|
|
CONTENTS
Chapter One: Pidgins and Creoles
1.1. Definitions
1.1.1 Lingue Franche, Pidgins and Creoles
1.1.2 Nigerian Pidgin: A Problem of Classification
1.2. The Inception of a New Academic Field
1.2.1 The Key Debate: Theories about the Origins of Pidgins
1.2.2 Other Topics of Discussion
Chapter Two: Research on the Internet
2.1 Who Uses the Internet?
2.2 Search Tools
2.2.1 Two Categories of Search Tools
2.2.1.1 Subject Directories and Other Internet Guides
2.2.1.2 Search Engines
2.2.2 Useful Features of Search Tools
2.2.3 Meta-Search Engines
2.2.4 On the Search Tools Covered in this Study
2.2.5 How Search Tools Deal with Questions
2.3 Critical Notes
2.3.1 General Reflections
2.3.2 Source Verification
Chapter Three: Search Results for Nigerian Pidgin - A Case Study
3.1 Possible Approaches
3.2 Analyzing and Comparing Search Results
Results from Individual Search Tools
3.2.1.1 Phrase Search “Nigerian Pidgin“
3.2.1.2 Phrase Search “Nigerian Pidgin English“
3.2.1.3 Advanced Search
3.2.2 Results from New Search Engines
3.2.3 Results from Meta-Search Engines
3.3 Other Starting Points
3.3.1 Starting from a Quality Link Site
3.3.2 Online Discussion Groups
3.3.2.1 Newsgroups
3.3.2.2 Mailing Lists
3.3.2.3 Re: Online Publishing
3.3.3 Contacting Experts
Bibliography
When carrying out research, people traditionally read reference books, conduct interviews, attend conferences and so forth. With the advent of the Internet, another important research source now opens up before us. One of the great features of the Internet is the amount of information it contains. You can look for a job in far-off places. You can look up a phone number for a restaurant in Paris. You can even read the “Complete Works of Shakespeare“ (the-tech.mit.edu/Shakespeare/works.html). But all this newly available information comes at a high price: confusion. The interconnected and dynamic nature of the Internet prevents it from being organized like a traditional medium, such as a book or a newspaper.
Due to the lack of organization to the avalanche of information that a typical search returns, finding relevant information reminds us of looking for a needle in a haystack and can be quite frustrating. It is thus very important to familiarize yourself with the various online research sources. Some people have never heard of meta-search engines, truncation, URL guessing, Usenet or FAQs. Only if we know the features of the Internet, we can benefit in full from it. While in some fields the Internet is already an established research tool, this is still different with relatively exotic disciplines such as creolistics. It is the objective of this study to examine what kind of information can be found on the Internet with respect to pidgin and creole studies in general and Nigerian Pidgin in particular. The paper is also intended to draw attention to how this material should be treated. Many Internet users take all the information on the Web at face value, when especially critical evaluation would be more appropriate. Are the data supplied accurate, complete, up to date? Last but not least, for those who are not familiar with pidgin and creole languages, this paper provides a primary overview about this fascinating subfield of linguistics.
The material-gathering took place between March and April 1999. Please note that since the Internet is a particularly dynamic and fast-changing medium, many Web sites that are mentioned in this study will have been updated, otherwise altered, moved to a new location, or vanished from the Web by the time you read this paper.
1.1 Definitions
1.1.1 Lingue Franche, Pidgins and Creoles
Pidgin and creole studies is a relatively new-established and relatively small research field. The material on this subject area is easier to survey than the millions of Web pages on certain other topics. It is therefore very suitable for this research paper. But above all, pidgins and creoles are a most fascinating phenomenon, the value and potential of which were largely ignored even by linguists before the 1950s. This chapter is designed to provide the novice with a general idea of the concepts of pidgins and creoles. In this connection, you will often come across the term “lingua franca“. Originally, this term was applied to the Proven(al-based trade jargon known also as Sabir, which was used by the Crusaders and merchants who visited eastern Mediterranean ports in the Middle Ages. Nowadays, it serves to denote any language that is used as a medium of communication among people who have no other language in common.
One example is the status of English in India, a multilingual society that naturally requires a common means of communication. So whenever two Indians with different linguistic backgrounds come together, chances are high that they make use of English, although it is native to neither side. In short, English is serving here as a lingua franca. In 1950, the Indian government declared Hindi as the “official language of the Union“ in an attempt to replace the language of the colonial power altogether. However, English, spoken by only four percent of the population, has retained the status of an associate official language (this status was prolonged by an amendment in 1967 for an indefinite time), which clearly shows that English enjoys a prestige out of proportion to the actual number of its speakers, thanks mainly to its function as a “world-language“ and “window on the world“ (cf. Hansen 215-220).
Now imagine two or more people using a language in a variety whose grammatical structure and vocabulary are sharply reduced, for instance grammatical gender is lost and the infinitive is used instead of an inflected form of the verb. This process is known as pidginization. This, incidentally, is exactly what happened during the development from Old English into modern English. Only an estimated 85 percent of the Old English vocabulary survived to the present day, due to the intensive contacts of the English language with other languages. Modern English exhibits a massive mixture of Germanic and Romance languages, often with a synonym from each group, for instance freedom and liberty. This is a widely recognized pidginization process, though it would be inaccurate to say that English is a pidgin. Like the other “ordinary“ languages, English emerged gradually (in the stammbaum sense), whereas pidgins are thought to have come into existence at some point at time. William Robert O’Donnell and Loreto Todd point out though that Middle English might have been a creole for the Normans who adapted it as a mother tongue (O’Donnell/Todd 48).
Ronald Wardhaugh argues that “The process of pidginization probably requires a situation that involves at least three languages, one of which is “dominant“ over the others. If only two languages are involved, there is likely to be a direct struggle for dominance, as between English and French in England after 1066 [...]“ (Wardhaugh 57). Interestingly enough, in this well-known case, it was the socially “inferior“ language that won the struggle, even if only after a long time of co-existence.
Many authorities claim that the word “pidgin“ is a Chinese corruption of “business“. Indeed, as we will learn soon, pidgin languages have very often arisen in short contacts between people who desired to trade with each other. Although this traditional etymology has been challenged by linguistics including Robert A. Hall, Jr., who suggests that the name derived from a South American Indian tribe (cf. Britannica, Volume 22, page 803), experts agree that the term was first applied to Chinese Pidgin English, and later to any language of similar type.
A very small vocabulary is one of the striking properties of pidgins, yet the size of the lexicon varies: Melanesian Pidgin has 2,000 words, whereas Chinese Pidgin English has only 700. Usually, approximately 90 percent of this vocabulary comes from the source languages, often English, French or Portuguese, but not necessarily a European language. Pidgins are therefore not mixed languages, as is often assumed. Since vocabulary is so much restricted, each word necessarily has a greater range of meaning than their counterparts of the respective donor-language. For instance, Nigerian Pidgin knows only one word for finger and toe, since the latter can be regarded as the leg’s finger.
Words such as “prick“ and “kont“ are likely to strike a speaker of Standard English, but are not in the least considered slang or bad language in Nigeria. These are in fact the appropriate terms in Nigerian Pidgin (cf. Faraclas 286). There is a simple explanation to this phenomenon: these must have been the words that the Africans learned from the white settlers:
For the urban poor, soldiers, and many others who made up the early settlers, slang was an important part of daily speech. Such words often became a part of the creole and lost their European connotations in the process: if a creole’s only word for “urine“ was piss, this word became as appropriate as urine in any domain, shedding the vulgarity of its etymon, e. g. Krio CE switpis “diabetes,“ pisbag “bladder,“ pisol “urethra.“ (Holm 78)
Unlike a pidgin, which is suitable only for limited conversation, because it functions only as an auxiliary contact language, a creole is the native language of most of its people. It goes without saying that both vocabulary and syntax of creole languages meet all communication needs important to their speakers. A creole extends its vocabulary by borrowing words from the respective donor language, from the local language(s), or by creating new words. Fine examples of neologisms and phraseologies include “decampee“ (person who switches to a different party, “maverick“), “to have long legs“ (to have influence) and “to smell pepper“ (to face a rough time). Given these observations, it is understandable that creolists want pidgins which have developed into creoles to be treated as any other “real“ language. In this context, it is important to point out that many creole speakers consider themselves to be speakers of the lexifier language, because they feel a sense of inferiority about their languages.
To sum up, pidgins spring from the initial, nonintimate contacts between speakers of different languages, when quick comprehension is more highly valued than grammatical correctness or fine shades of meaning. As contacts grow closer, normally one group learns the other’s language in more detail, which accounts for the fact that pidgins are usually short-lived.
An interesting exception to the rule is the case of Chinese Pidgin English. It survived for three centuries, primarily because each side wanted to keep the other at arm’s length. Hall describes the peculiar situation as follows:
The English regarded the language of the “heathen Chinee“ as beyond any possibility of learning, and began to pidginize their own language for the benefit of the Chinese. The latter held the English, like all “foreign devils,“ in extremely low esteem, and would not stoop to learning the foreigners’ language in its full form. They were willing, though, to learn what they perfectly knew to be an “imperfect“ variety of English or of some other Western tongue, and considered that this was abasing themselves less than learning “real“ English. (Hall 8)
Pidgins persist, however, where a dominant group regards another as childlike or capable only of a simplified version of the “superior“ language, as in the relations between Europeans and American Indians, West Africans, or South Sea natives. Only when a pidgin gradually becomes the native language of a speech community, it is to survive as a creole. This phenomenon has been dubbed creolization. As a matter of fact, many pidgins became creoles in this way, for instance the French-based creoles of Louisiana (known as Cajun), Haiti and the Lesser Antilles, or the Papiamento of Cura(ao (based on Spanish and Portuguese), to name but a few of them.
