is added after the explanation which is placed in a grammar chapter at the back. The book exemplifies six instances where capital letters occur in English but not in Swedish: nationalities (and words thereof, such as languages); months, days and holidays; titles followed by a name; words connoting famous things, people or institutions; important words in titles of books, titles and headlines, etc., and finally both words in street names, etc.

Some of the books/series that were shown not to have a section on capital letters were:

Project X (Casselryd et al.: 1998, upper secondary school), Wings (Glover et al. 2003, lower

(28)

5. Summary of the results

The aim of this investigation was threefold. The first aim was to identify what mistakes caused by possible L1 interference were made by Swedish learners of English who had left the Swedish school system. Secondly, the writer of the essay wanted to examine one of these mistakes, namely the use of capital letters, in more detail in order to search for L1 interference patterns. Thirdly, a closer look was to be made at a few school books in order to see how the problems caused by differences between Swedish and English use of capital letters have been dealt with and how this particular area is presented to Swedish learners of English. The study was conducted by looking at fifty student essays from different Swedish universities, correcting them and subsequently analysing and classifying the mistakes. Also, a number of school books teaching English to Swedish learners were reviewed.

In order to reach the threefold aim presented in the previous paragraph, four research questions were produced. Firstly, the writer of this essay wanted to look at what mistakes were made in university essays by Swedish ESL learners and secondly, if the misuse of capital letters was a common mistake made by the same learners. The third question dealt with in what contexts the capital letter mistakes were made and the final question was about how the differences between Swedish and English use of capital letters had been presented in Swedish school books. This paragraph summarizes the results and briefly tries to answer the questions.

The answer to the first research question (i.e. which were the most common mistakes for Swedish ESL learners in the study to make) was, not surprisingly, spelling mistakes with almost one hundred attested occurrences, followed most prominently by capital letter mistakes, choice of word mistakes and verb mistakes. Article, preposition and punctuation mistakes were also common, followed by the slightly less common mistakes on nouns, word order and pronouns. For more detail, see Section 4.1 or the overview of all the mistakes in Table 1 in Appendix A.

(29)

Many of the school books examined for the purpose of finding out how the issue of capital letters is presented did not address the topic at all. The majority of those that did went over the issue briefly. Interestingly, the only book that both explained the different occurrences and allowed the learners training on the topic with an exercise (Engelska 2, Bergman et al. 1992), was the only book that the essay writers in the study could have had as a school book themselves, having been published at the beginning of the 1990s.

6. Discussion

In the essays, clear influences (or transfer) of L1 may be seen in examples (4), (5), (7), (19) and possibly also (16). Additionally, almost all examples of underuse of capital letters are examples thereof. This may indicate that most writers were, when producing the essays, within a language continuum (see Section 2.3) and at the same time, had mental grammars with parts from both their L1 and L2. This seems to prove that there is, in fact, an

interlanguage based on the L1, evolving into an L2 and that this interlanguage is completely

detached from both the L1 and the L2. Since the study was synchronic and not diachronic, however, it is not possible to prove whether the writers were still evolving in the continuum or if their L2 had fossilised. Some writers made no mistakes and their texts were close to being, or even were, on the same level as that of a native speaker of English. However, without a larger sample of texts produced over time, it cannot be proved that they were within the five percent Ellis (1997: 34, see Section 2.3) established as the group that could move on to having the same mental grammar as native speakers.

There were several examples indicating overgeneralization (see Section 2.4), most prominently the overuse of capital letters. Some writers used capital letters where they were not needed which was contrary to what could be expected, but this might simply be because they had overextended the use of a rule known to them.

I was not able to find any clear examples of either under- or overproduction (see Section 2.4) of pieces of language in the essays. In order to quantatively prove the occurrence of such structures I would need a considerably larger corpus together with a corresponding corpus with text written by native speakers of English, something which the scope of this study did not allow.

(30)

with other forms. Another possibility is that they simply put a question mark in place of the word they could not retrieve (see Section 4.1.11). There was at least one clear example of

calque in the study (example (7)).

The mistakes in the essays seem to correspond well to the typical mistakes performed by Swedish learners of English presented in Swan & Smith (2001: 21-36, see Section 2.5). Most mistakes were represented, be that on a slightly higher level of competence than that described in the book. The more specific mistakes in Ljung & Ohlander (1992: 322, see Section 2.6), dealing with capital letters, where less often represented, however. Half of the categories of differences between Swedish and English capital letter usage were not represented in my data. This might also, of course, be due to the possibility that the learners had acquired some rules but not others. The investigation was only focused on what the learners had done wrong, not what they hade done correctly. Some mistakes, such as that of holidays, for example, were represented to a higher degree than I had expected however and the capital letter mistake was the second most common mistake, following spelling mistakes. The fact that the capital letter mistake was so common in comparison to other mistakes leads us to the final question posed in this study, namely how the use of capital letters is presented in school books. The school books did not reveal many examples of paragraphs or sections dealing with capital letters and the differences of use of them between English and Swedish. Considering the number of mistakes that might not occur with a more thorough presentation of the problem in school books, more emphasis placed on the use of capital letters in Swedish school books for ESL learners could be advocated. That is, if the researcher is only looking at reducing the number of mistakes, no matter their type. On the other hand, the reason other mistakes are so infrequent in comparison might be that they have been taught thoroughly and that the use of capital letters might be considered a relatively small problem as compared to, for example, verb inflection or learning what nouns are uncountable in English. It might just be that capital lettering is a luxury problem which may be an issue for strong learners at a late stage of learning, but is hardly a problem everyone needs to deal with since it does not hamper the understanding of a text. This notion is supported by the fact that sections on capital letters were more common and more extensive in books for upper secondary school than in those for lower secondary school.