Pidgins and creoles are spoken all over the world. Creolists are interested in Greenlandic Inuit pidgins and the now extinct Russian-Norwegian trader jargon of the Arctic Ocean, as well as pidginized Bantu languages spoken in central Africa and the creoles of Mauritius and Hawaii. The most well-documented restructured languages are those which arose as a result of European colonial expansion in the past few centuries, notably on islands in the Caribbean, in the Indian Ocean and in the Pacific. The two most widely spoken creoles in terms of first language speakers, Haitian Creole (Haiti) and Tok Pisin (Papua New Guinea), draw most of their lexicon from French and English respectively, but note that their grammatical structures are quite unlike their lexifier languages.
1.1.2. Nigerian Pidgin: A Problem of Classification
Creolists justifiably point out that the distinction between pidgins and creoles is not always easy to draw. This is due to the fact that some extended pidgins are beginning to acquire native speakers. This is happening for instance with Tok Pisin, Krio (in Sierra Leone) and Nigerian Pidgin, to name but the three best-known cases. In particular this phenomenon has tended to occur in urban environments, where speakers from different ethnic groups have daily contact with each other. What develops then is a somewhat distinct language of that town. The children of mixed marriages frequently grow up speaking the pidgin as their native language. In the following, I would like to illustrate the things said with an example. For this purpose I chose a common situation in Nigeria, which is the most populous country on the African continent, comprising a large number of different ethnic groups who speak 400 or so different languages. For a Yoruba market woman whose use of Nigerian Pidgin is strictly restricted to business transaction, the language is a pidgin in the true sense of the word. However, apart from the estimated 40 million people who speak it as a second language, more than one million people are estimated to speak it as their first language (These are estimates made by Nicholas G. Faraclas. Later we will see that there are no official figures). For those people, Nigerian Pidgin has become a creolized speech form. This is especially true for children who grow up speaking Nigerian pidgin, because their parents come from different speech communities, and because their playmates from the neighborhood again speak another language. As a matter of fact,
[...] intermarriage, trading and travel have brought Nigerians who speak different languages into close contact with one another for thousands of years. Bilingualism and multilingualism have always been the norm rather than the exception in most parts of Nigeria. For these reasons, it is very likely that pidginized versions of Nigerian languages were widely used in many areas. In fact, pidginized Hausa is still spoken by non-native speakers of Hausa in the markets around Lake Chad while a pidginized form of Igbo is used at present in some Niger Delta markets“ (Faraclas 3).
From this we clearly see that the term Nigerian Pidgin is somewhat misleading, as it is the first language, and thus a creole, for a growing number of Nigerians and serves as a lingua franca for many others. Yet another aspect is that for the pidgin-speaking child from an elite family who hears Nigerian Standard English at school and on the radio, Nigerian Pidgin is in all probability a somewhat decreolized speech form already (cf. chapter 1.2.2). Given these observations, Nicholas G. Faraclas postulates that:
If present trends continue, Nigerian Pidgin will be spoken by most Nigerians by the year 2000 and it is already the most widely spoken language in the country. Nigerian Pidgin is distinguished from the other 400 or so languages by the fact that it is spoken by members of every regional, ethnolinguistic and religious group in the federation. Nigerian Pidgin is distinguished from Nigerian Standard English by the fact that it is spoken by members of every socioeconomic group, while only those with many years of formal education can claim to speak Standard English with any proficiency. For an understanding of Nigerian affairs and for practical communication in Nigeria, a knowledge of Nigerian Pidgin is fast becoming indispensable. (Faraclas 1-2)
To the present day, the Nigerian authorities regard a pidginized form of English as inappropriate to serve as the national language. This is partly due to the fact that pidgins are still widely seen as “broken Englishes“, but it also serves as a means for the Standard English-speaking elite to keep the common people out from politics. In his article “The National Language Issue: A Revisit“ (see chapter 3.3.2.1), Ejike Eze claims that “the social and political elite who are generally highly educated in the colonial language have a major stake in its propagation since its continued use allows them a major share of lucrative jobs and advances their social position and power“. It is notable that Tok Pisin already enjoys the prestige Faraclas wants for Nigerian Pidgin:
The vast majority of the speeches on the House of Assembly of Papua New Guinea are made in Tok Pisin, which is but one indication of the importance of this language in the emerging urban and national society of the country. New Guineans are well aware of the financial and other social advantages that may accrue to fluent speakers of Tok Pisin and are accordingly anxious for their children to learn it. (Kay/Sankoff 64)
I decided on Nigerian Pidgin for my Internet research, because it is on the one hand the most widely spoken pidgin in the world. On the other hand, pidgins and creoles are a rather exotic topic on the Internet. The combination of these two factors guarantee a relatively clear number of online documents. Nigerian Pidgin is widely used in the broadcasting media as well as in novels, plays and poetry. In addition, fellow students from Nigeria informed me about the existence of some pidgin newspapers in Rivers State and Lagos. Moreover, it is remarkable that Nigerian Pidgin is one of the two pidgin/creole languages in which the “Universal Declaration of Human Rights“, available in all major languages, has been translated. The other one is Haitian Creole.
Nigerian Pidgin owes much of its importance to literary works by Nigerian anglophone writers who deliberately employ African-specific elements to establish local color. As a matter of fact, modern Nigerian literature is nowadays considered the most important one in Black Africa. Some of the material about Nigerian Pidgin is connected with these writers, so it is a good idea to familiarize yourself with them: Chinua Achebe, the author of “Things Fall Apart“ and “A Man of the People“, uses Igboisms that can be understood from the context. But the non-Nigerian reader often cannot do without a glossary of Yoruba words and pidgin words when reading plays by Wole Soyinka, who is also known to frequently employ gods from the Yoruba pantheon without explaining them for his non-Nigerian readership. Soyinka achieved world-wide recognition when in 1986 he became the first Black African writer to be awarded the Nobel Prize for Literature. While Chinua Achebe once stated that “the English language will be able to carry the weight of my African experience. But it will have to be new English, still in communion with its ancestral home but altered to suit its new African surroundings.“, Gabriel Okara, another Nigerian writer, asks more directly: “Why shouldn’t there be a Nigerian or West African English which we can use to express our own ideas, thinking and philosophy in our own way?“ (cf. Kachru). Ken Saro-Wiwa, playwright, environmental activist and President of the “Movement for the Survival of the Ogoni People“, executed by the Nigerian regime in 1995, wrote his novel “Sozaboy“ in a language he himself labeled “rotten English“. In the preface, he describes it as
[...] a mixture of Nigerian Pidgin English, broken English and occasional flashes good, even idiomatic English. This language is disordered and disorderly. Born of a mediocre education and severely limited opportunities, it borrows words, patterns and images freely from the mother-tongue and finds expression in a very limited vocabulary [...], and is part of the dislocated and discordant society in which Sozaboy must live [...].“
Personally, I find the coinage “rotten English“ unfortunate. It reminds me too much of the terms ignorant people use to refer to what they erronneouly consider merely as poor attempts at mimicking Standard English. The next chapter will bring your attention to similar terms.
1.2 The Inception of a New Academic Field
1.2.1 The Key Debate: Theories about the Origins of Pidgins
Creolists are interested in contact languages, mainly pidgins and creoles, which have arisen through the need of a medium for interethnic contact. Many creolists try to understand the nature of this development process. Most are also interested in how the pidgin and creole languages are used today. Besides, any subfield of linguistics can be applied to pidgin and creole languages, of course.
The significance of pidgins and creoles to general linguistics, anthropology, sociology and related fields has long failed to be recognized. Derisive terms such as “broken English“, “bastard Portuguese“, “nigger French“, “mongrel lingo“ or “baby talk“ all too clearly reflect what earlier generations thought of pidgin and creole languages. As I mentioned before, this partly explains also why Nigerian Pidgin, understood by a high percentage of the country’s population, is still considered “inappropriate“ as the official language by the authorities, and consequently Standard English, spoken by only a small percentage, retains that status.
For a long time, even linguists dismissed pidgins and creoles as merely defective versions of their lexifiers, and unfortunately, the equation of pidgin with bad language has still survived in the minds of some people. In her report “Auf Wiedersehen, English“, Ursula Sautter quoted Walter Krämer, the president of the Society for the Protection of the German Language, an organization that takes up arms against anglicisms, as follows: “The German tongue is deteriorating into a pidgin dialect which will soon no longer be usable as an independent cultural language. We fight against this kind of chimpanzee language.“ Firstly, pidgins are no dialects, and secondly, certain pidgins are widely used in parliament, radio programs, novels, etc. The abuse of the term “pidgin“ in Krämer’s statement reveals his ignorance.
It was not until the first international conference on creole language studies, held in Jamaica in 1959, that pidgin-creole studies became a respectable academic field. Finally, the possibility of a monogenetic theory of pidgin-creole origin, namely the Portuguese-origin hypothesis, advanced sixty years earlier by Dirk Christiaan Hesseling, became a topic of discussion. As to the next conference, David DeCamp writes:
If pidgin-creole as a separate discipline was born in 1959, then it came of age in 1968. The second international conference, also held in Jamaica, revealed how much the field had grown in the intervening nine years. The second conference was attended by several times as many scholars as the first, representing a much greater number of territories and interests.
The main research effort in pidgin and creole studies is to find out how those languages evolved, for example whether there has been a sort of “proto-pidgin“, a single language which has developed distinct and mutually unintelligible varieties under the influence of English, French, Portuguese, etc., or whether each pidgin and creole is genetically related to the corresponding standard language, from which it diverged under the influence of a similar sociolinguistic situation. These two opposing notions are called monogenesis and polygenesis.
Early polygenetic approaches involved the transformation of the European source languages, claiming that each pidgin began as a sort of baby-talk used by masters, plantation owners and merchants to communicate with their servants, slaves, and customers. This reflects the eurocentric and racist attitude that speakers of “inferior“ languages are also less capable of learning. As DeCamp sees it, this “baby-talk theory“ is very easy to refute:
If each European had indeed improvised his own variety of baby-talk to communicate with his servants and slaves, how could one explain the fact that all dialects of creole French, including those in the Indian Ocean, are mutually intelligible? The typological similarities shared by creole French, English, Spanish, etc., are too great for coincidence, and when we consider that these creoles also share many common vocabulary words, including syntactic function words, the baby-talk hypothesis completely collapses. (DeCamp, 19)
These similarities appear to favor monogenesis, which claims a single origin for pidgins and creoles. Basically, there are two primary versions of this. The first derives all creoles from the West African Pidgin Portuguese. In the early 1960s, Douglas Taylor and W. Thompson assumed that this Afro-Portuguese jargon was widely spoken from the 15th century to the 18th century in and around the numerous forts and trading settlements founded by the Portuguese along the West African coast. In order to explain the striking similarities between these creoles and creoles with different lexical bases, it was hypothesized that the French, English and other creoles were “relexified“ in a later stage, meaning that the Portuguese lexical items were replaced word for word with French or English items (cf. Arends 88). The similarities of the creole languages would then be due to the underlying Portuguese jargon. The second version incorporates the first and assumes additionally that this pidgin in turn was derived form the medieval Sabir. Monogenesis thus has as its key mechanism the concept of relexification or vocabulary shift. In this connection, it should be pointed out that this theory obviously cannot apply to those pidgins and creoles with non-European lexifiers.
A more restricted approach is to assume that English and French-based creoles derived from a Western African Pidgin English and West African Pidgin French, respectively. The fact that each of these two families of creoles display significant parallels which are not shared between the two groups would suggest a common origin for either of the creole families.
One of the thoughts that presently dominate the creolistic world is the belief that creoles look the way they do because various features were transmitted from the languages spoken by the ancestors of today’s creole speakers, the substrates, much of the non-European looking features in e. g. Haitian Creole would thus be of African origin. Most believe that the socioeconomically disfavored populations who had to adopt a new language were incapable of learning the superstrate (another term for lexifier language), simply because it is difficult for adults to learn a new language.
Taylor suggested that the Atlantic creoles started out as an Afro-Portuguese pidgin that later changed its lexicon under the pressure of various other languages, but more or less maintained their grammatical structures. His comparison of linguistic features common to Yoruba and Atlantic creoles showed that “while African loanwords are relatively few in most West Indian creoles [...] African loan constructions are both common and striking“ (Holm 65). Given these structural similarities, he concluded that:
Lesser Antillean Creole French in its formative period was in close contact with a language or languages very like Yoruba; and as French-based Haitian and Cayenne creoles, English-based Sranan and Saramaccan, and Iberian-based Papiamentu and Saotomense (Gulf of Guinea) show very much the same and other resemblances to Yoruba [...] we conclude that these creoles have diverged from what may well have been a common pidgin by lexical replacement from the languages of the slaves’ European masters and overseers. (Holm 66)
Thus, Taylor’s findings attributes the West African substratum to a language very closely related to Yoruba, which is a typical example of the Kwa branch of languages. This theory is supported by a comparative study of various features in creole and African languages, that demonstrated that a significant proportion of the slaves brought from Africa to the New World during the period in which the creoles emerged were speakers of Kwa languages.
With respect to Nigerian Pidgin, the exemplary pidgin/creole in this study, Faraclas suggests a development process that also helps to support Taylor’s theory, considering that most of the slaves who were deported to the New World, the discovery and exploration of which had led to a huge demand of cheap labor to work on sugar cane and cotton plantations, came from the West African coast:
[...] it is very likely that pidginized versions of Nigerian languages were widely used in many areas [...]. Nigerian Pidgin may very well have developed from one or several such pidginized Nigerian languages that were spoken along the coast before the Europeans arrived. Because of the importance of the European trade and the reluctance of Europeans to learn other languages, European words would have been substituted for Nigerian words to facilitate communication. Since the Portuguese arrived first, a few Portuguese-derived items such as sàbi “know“ and pìkîn “child“ would have been initially adopted, but a the British consolidated power over Nigeria, more and more English words would have been integrated into the language [...]. (Faraclas 3)
Polygenesis used to offer no explanation why pidgins evolved from different source languages can share similar structural features. Here, Noam Chomsky’s theory that children are born with an innate predisposition to recognize certain universals in simplification that facilitates their acquisition of the language of their particular speech community could provide supportive arguments. In the late 1970s, Derek Bickerton proposed the Language Bioprogram Hypothesis. Influenced by Chomsky’s theory, he concluded that creole languages reflected structures innate to the human brain. The children who were to constitute the first generation of creole speakers were exposed, he claimed, to a completely chaotic linguistic input. Their parents had different linguistic backgrounds and communicated with each other in a primitive makeshift jargon. The children expanded the jargon, so that it could serve all purposes. This expansion was done not with existing languages as the pattern, but a genetically defined language faculty of the human brain. Therefore, creoles reflect the universal linguistic principles of the human mind, and that is why they are so alike all over the world.
Todd noted possible parallels in child language acquisition and the formation of pidgins:
It is a stimulating thought that pidgins may result from such “simple“ exchanges. Babies soon discard the simple “idealized dialect“ because social pressures put a premium on their acquiring the language of the adult community. But such pressure did not, in the past, prevail in pidgin situations and so the urge to modify towards a more “acceptable“ norm was not a factor in the formation of pidgins. (Todd 47)
Whereas in the past either the monogenetic or the polygenetic approach was held as the only possible explanation, many creolists today prefer a mix of theories. Many of those who are skeptical to the Language Bioprogram Hypothesis, for instance, agree that universal principles are relevant in choosing structures provided by the input languages, but reject the possibility that these universal principles could create such structures. Departing from an essay by Salikoko S. Mufwene, Karl-Heinz Stoll concludes that:
Es scheint, daß beide Ansätze, die sich als “substrata versus universals“ (vgl. Muysken/Smith) kontrastieren lassen, jeweils für Teilbereiche ihre Gültigkeit haben und so einander ergänzen. Der in den USA lehrende zairische Linguist Mufwene meint in einem Aufsatz mit dem programmatischen Titel „The Universalist and Substrate Hypotheses Complement One Another“, die weltweiten Ähnlichkeiten von Pidgin- und Kreolsprachen könnten am besten durch Universalien, die gemeinsamen Verschiedenartigkeiten von einigen dieser Sprachen aber durch Substrateinflüsse erklärt werden. (Stoll 194)
1.2.2 Other Topics of Discussion
The creole origins issue continues to inspire more controversy and new research. John Holm once wrote that:
An adequate theory of pidgin/creole origins might contain elements of all the above approaches, and each of them may explain different aspects. In any case, attempts to develop such a theory will be beneficial to general linguistics (including the field of natural language acquisition) as well as to related fields.
Modern creolists also conduct research on many other aspects. Now that the value of pidgin-creole studies is recognized by a growing number of linguists, sociologists, anthropologists, educators, etc., further publications in this field will appeal not only to scholars, but to just about everybody who takes some kind of interest in languages. Take Charles Mann’s work on Nigerian Pidgin for example. His publications include titles such as “The sociolinguistic status of Anglo-Nigerian Pidgin: An overview“, “The place of Anglo-Nigerian Pidgin in Nigerian Education: A survey of Policy, Practice and Attitudes“ and “Language, mass communication and national development: The role, perceptions and potential of Anglo-Nigerian Pidgin in the Nigerian mass media“. At present, he is conducting research on the sociopsychological attitudes of the urban population in southern Nigeria to Nigerian Pidgin. (In chapter 3.3.3, we will learn why Mann favors the term Anglo-Nigerian Pidgin.)
Or consider the issue of Ebonics: The discussion whether non-standard Black English should be considered a creole, and by this an independent language, dates back to the time of the first international conference on creole language. In 1997, it became a highly controversial topic of discussion in the United States, after the School Board of Oakland, California, decided to use Ebonics in class. Advocates of the use and expansion of Black English claimed that students can learn to read and write sooner if they are learning in their own speech variety, which is beneficial to learning in general. They argued that ethnic varieties of speech function is as a symbol of ethnic identity, and can enhance the sense of group solidarity that is important for the empowerment of minorities.
Others took the view that it is indispensable to master Standard English to be successful in work, and that encouraging the use of nonstandard varieties may make it harder to learn Standard English. On the basis of a study involving black speakers in the United States, John R. Edwards indicated that “ [...] speech patterns of regional speakers, ethnic group members and lower-class population [...] evoke unfavorable reactions, at least in terms of status and prestige, from judges who may or may not be standard speakers themselves (Edwards 26).“ So far, the issue has not met with the necessary political (and by that financial) support to become implemented, and fears that official recognition would lead to undesirable consequences cannot be simply brushed aside.