(31)

interesting and important for future teachers in Sweden, but such a study would also require a different theoretical framework and many more essays and much more time that was granted for this study. There is, however, always the possibility of complementing the study later. Such a study could, for example, include people of other ethnical (i.e. language) backgrounds and other features such as gender, age or even time spent abroad in English speaking countries. A researcher could also complement the didactic angle of the essay by studying how the results may best be used when teaching.

It is important to remember that the essays selected for the study were produced by writers who had been accepted to various universities which possibly meant they had better grades than their schoolmates who were not accepted or did not apply. Moreover, they had studied English at university level for more than a year. The results might have possibly differed if I had looked at essays written by a random selection of people who had just left upper secondary school rather than the ones that went to study at university level. As no such corpus is known to me, however, the way the study was conducted seemed most efficient for the purpose of this study. Even though the frequency of mistakes, or even the relation between the mistakes displayed in percentages, might not be the same as in a study with a more random sample of writers who had left upper secondary school, some indications of common mistakes made by Swedish learners of English may be seen in the results.

The frequency of mistakes involving capital letters might have been affected by the choice of material. Since the students did not get to choose the topics of the essays freely, but rather were given a few topics to choose from, some occurrences of capital letters might be more frequent in the essays chosen than in a random sample of language. One such example is that many essays were about how to celebrate Christmas, thus naturally increasing the frequency of the word Christmas opening the door to more mistakes, showing that the writer had not acquired the rule that the names of all holidays start with a capital letter in English. It is also possible that the language of the essays was influenced by the more informal language expressed on the Internet and other forms of modern communication, which the writers may have come into contact with via text messages, e-mails, chat programmes and the like.

(32)

The form of the writing also contributed to making the material more representative of the actual knowledge of the writers. Since all essays in the study were written without the use of a monolingual or bilingual dictionary or grammar, the spontaneity of the results was ensured. Similarly, no computers with spelling and grammar checkers were used. What the students produced were essays representative of their knowledge, no more, no less.

(33)

List of references

5

Bergman, Lars, Ian Hinchliffe & Ingvar Sjöholm. 1992. Engelska. Göteborg: Akademiförlaget.

Cambridge advanced learner’s dictionary. 2005. Cambridge: Cambridge University Press

Casselryd, Per-Ola, Astrid Jägfeldt & Kjell Ljungberg. 1998. Project X. Stockholm: Natur och Kultur.

Cutler, Tony & Dennis Miller. 1994. Get it together. Malmö: Gleerups.

Ellis, Rod. 1997. Second Language Acquisition. Oxford: Oxford University Press.

Glover, Mary, Richard Glover, Bo Hedberg & Per Malmberg. 1997, 2003. Wings. Stockholm: Natur och Kultur.

Hedencrona, Eva, Karin Smed-Gerdin & Peter Watcyn-Jones. 2007. Action in

Progress. Lund: Studentlitteratur.

James, Carl. 1980. Contrastive Analysis. Harlow: Longman Group Limited.

James, Carl. 1998. Errors in Language Learning and Use – Exploring Error Analysis. Harlow: Addison Wesley Longman Limited.

Ljung, Magnus & Sölve Ohlander. 1992. Gleerups Engelska Grammatik. Malmö: Gleerups. Lund University. 2007. The Swedish Component of the International Corpus of Learner

English. Lund. http://www.englund.lu.se/corpus/corpus/swicle.html. Accessed 19 March

2008.

Odlin, Terence. 1989. Language Transfer – Cross-linguistic Influence in Language Learning. Cambridge: Cambridge University Press.

Skoog, Petra. 2006. Native Language Interference – a Study of Interference Patterns in

Swedish Students’ English Writing. Karlstad: Karlstads Universitet. Found at:

http://www.diva-portal.org/kau/abstract.xsql?dbid=418 Accessed 19 March 2008.

Tholin, Jörgen, Rigmor Eriksson & Moira Linnarud. 1996. A piece of cake. Stockholm: Almqvist & Wiksell.

5

(34)

Appendix A – Tables of mistakes

Table 1 – All mistakes in the essays, excluding capital letter mistakes

Table 2 – All capital letter mistakes in the essays Overuse of capital letters Weekdays, months, holidays Nationals, languages, nationalities First person singular Religions, political parties and members Words unaccounted for Sum 18 17 16 2 1 7 % of total no. of capital letter mistakes 30 28 26 3 1.6 11 Spelling Choice of words and idioms Verb forms/ tenses

(35)

Appendix B – SWICLE essays used

The following essays from the SWICLE were chosen for the essay and all examples in the thesis are taken from them.

Capital mistakes The analysis of mistakes in the written production of advanced Swedish ESL learners at university level with focus on the use of capital letters