Language is exposed to permanent changes, and as language “moves on“, pidgins and creoles will undergo further changes as well. At present, language experts discuss whether the numerous varieties of English (or “Englishes“) will eventually develop into different languages, as it was the case with Latin in the past. Indeed, speakers of Indian English or West African English no longer consider their way of speaking inferior to British English, but even take pride in it. The issue of Ebonics is a clear sign of this new self-confidence. In his paper “Norms, Models, and Identities“, Braj B. Kachru argues that “English has gradually developed new local centres for authentication of its models and norms. In other words, it has become a pluricentric language with Asian and African norms and models for its acquisition, its teaching, and creativity in the language.“
On the other hand, intensifying contacts and the modern mass media bring about a post-creole continuum that work against such a diversification development. This process, known as decreolization, can be expected to bear especially great influence on territories that are politically and culturally connected to the “home country“. When a creole exists in a community where its lexifier language is the language of education and politics, the two linguistic systems inter-influence each other and people usually start to “improve“ their creole using the standard language as their model (cf. Wardhaugh 78). There is some evidence that Black English in the United States is an example of a creole English which has been exposed to standard for such a long time that it is now in an advanced state of decreolization (cf. O’Donnell/Todd 52).
Various other things can happen to a creole. It can reach a relatively stable relationship with the other languages of the community, it may be extinguished by the standard language, and in some cases a creole has become the standard language, e.g. Afrikaans, Swahili and Maltese (cf. Wardhaugh 82). What happens to a creole under which circumstances? Why is a standard language or established dialect not acquired? These are only some of the questions that creolists try to answer.
A clear proof that creoles have gained some importance is the fact that they have recently entered the domain of machine translation. For instance, Jeff Allan, a creolist who has been involved in research on high-quality multi-lingual machine translation, reported on a project for areas like Miami, where Haitian children attend school in English, “but their parents may be Haitian monolingual. I have heard of a project for developing an English-Haitian Creole system to translate grade reports and correspondence from the schools to the parents and vice versa as this is sometimes necessary in such communities.“ (Jeff Allan in his e-mail from March 26, 1999).
When doing research, people traditionally read reference books, conduct interviews, attend conferences, etc. With the advent of the Internet, another important research source now opens up before us. It is the objective of this study to show what kind of information can be found on the Internet, and how this material should be treated.
2.1 Who Uses the Internet?
“In 1973, the U.S. Defense Advanced Research Projects Agency (DARPA) initiated a research program to investigate techniques and technologies for interlinking packet networks of various kinds. The objective was to develop communication protocols which would allow networked computers to communicate transparently across multiple, linked packet networks. This was called the Internetting project and the system of networks which emerged from the research was known as the Internet“. This is how Vint Cerf, one of the researchers and sometimes called “the father of the Internet“, remembers the early stage of the Internet in his paper “A Brief History of the Internet and Related Networks“ (www.isoc.org/internet-history/cerf.html).
The development of the Internet was largely financed by the U.S. Federal Government, since the Internet was originally part of a federally-funded research program and, subsequently, has become a major part of the U.S. research infrastructure. During the late 1980s, however, the population of Internet users and network constituents expanded internationally and began to include commercial facilities. Indeed, the bulk of the system today is made up of private networking facilities in educational and research institutions, businesses and in government organizations across the globe.
The World Wide Web (often just called the WWW or the Web) is the fastest growing Internet service. Through the Web, you can view images, look at film clips, hear sound recordings, and find valuable and interesting information on any given topic. In fact, the Web has become so popular that very often people are only thinking of this service when they talk about the Internet, which, to be accurate, also comprises services such as e-mail and newsgroups.
Tim Berners-Lee invented the World Wide Web in late 1990 while working at CERN, the European Organization for Nuclear Research in Geneva, Switzerland. He is credited with developing the idea of combining hypertext with the speed of today’s electronic networks. Working with a small team, he developed the Hypertext Transfer Protocol (HTTP) on which the Web is based. Berners-Lee is now the Director of the W3 Consortium, an open forum of companies and organizations with the mission to realize the full potential of the Web (cf. www.w3.org/People/W3Cpeople.html and whatis.com/cern.htm).
The Web’s special feature is that its hypertext documents allow you to move from one document to another by selecting highlighted links. Document authors can thus allow you to access related Internet sources. These documents were created using Hypertext Markup Language, or HTML, and are often referred to as HTML documents or Web pages. According to Torben Kjaer, a Web site denotes a connected collection of such Web pages:
A Web site is a collection of pages centered around a particular subject, company, organization, service or person. A Web site can be many thousands of pages, or very few. The Web pages are usually all sitting on one Web server, and there is always a home page. This home pages is usually some sort of welcome, perhaps containing a map on the entire Web site together with buttons or links to navigate around the pages, and information on who owns the site. The home page is usually the page which is sent if you just type in the address of the Web server without giving a file name. (Kjaer 11)
The first computers to be attached to the network were computers on American university campuses. Today there are millions of computers attached to the Internet. Now as ever, scientists, scholars and students are among the most vigorous users of the Internet. It enables them to gather information and to communicate with collaborators from far-off places. Other groups who benefit from the masses of scientific documents on the Web, from the numerous online dictionaries and encyclopedias, etc., are e.g. journalists and translators. The Internet is also a rich source for current local, national, and international news. There are hundreds of individual news sites on the Web, from broadcast sources such as CNN or ABC to newspapers including the Washington Post and the New York Times. CNN Interactive, for example, is even searchable. Some search tools cover multiple news sites and news wires, including Yahoo! News (dailynews.yahoo.com) and HotBot News Search (news.hotbot.com). In my search for Nigerian Pidgin, they both retrieved no news articles, so I excluded them from this study. They are certainly a valuable source for research on other topics.
A relatively new market for the computer industry is the Intranet, an internal company version of the Internet, introduced in 1996. Intranet allows workers to electronically access information from company computers via the same user-friendly browsing software used on the Internet. Corporations that adopted this approach said Intranets simplified the work of their employees and thus led to higher worker productivity and lower frustration levels (cf. Alexander 181).
Companies, university members and journalists not only consume the Internet, but in return supply much of the information available on the Web. Universities present their research results, libraries disseminate some of the most valued information, and magazines and newspapers often have online versions. Many companies include a Web site address in their print advertising and television commercials. They make use of the Internet to reach a large audience. 56 percent of U.S. companies will sell their products online by 2000, up from 24 percent in 1998, according to a survey by the Financial Executives Institute and Duke University (cf. www.nua.ie/surveys). All major firms and banks run their own Web sites, and some companies offer free information about a certain subject in the hope that you end up buying their products. Consequently, it is often hard to differentiate between marketing and real information on the Web. And in addition to this, many Web sites are supported by advertising revenue. At other Web sites you receive free information only after giving personal information about yourself, your age, your income, etc. As Kjaer points out, the purpose of collecting these data is usually to be able to come back and try to sell you something. He also warns us that the information could be sold on to other companies (cf. Kjaer 47).
Other groups that seek to present their work on the Web are all kinds of organizations (including political parties) and institutes. Last but not least, the Internet offers innumerable personal home pages, which deal with any conceivable topic.
2.2 Search Tools
2.2.1 Two Categories of Search Tools
In order to go directly to a specific Web site, you need to know its Uniform Resource Locator. URLs are the addresses of documents available on the Internet. As they are built up very logically, it is often possible to guess addresses. For many sites, you do not have to bother with a subject directory or a search engine. When you are searching for a specific organization’s Web site, first try guessing the central URL for the organization. Use the name, acronym, or brief name of the organization (e.g. “un“ for United Nations) in the middle and then add the appropriate top level domain. Incidentally, if you enter uno.org instead of www.un.org into the address field, you will be taken to the home page of the United Networks Organization, which is in no way associated with the United Nations. They are, however, kind enough to offer a hyperlink to the latter! However, if you enter vlib.com instead of vlib.org in your search for the WWW Virtual Library, then you will find yourself at a commercial site, which has not taken the effort to forward your query. All in all, there are six common top level domains: com (for commercial companies), edu (for educational institutions), org (for non-profit organizations), gov (for U.S. federal government), mil (for U.S. military) and net (for Internet service providers and networks). The com and org extensions are not limited to the USA, but can be used by anybody who is willing to pay for them. The URL of a Web page can provide some assurance that its information is reliable. If the Web page resides on a computer belonging to an official organization, this is a good sign. For URL guessing, it is also important to know the most common country codes, e.g. de for Germany and fr for France. The country code for Nigeria is ng. If you cannot guess the URL, you have to search for the document with the help of one of the numerous search tools.
Web search tools fall into two large categories: subject directories (or indexes) and “ordinary“ search engines. One difference between the two is that you can browse the vast collection of categories and sub-categories of a directory such as Yahoo! If you are lucky, you find by browsing a complete listing of all the sites that cover a particular subject. Subject directories include human-selected Internet resources and are arranged and classified in hierarchical topics. An increasing number of universities, libraries, companies, organizations, and even volunteers are creating subject directories to catalog portions of the Internet.
A good rule of thumb is: If your topic is general and consequently widely covered, go for a subject directory first. If your topic is very specific or you want more than a general overview, go for a search engines such as AltaVista, HotBot and InfoSeek. They are sometimes called portals, as they are usually the starting point for search queries. Most search engines are sponsored by advertisers who purchase advertising banners. As a matter of fact, portals are great money-makers.
The major search tools tend to have overlapping but different databases. For instance, if Yahoo! cannot find Web sites within its own categories, the query will be automatically transferred to the Inktomi database, which provides the underlying technology also for other search engines such as HotBot.
2.2.1.1 Subject Directories and Other Internet Guides
Subject directories are valuable for their smaller size of hand-picked sites. For topics with vast quantities of information on the Web, beginning in a subject directory often helps sort out meritorious sites from those that may mention your topic without in-depth treatment. Probably the most complete hierarchical, topical index of Web sites is Yahoo! In addition, it features a sophisticated search facility, which means that it is not only browsable, but also searchable: You can either click through the various categories or enter one or more key words into the search box, which will supply you with a list of categories that contain these words in the title or description.
Yahoo!’s 14 main categories that serve as starting points are Arts and Humanities, Business and Economy, Computers and Internet, Education, Entertainment, Government, Health, News and Media, Recreation and Sports, Reference, Regional, Science, Social Science and Society and Culture. Everything contained within Yahoo! lives somewhere in these categories. And this is how the subject tree principle functions: Click on the category that is the nearest match for the subject you are looking for. For example, the main category “Social Science“ combines sub-categories such as Ethnic Studies, History, Psychology and Linguistics and Human Languages. This will display a new page containing sub-categories of the category you chose. A click on “Linguistics and Human Languages“ will show you sub-categories such as Institutes, Journals and Sign Language. Say you are searching for Nigerian Pidgin. Logically, your next clicks are Languages and then Specific Languages. Under Specific Languages I found a list which contains a choice of 90 different languages. There was even a link for “Pidgin“ (we of course know that the term pidgin serves to denote a number of very different languages), and I was eager to see which pidgins/creoles would be mentioned. However, it turned out that this link’s only Web site was a Tok Pisin/English dictionary.
Later I discovered the short list of pidgin-related links that Yahoo! has drawn. To see it, you have to click through to Contact_Languages via Social_Science, Linguistics and Human Languages, Languages, and Language Groups. It is also possible to access this page directly by entering dir.yahoo.com/Social_Science/Linguistics_and_Human_Languages/Language/
Language_Groups/Contact_Languages into the address field. These were the links that this list presented in late March 1999: Tenas Wawa (a semi-monthly newsletter about the Chinook Jargon, a pidgin of Northwest Native American languages, English and French), Glossary of the Chinook Jargon, Creolist Archives, Journal of Pidgin and Creole Languages, Kreyol Grammar (a short description of Haitian Creole), Pidgins and Creoles, Post-Contact Languages of Western Australia, Reference Guide for Pidgin and Creole Languages.
This demonstrates that it is not worthwhile browsing a subject directory when searching material about Nigerian Pidgin, for this topic is apparently still too specific. Yahoo! serves very well as a general starting point, though, as we will explore in chapter 3.3.1.
There are also other Internet guides from which you can start your search. The best-known one is the WWW Virtual library at vlib.org, maintained by CERN and later by the W3 Consortium. It is the oldest catalog of the Web, started by Tim Berners-Lee, the creator of the Web itself. Unlike commercial catalogs, it is run by volunteers who compile pages of links for areas in which they are experts. Categories include “Agriculture“, “Computer Science“ and “International Affairs“, sub-categories include “Gardening“, “Journalism“ and “Linguistics“. The latter is maintained by the Linguist List (cf. chapter 3.3.2), to which you will be automatically transferred. Other examples of Internet guides are Galaxy (galaxy.einet.net/galaxy.html) and Clearinghouse (www.clearinghouse.net/index.html). All these highly selective indexes have the disadvantage that they, unlike Yahoo!, do not run spiders that constantly update and enlarge their databases (see next chapter).
2.2.1.2 Search Engines
When you use a search engine, you search the contents of its database - not the World Wide Web directly. Each search engine operates on its database of URLs, texts, and descriptions selected from the entirety of the World Wide Web. Since none of these databases includes all the Web pages in existence, you get different results from different search tools. If the Internet were a book, search engines would be the index. In order to combat the dynamic nature of the Web, search engines are constantly running software called “spiders“, “robots“ or “crawlers“ that read entire Web sites and update the index entries for that specific search engine. If you want to sort out “the best“ pages on an extensively covered topic, you may profit from the selectivity of some of the smaller, hand-picked databases, whereas large databases are useful for more comprehensive searches.
To use a search engine, you access the search engine’s site, type one or more key words, and click the submit button. If the search engine is able to locate any documents that match your search query, you will be offered a list with hyperlinks that you can click to access the listed documents. Usually, relevancy is computed by examining the frequency of your key words in the retrieved documents. However, a couple of more recently introduced search engines take other approaches.
The large number of available search engines makes it tremendously difficult to choose the right one(s) for your query. Some are very comprehensive, such as HotBot and AltaVista, others are very specific, for instance “the cool gay search engine“ at www.pridelinks.com. All the more reason why you should familiarize yourself with at least the most popular search tools and keep track of new search engines that take innovative approaches. Two newer search engines, Google and Direct Hit are included in this study. A third one, GoTo, launched in June 1998, is according to my experience less promising, but I think it is quite interesting to give you a brief outline of its special approach: Rather than ranking search results according to where and how often certain key words appear, as conventional search engines do, GoTo ranks results according to how much sites are willing to pay. This allows GoTo to do without an advertisement-oriented interface. It has no stock tickers, free e-mail, horoscopes, or other extraneous content that most other search engines offer.
2.2.2 Useful Features of Search Tools
Every search tool is different. They vary in features and size/comprehensiveness. Each has its special features and drawbacks. The most important features in selecting a search tool are those which allow you to refine or focus your search when you need to. I would now like to introduce the most helpful ones.
The way to combine terms using “AND“, “OR“, and “NOT“ is called Boolean logic (after Victorian mathematician George Boole). AND requires all terms to appear in a document, OR retrieves pages with either term, NOT excludes terms. It is important to note that the absence of a symbol is also significant, as the space between key words defaults to either OR logic (e.g. in AltaVista, Excite, InfoSeek) or AND logic (e.g. in HotBot, Lycos, Google). Include synonyms or alternative spellings in your search statements and connect these terms with OR logic, for example “freebie OR freebee“. In some rare cases, using only one of the synonyms is more effective. If, for instance, you use the word homosexual, you are likely to run across more scientific information, whereas the key word gay will return you cultural information on happenings, etc.
Certain search engines allow you to use a proximity operator. This is a type of AND logic which specifies the distance between words in a source file. AltaVista and Lycos are among the search engines that let you use the NEAR operator. For instance, say that you are looking for United Nations documents regarding human rights abuses in Nigeria, you may try the following search string: “united nations“ AND “human rights“ NEAR nigeria. In AltaVista, the two terms must be within 10 words of each other in the source file. Lycos allows user-specified distances. Use of this option can help you gain relevance in your search results.
It is a good idea to make use of these more advanced techniques from the outset, instead of only using key word search. Simple key word searching often retrieves irrelevant or too many documents in large databases. Advanced queries can reduce irrelevant hits by making a query more specific. In small databases and in subject directories, however, simple key word searching is usually the best approach, since the small size of the databases makes more complex searching unnecessary. AltaVista instructs its visitors to use advanced search “for very specific searches and not for general searching. Almost everything you need to search for can be found quickly and with better results using the standard search box, where the AltaVista search services sorts the results by placing the most relevant content first. However, if you need to find documents within a certain range of dates or if you have to do some complex Boolean searches there isn’t a more powerful tool on the Web“ (www.altavista.com/av/content/help_advanced.htm).
There are other useful features of search engines. “Phrase searching“ is a feature that requires all the terms to appear in exactly the order you enter them. You only have to enclose the phrase in double quotations, e.g. “World Wide Web“. Otherwise you will get millions of irrelevant Web sites that contain at least one of these three frequent words. Are you looking for terms with many possible endings? Truncation permits retrieving all these variations in one search term. All you have to do is to enter the first part of a key word and insert a symbol, usually the asterisk (*). Examples: femini* retrieves feminine, feminist, feminism, etc., sara*evo retrieves any variant internal spelling. Besides, truncation also enables searching for singular and plural words at the same time. A search for glossar* consequently finds the words glossary, glossaries as well as the German words Glossar and Glossare.
The use of parentheses can also be very helpful. For instance, if you wanted to look up the income tax information for the United States, an advanced query might look like this: “income tax“ AND (“united states“ OR usa OR america). This query instructs the search engine to search for the phrase income tax whenever it appears on the same page with United States, USA, or America.
Furthermore, some search tools let you limit your search to the title field. And indeed, the search title:Krio on AltaVista, HotBot and Yahoo! retrieved only those documents that contained my key word in the title when I tried out this special restriction method in April 1999. The search title:“nigerian pidgin“, however, resulted only on HotBot with one matching Web page: “OHCHR: Nigerian Pidgin English Universal Declaration of Human Rights. The United Nations Universal Declaration of Human Rights, translated into Nigerian Pidgin English by United Nations Information Centre (UNIC, Lagos Nigeria).“ In chapter 3.2.1, I will explain why this is one of the few top quality pages. For this particular query, the other two search tools turned out to be complete failures. AltaVista and Yahoo! not only yielded no single match for the query title:“nigerian pidgin“, they did not even have the above page indexed. The conventional key word search merely retrieved the “Alphabetical Listing of All Translations“ on AltaVista (in 22nd place) and the “Regional Listing of All Translations“ on Yahoo! (in 2nd place), from which you then have to click your way to the Nigerian Pidgin version. This clearly shows that search engines can differ enormously in their results and are consequently worth a comparative examination.
Some search engines allow choosing a certain language. But be cautious when using this option. Most of the time, this function to limit the number of documents orientates itself to the country codes. This means that a search in German retrieves only documents that carry the extensions de (Germany), at (Austria) and ch (Switzerland), but neglects German documents from overseas and the home pages of German companies that operate on a global scale, since their addresses usually end with com.
2.2.3 Meta-Search Engines
In ordinary search engines, you submit key words to a single database of Web pages owned by the search tool, and you get back a different display of documents from each search engine’s unique database of Web pages. Results from submitting very comparable searches can differ widely, but also contain some of the same sites.
Meta-search engines, also called multiple search engines, claim to search all of the big search databases in parallel and to give you a consolidated report of their findings. You enter key words into the search box, and they transmit your search simultaneously to most of the popular search engines and their databases of Web pages. Then you get back a compilation of results containing matching sites from all of the search engines queried. This can save you a lot of time and provide an overview of the kinds of documents available on the Web matching any term or phrase. It may result in locating exactly what you want, especially if you are searching a unique term or phrase.
Meta-search engines do not own any database of Web pages; they use and deliver the databases and searching programs of each of the popular, individual search tools they query. Meta-search engines act as intelligent middle-agents to pass your search through, gather the responses from the individual search tools they query, and then give you a more unified report of results from many different resources. By necessity, meta-search engines are more powerful than the average single search engines, so the fact that most people I asked made exclusive use of individual search engines kept me pondering for a long time. Finally, I turned to Laura Cohen for expert guidance, as she maintains an excellent Internet tutorial at www.albany.edu/library/internet. Shortly after, she replied to me via e-mail and provided me with the following authoritative answer:
The disadvantages are:
With most meta-engines, you are getting a small number of hits. You have to be satisfied with a small number of (usually) relevant hits.
You can’t do complex searches. There is very little fields searching or limiters you can employ on your searches with meta-engines.
There are so many fascinating and useful new engines, such as Direct Hit and Google. If you spend too much time on a meta-engine, you miss some of the latest technological advances.
Meta-engines are very good for what they do, however. They are especially great for obscure topics, or for a picture of what the Web might offer in your topic area. (Laura Cohen in her e-mail from March 2, 1999)
Later I found other sources that outline the significant limitations of meta-search engines as a comprehensive search tool. But none of them impressed me as much as Cohen’s reply did, which was concise and absolutely “hitting the nail on the head“.
2.2.4. On the Search Tools Covered in this Study
There are innumerable search tools to choose from, and their number is incessantly growing. For this comparative study, I decided to use three search tools that are extremely popular at present: the subject directory Yahoo! and the two search engines AltaVista and HotBot. According to a study conducted by the U.S. research institute NEC, no search engine covers more than 45 percent of all Internet sites (cf. Focus 6/1999, page 221). Greg R. Notess estimates that Northern Light indexed almost 130 million pages, AltaVista 106 million, and HotBot 99.5 million until March 1999 (cf. notess.com/search/stats/sizeest.shtml). As their methods of indexing new Web sites differ, the search results will differ too. Departing from his analysis, Notess concluded that Northern Light found the most unique hits, unfortunately also the most dead links, which are URLs that result in a 404 error message (file not found) from the server. The catch to these unique links, dubbed “Special Collection“ and depicted as a “unique combination of premium data representing over 5,400 journals, books, magazines, databases and news wires not easily found on the World Wide Web“ by Northern Light, is that you have to pay in order to see the entire document.
In chapter three, I will also put two new search engines to the test to see whether their approaches to relevancy ranking can guarantee more accuracy. In addition, I will compare the search results yielded on two popular meta-search engines to see whether they really make searching more efficient. But first, let me give you a short introduction to the search tools involved in this study.
Yahoo! (www.yahoo.com), launched in December 1995, is a relatively small database. All documents are selected by Yahoo! staff and assigned to a subject classification. Note that these editors do not evaluate the documents. Still, its selectivity can provide an excellent introduction into topics widely covered on the Web. Yahoo! is both browsable and searchable. You can browse Yahoo! by simply clicking on the various categories listed on each page. If your search is too specific for Yahoo!, as it is the case with Nigerian Pidgin, you can also search Yahoo! by entering key words into the search box. Yahoo! then searches for matches in its database and after that ranks the results in order of most relevant to least relevant. A factor that affects relevancy is the number of search words matched (the more words matched, the higher the rank). Besides, a match in the title of a site is ranked higher than a match in the comments or the URL (cf. howto.yahoo.com/chapters/7/3.html)
When you search material by browsing and no categories match your search terms, because your search is too specific for Yahoo!, your query will automatically be transferred to the Inktomi database, a search engine that specializes in indexing every single Web page it can find. This gives it a lot of raw data. As search engines provide good results with very specific requests (and often poor results with general requests), Inktomi will then find some suitable matches most of the time. In addition to this, Yahoo!’s Advanced Search searches DejaNews for Usenet messages (cf. 3.3.2).
AltaVista (www.altavista.com), equally launched in December 1995, is one of the largest databases. It offers two levels of searching: Simple and Advanced. The advanced search offers Boolean logic and results ranking, two of the finest features available in any of the search tools currently in existence, for refining searches and extracting the documents you want. AltaVista offers unusual options such as searching by language and translating search results. It is possible to have any Web page translated by the AltaVista translation service Babelfish (cf. chapter 2.3.1). Moreover, AltaVista is surprisingly good at answering plain-English queries such as “Where can I find a map of Nigeria?“ I will analyze this amazing possibility in chapter 2.2.5.
HotBot (www.hotbot.com), launched in May 1996 and now part of The Lycos Network, is another very large database with considerable potential to refine searches. It lacks truncation, but supports optional Boolean logic and phrase searching. HotBot permits geographical, media-type, and domain searching not available on other search tools. For instance, it is possible to search for only those documents that are hosted by European servers and contain images or videos. These are very useful functions that allow you to significantly narrow down searches based on the medium of the material you are looking for. There is also a date delimiter on HotBot, so one can specify the time frame of published material. Inktomi will continue to provide the underlying technology for HotBot searches. However, about half the first page of searches returned will be branded by Direct Hit’s ten most popular sites for a particular search.
Google and Direct Hit, both launched in 1998, are two of the newer search engines that have been attracting much attention. In chapter 3.2.2, I will put them to the test to see if they can live up to what several reports suggested.
2.2.5 How Search Tools Deal with Questions
Calling itself the “most powerful and useful guide to the Net“, AltaVista prides itself on answering plain-English queries such as “Where can I find a map of Nigeria?“ It has made preparations to answer such frequently asked questions. What happens is that AltaVista shows you the prearranged directory “Where can I find a map of + African country“, whereby you can choose any country ranging from Algeria to Zimbabwe. In this chapter, I will try out how AltaVista deals with these two questions:
Who won the Nigerian presidential election in 1999?
How many people speak Nigerian Pidgin?
Let us begin with the first question. AltaVista returned me two directories for this query: “What are the most recent election results in + African country?“ and “Who is the head of state of + African country?“ A click on the answer button next to the first directory took me to the CNN Web page cnn.com/WORLD/election.watch/africa/nigeria.html. This page is full of valuable information about the Nigerian election in February 1999, but contains no data concerning the election results! What about the second directory? This time AltaVista retrieved a Web page that listed all Nigerian leaders from 1960 to 1998, thus not answering my question, either. It is a pity that AltaVista obviously failed to update this search possibility. AltaVista is doing better with more popular countries, though. When I submitted the same query for Germany, I was already informed that Hans Eichel was the new Minister of Finance, some time before he actually took up the office.
When I entered the first question into HotBot’s search box, HotBot retrieved 230 matches. Most of them were completely irrelevant, however the second document already informed me that Olusegun Obasanjo was the winner of the Nigerian presidential election. The subject directory Yahoo! did even better: It showed me a list of nine relevant Web sites, the first one being the CNN Web page “Obasanjo wins Nigerian presidential election“ of March 1, 1999. Incidentally, it is possible to subscribe to the free CNN service “Headline News Mail“ at cnn.com/EMAIL. Then you will be sent a summary of the current news every day. And, not very surprisingly, on 1 March 1999, the top story of this service was “Obasanjo Wins Nigerian Presidential Election“.
Incidentally, Olusegun Obasanjo, just like Alex Ekwueme, another contender for the presidency of Nigeria, had launched a campaign Web site. However, in a country where few homes have a telephone and with an estimated 1,000 Internet subscribers, the candidates were unlikely to reach most of the electorate through that online campaign (cf. www.nua.ie/surveys).
Now to the second question: How many people speak Nigerian Pidgin? Again, AltaVista provided two directories: “Where can I find demographic information for + African country?“ and “Where can I find extensive historical, economic, and political information about the country + African country?“ A click on the answer button next to the first directory took me to a CIA (Central Intelligence Agency) Web page that gives very comprehensive information on Nigeria, for example that only 57.1% of the population are literate. Under “Languages“ we are supplied with the following information: “English (official), Hausa, Yoruba, Ibo, Fulani“, but with no details as to the actual number of speakers. (cf. www.odci.gov/cia/publications/factbook/ni.html
#people). For the second directory, and this is more interesting, AltaVista took me to the searchable Web site “Nigeria - A Country Study“ (lcweb2.loc.gov/frd/cs/ngtoc.html), which oddly enough none of the traditional searches with search tools yielded so far. I entered pidgin into the search box and was returned a fine article on the ethnic variety on Nigeria, unfortunately somewhat obsolete (“data as of June 1991“). Here is the relevant passage:
The official language of the country is English, which is taught in primary schools and used for instruction in secondary schools and universities. All officials with education to secondary school level or beyond spoke English and used it across language barriers formed by Nigeria's ethnic diversity. Many in the university-trained elite used English as one of the languages in their homes and/or sent their children to preschools that provided a head start in English-language instruction. In addition to English, pidgin has been used as a lingua franca in the south (and in adjoining Cameroon) for more than a century among the nonschool population. In 1990 it was used in popular songs, radio and television dramas, novels, and even newspaper cartoons. In the north, southerners spoke pidgin to one another, but Hausa was the lingua franca of the region and was spreading rapidly as communications and travel provided a need for increased intelligibility. Counting English, the use of which was expanding as rapidly as Hausa, many Nigerians were at least trilingual [...].
(cf. lcweb2.loc.gov/cgi-bin/query)
Here, AltaVista outperformed other search tools. Still, the question how many people speak Nigerian Pidgin is left unanswered. I also queried a couple of Nigerians and creolists, and in the following chapters, when I will report on their replies, we will see why this question is indeed so difficult to answer.
2.3. Critical Notes
2.3.1. General Reflections
Contrary to popular belief, the Internet is not a cure-all. It currently offers approximately 320 million Web sites (this figure is given both by Focus 6/1999, page 218, and in the paper “Informationsrecherche im Internet“ by Martin Kunz), but that does not mean that you will automatically find much valuable information on any given topic. Many of the resulting items will be peripheral or useless for your research. For many topics, there is simply a lack of quality information on the Web. The Internet is often good for the type of information that you would find in a product brochure, but quality sources of reference information are often non-existent on the Web. Where there is quality information, it is often available only from subscription-based sites, or you have to pay for them, which is true for many documents from the above-mentioned Northern Light.
Whether a specific search query will be crowned with success depends heavily on the kind of information you need, and on your search strategy. Doing research on esoteric topics that are hardly covered on the Internet can be as much time-wasting and frustrating as is dealing with lengthy lists loaded with irrelevant links. Very often, the chaff is out of proportion to the wheat, and the investment of time does not justify the paltry amount of information that the Web yields. Sometimes a trip to a traditional library would have been more helpful. Detlev Kalb, project director of the German search engine Fireball, even estimates that about one third of the German Web sites are trash (cf. Focus 6/1999, page 218).
Be aware that the addresses of Internet sites frequently change and that Web sites can vanish altogether. Do not expect stability on the Internet. A Web site that you find most valuable today might disappear tomorrow, perhaps for no other reason than that the maintainer of that site grew tired of the Internet. Or a certain Web site is currently being updated and therefore not accessible. In this case, the new search engine Google might come in handy: It keeps a copy of many of the Web pages it returns. If the server is down or otherwise not available, this so-called cached version might be helpful. Google claims that using these cached links is often “much faster than following the regular links, though the information you receive will be less up to date. But in many cases, no more frustrating 404 Not Found errors!“
Besides, it is a well-known fact that certain Web sites on the Internet are offensive to some or even most users. Unfortunately, the Internet also serves hate groups by providing them with an unprecedented opportunity to market themselves. According to a report by the Simon Wiesenthal Center, there are currently more than 1,400 Web hate sites on the Internet. This compares to 600 hate sites at the end of 1997, and only one hate site in 1995. The list included sites that were anti-Semitic, anti-Catholic, anti-Moslem, anti-abortion, as well as sites that promoted racism, hate music, neo-Nazism, and bomb-making (cf. www.nua.ie/surveys).
Those who are concerned about the proliferation of e.g. child pornography on the Web demand the elimination of these sites. On the other hand, however, we also find people who distrust censorship and fear that freedom of speech rights will be abridged. The Chinese government, by the way, makes short work of unwelcome views on the Web such as tibet.org, amnesty.org or cnn.com. These are simply blocked and cannot be visited by the ordinary people. Nevertheless, the Chinese government is wise enough to promote the Internet, being conscious of its economic potential (cf. Strittmatter).
Another point is that you also have to be literate in two senses in order to fully benefit from the Web: computer literacy and a sound knowledge of English are prerequisites. The marketing magazine Emarketer (www.emarketer.com) estimates that 70 to 80 percent of the Web pages are written in English (cf. bild der wissenschaft 3/1999, page 34). The computer industry is consequently another field that helps spreading anglicisms throughout the world. Even the French, normally known as language purists, are eager to use vogue words such as le webmestre or très cyber.
At www.euromktg.com/globstats you will find “the latest estimated figures of the number of each language population on the Internet (native speakers): those who have access to the Internet on a worldwide scale (that is, who have e-mail access) or to the Web.“ The chart is classified by languages. You will learn that 103.6 million English-speaking and 80.2 million non-English speaking people have Internet access. The biggest groups among the non-English speaking people online are the Japanese and the Spanish-speaking community (both 14.2 million), followed by the German-speaking people (13.8 million). Most of the included languages are highlighted and a click on these links shows you the countries where a given language is spoken and some details of each country’s market (population, GDP, etc.). A pleasant thing is that all data is accompanied by sources including the World Statistics Pocketbook (issued by the United Nations) and government Web sites.
According to a report by Computer Economics, Inc., China will have the second largest Internet population by 2005 (with a predicted 37,3 million users), behind the United States (126,6 million users), but pushing Japan into third place (cf. www.nua.ie/surveys).
To address the language barrier problem, AltaVista has recently introduced its free translation service called Babelfish. Whereas the earliest translators just looked up each word or phrase in the target language, Stephen Budiansky feels that most translations today are “surprisingly understandable“. To illustrate this evolution, he recounts the following story:
When the field was still in its infancy, in the early 1960s, an apocryphal tale went around about a computer that the CIA had built to translate between English and Russian: to test the machine, the programmers decided to have it translate a phrase into Russian and then translate the result back into English, to see if they’d get the same word they started with. The director of the CIA was invited to do the honors; the programmers all gathered expectantly around the console to watch as the director typed in the words: “Out of sight, out of mind“. The computer silently ground through its calculations. Hours passed. Then suddenly, magnetic tapes whirred, lights blinked, and a printer clattered out the result: “Invisible insanity“. [...] Babelfish handled that highly figurative phrase with aplomb, rendering it in idiomatic, even nuanced, French as “Hors de la vue, hors de l’esprit“... (Budiansky 81-2)
Somehow, Budiansky did not notice that the idiomatic translation would have been loin des yeux, loin du coeur. Of course, I was very eager to put Babelfish to the test and see for myself. As a trial sentence, I chose the beginning of the above anecdote, which Babelfish rendered into German like this: “Als das Feld noch in seiner Kindheit, in den frühen sechziger Jahren, in einer apocryphal Geschichte war, ging über einen Computer umher, den der CIA aufgebaut hatte, um zwischen englischem und russischem zu übersetzen“. Although the quality of the translations obviously still needs improvement, Babelfish is for some people the only chance to access foreign language documents in that it allows them to understand the gist of the text. Unfortunately, the service presently covers only a couple of European languages.
It should be mentioned that the internationalization of the Web is already in full swing. By 2003 non-English material will account for over half the content published on the Web, up from the current estimate of 20 percent, according to a report in TechServer. Foreign language newspapers are the major contributors to this growth, rapidly catching up on their U.S. counterparts. Therefore, we can anticipate a great demand for translation services like Babelfish (cf. www.nua.ie/surveys).
Keep in mind that the spiders that compile the databases are indiscriminate. Be aware that some of the resources they collect may be outdated, inaccurate, or incomplete. Others, of course, may come from responsible sources and provide you with valuable information. It is essential to evaluate all search results carefully, as far as their accuracy and currency is concerned. Spelling mistakes constitute a problem which is largely neglected, probably because they are usually easy to detect, if that. Proper names that are spelled wrongly, however, can be quite confusing. Consider this example: At www.research.cornell.edu/VPR/CWC/Spring96.html you are told that “Chinua Ahebe“ used Nigerian Pidgin in his book “A Man of the People“. Those who have never heard about African writers before and want to search for “Ahebe“ on a search engine will meet with difficulties. The text does not contain the correct spelling “Achebe“ in other passages, which would have probably helped to clarify the problem.
And then there often remains this gnawing question: How do you know when you are finished or have successfully concluded your research on the Internet? How can you be sure that you did not miss a relevant document? This can be very problematic, due to the large scale of information regarding some topics. It is unproblematic with plain questions such as “What is the name of the legal tender in Nigeria?“, of course.
Last but not least, you must maintain a high degree of discipline when conducting searches on the Internet. With an overwhelmingly large number of Web sites available on “The Information Superhighway“, it is all too easy to find yourself sidetracked and wandering away from your original topic. So the beauty of the Internet, namely the hyperlinks that take you to other servers across the globe within seconds, is at the same time one of its dangers.
2.3.2 Source Verification
The biggest challenge, however, is the fact that much of the information on the Web is unreviewed. Anybody with a small amount of technical skill and access to a host computer can publish on the Internet. This is especially true with Usenet messages (see chapter 3.3.2.1). Some sites demonstrate an expert’s knowledge, while others are amateur efforts. Some may be updated daily, while others may be outdated. As with any information resource, it is important to evaluate the material when conducting research on the Internet. As Gary W. Selnow points out: “The sites often look as authentic as the New York Times, but appearances on this medium can be painfully deceiving.“ (Selnow 173). He has this story to tell about the impressiveness of the Internet:
Look at the conspiracy sites that sprang up in 1996, claiming that TWA flight 800 was destroyed by U.S. Navy missiles. They were so convincing that even veteran newsman Pierre Salinger was duped into believing their validity when he carried their charges to the mainstream press and put his professional reputation on the line. The sites were wrong and Salinger was wrong - and he had the time and the competence to check the facts. (Selnow 173)
This story illustrates that journalists have especial responsibility to verify the information on the Web before using them in their own reports. This puts them into a difficult situation: On the one hand, they risk their credibility when for instance quoting unverified poll results posted on research sites, but on the other hand, there is an immense time pressure imposed on the reporters. Former CBS newsman Marvin Kalb said on one occasion:
On this new medium the problem compounds into a journalistic nightmare. Journalists can turn to the Internet, where there is so much information. But it is often undigested and unsourced, and even if it is sourced, what is the validity of that source? If you use it, you then give it legitimacy through your own stamp, and spread the error not only to a limited group, but to tens of millions of people around the world who then are fed bad information, bad insights, bad perspectives. The tyranny of the file is compounded in the nightmare of the Internet. (Selnow 175)
Another closely related problem is that Web sites almost always reflect the author’s inclination. Here again, it is the users’ responsibility to judge the validity of the information. Charles Kuralt, another CBS newsman, told the following anecdote about the interpretation of facts:
I was reading a piece about Galileo’s dispute with the Catholic Church, and I thought it was a bit different from what I had remembered. It put the Church in a little better light than I recall. I discovered at the end of the site that the article was written by someone from the Vatican. (Selnow 173)
The users of the Internet must always be aware of the fact that the documents on the Web are unreviewed. It is thus in their own interest that they never forget to ask questions like: Who is the author? (does he or she has expertise on the subject as indicated on a credentials page?, is he or she a native speaker?). And who is the author’s target audience, or expressed differently, for which purpose did the author write the article? It is usually quite helpful to see who the “sponsor“ is. Educational institution sites are usually more reliable than personal home pages. Bryan Pfaffenberger points out that in academic circles, information does not get published unless researchers in the area judge the document to be worthwhile. This is called peer review (cf. Pfaffenberger 3). Some home pages contain links to resumes. You may find out that the page was created by a 19-year-old college student with too much free time, or by one of the leading experts in this field. It is often a good idea to contact the author more detailed information. Gaga Ekeh comments on his compilation of “Some pidgin sayings and their translations“ (cis.upenn.edu/~ogunyemi/pidgin.html):
Those sayings on the page are more humorous than “syntactically“ correct (pidgin-wise that is). But the reason Nigerians find them humorous might be of interest. In a place called Warri, in Mid West Nigeria, the pidgin dialect is continually morphing. It almost seems as though the “waffarians“ as they are known, are in an eternal quest for simplicity, using as few words as possible to explain a concept in pidgin. For instance, the phrase: “No face, everywhere tinted“ refers to someone wearing large, dark shades. “Everywhere tinted“ is something of a reference to the cars, mostly military, with tinted windows, not revealing the face of the owner. So, in Warri, you will actually hear people use such phrases, but it’s not widespread or accepted. It’s kind of like underground hip-hop, that’s what Warri or “waffi“ pidgin is like. So, you might not be able to use my sayings as being generally accepted pidgin... (Ekeh in his e-mail from April 4, 1999)
If I had not contacted Ekeh, I would not have obtained this valuable extra information. Or I would even have cited the sayings as standard idioms, which could have given a false impression of Nigerian Pidgin.
One exemplary site that contains peculiar statements is accessible at www.geocities.com/Athens/Crete/6503/index.html. It claims to “give an insight to that type of English spoken in West-Africa“. There are six Web pages altogether: “Pidgins“, “Creoles“, “West-African English“, “Other Sources on the Web“, “More about the funny World of Linguistics“ and “The Books we used“. On the two last pages we learn that this site was created on the basis of a linguistic project of the Technische Universität Chemnitz, and we are given the titles of seven reference books about pidgins and creoles. So far, so good. Two things on the other pages struck me immediately:
1.) At one time, the authors claim that pidgins and creoles are “no second class languages as they were and still are the only mean [sic] of communication in some parts of the world. Even today there are considered to be between six and twelve million people still using Pidgin languages“, but later they claim that “Nowadays WAPE [= West African Pidgin English] in Nigeria has more than 20 million speakers.“ Not only is the English peculiar, but also the numbers quoted are contradictory.
2.) The authors’ assertion that “[s]peakers of pidgin languages prefer to use long and latin [sic] words“ is even stranger. The examples they supply are epistule for letter and purchase for buy. First, the authors should have known that by far not all pidgins have European languages as their source languages, and secondly, even closely related variants of contact languages possess a quite different vocabulary stock, so that generalizations simply cannot be made. Mark Sebba, whose books the authors allegedly used for research, says himself: “I don’t see how you could generally verify a statement that Pidgin speakers like to use long and elaborate words. It may be true for certain registers of Nigerian Pidgin - it seems to be part of the stereotype evoked by some of Wole Soyinka’s pidgin-speaking characters, but I haven’t done any research on this. [...]. It would definitely not be generalisable to all pidgins.“ (Sebba in his e-mail from April 14, 1999)
To find more about GeoCities, I chopped off all parts of the URL that come after the www.geocities.com. I was then taken to the GeoCities home page and I discovered that it is a company that provides free personal home pages in one of its 41 themed communities to anyone with access to the Web. These communities are called Neighborhoods and include “Athens“ (education, literature, poetry, philosophy), “BourbonStreet“ (jazz, Cajun food, Southern culture), “Pentagon“ (military men and women), “Pipeline“ (extreme sports) and “Vienna“ (classical music, opera, ballet). Although in principle I very much welcome the idea of promoting home pages, it certainly incorporates the danger that anybody can publish material on topics he or she is not necessarily familiar with. It is also annoying that GeoCities’ sponsors consistently bombard the visitors with advertisement banners that load automatically each time you click on a link.
3.1 Possible Approaches
When you search for a specific topic, you not only can employ different search strategies, but also use different parts of the Internet. The most common way to find material is to enter key words into the search box of a search engine. Another common approach is to browse a subject directory in the hope of finding a quality link site. Consequently, search engines and subject directories are also called Web portals. In addition, I will examine the effectiveness of searching newsgroups, mailing lists, and above all, contacting experts.
I will analyze a large portion of what can be found on the Internet about Nigerian Pidgin in terms of quality, accuracy and usefulness. I will start with the traditional search via search tools in chapter 3.2. and then go on to introduce the other starting points in chapter 3.3.
3.2 Analyzing and Comparing Search Results
Results from Individual Search Tools
3.2.1.1 Phrase Search “Nigerian Pidgin“
These were the number of Web pages found when I searched three of the most popular search engines for a) nigerian pidgin (simple key search), b) “nigerian pidgin“ (phrase search), and finally c) „nigerian pidgin english“ on March 30, 1999:
search toolnigerian pidgin“nigerian pidgin“„nigerian pidgin english“Yahoo!/Inktomi67135HotBot190
33
(English items: 13)15
(9)AltaVista56,18566 (64)30 (29)
The search result with AltaVista is breath-taking, but note that almost all of the 56,185 Web sites retrieved in my search for nigerian pidgin include either the word “Nigerian“ or “pidgin“. I will first focus on the sites found when employing the phrase search “nigerian pidgin“ and then, in the following chapter, analyze whether you are likely to miss relevant matches when using the more specific key word “nigerian pidgin english“.
I would like to begin with Yahoo! Since the topic of Nigerian English is still too specific for Yahoo!, the search results discussed here were provided by its search engine partner Inktomi (cf. the chapter on search tools). I will thus treat Yahoo!/Inktomi like an ordinary search engine in this study. Yahoo! found the smallest number of Web pages. I will analyze whether these pages are, as it might be assumed, top quality Web sites in their majority. The first ten Web sites recommended by Yahoo! were:
Although Yahoo! retrieved 5 matches less than on 25 February, its collection seemed to have improved. The document “Nigerian Languages“ was added to the database and the “Regional Listing of All Translations“ is now in second place. Oddly enough, the article “Pidgin Vs. Rotten English in Soyinka and Saro-Wiwa“, in 3rd place only a month ago, has disappeared from the list altogether. Later we will see that it is in 2nd place on Google’s list. I will discuss the weaknesses of this page in chapter 3.2.2.
Before I go on to discuss the material, let us have look on the first ten search results of Hotbot in order to see how many overlapping results it returns:
The first three matches on Yahoo! are also included in HotBot’s top ten, though ranked in a different order. Some of the other matches are listed in the lower ranks, e.g. “Nigerian Languages“ is in 17th place and “Alphabetical Listing of All Translations“ is in 19th place on HotBot. The others, however, are unique to the respective search engine, so we can say that it makes sense to query more than one search engine. Interestingly, “ Pidgin Vs. Rotten English in Soyinka and Saro-Wiwa“, in 5th place a month ago, was also removed from HotBot’s list of 33 matches. HotBot’s top ten list has not changed as much as Yahoo!’s did: the four first matches were the same, Shana Poplack’s home page moved from place 6 down to place 9. The most important questions are, of course, which Web pages were to at least some extent useful, which pages were completely irrelevant, and which search engine yielded more relevant results?
Yahoo!’s top match “Contents“ is maintained by the Journal of African Music and Popular Culture of the African Music Archive at Mainz University, my alma mater. Basically, it deals with songs by Nigerian musician Fela Kuti who mixes up elements of Standard English, Nigerian Pidgin and Yoruba. Four links contained the term Nigerian English in the title: “Typical Features of Nigerian Pidgin in Original Sufferhead“, “Selected Problems of Fela Kuti’s Language as a Mixture of Nigerian Pidgin and Standard English in Fear Not For Man“, “Features of Nigerian Pidgin in Fear Not For Man“ and “Aspects of Tense in Sorrow, Tears, and Blood with Respect to the Aspect and Tense Markers in Nigerian Pidgin“. The lyrics to the three above songs are given in both Nigerian Pidgin and Standard English. Here is a selection of excerpts of the linguistic approach to Fela Kuti’s lyrics: 1.) “The language of Fela Kuti's lyrics consists of at least three languages. [...] It is more difficult to evaluate substratum influence regarding grammar, for one has to know the indigenous languages well. Yet, this is to a great extent what the linguistic research of pidgins and creoles is about. 2.) “There are good examples for the use of /wetin/ and /wey/ which are typical NP words. /Wey/ functions as relative pronoun. This is obvious in sentences like those wey dey for London and we wey live for Africa. /Wetin/, which can also be a subordinative conjunction in [Nigerian Pidgin], appears in the phrases Wetin do them and I go know wetin. Bearing the meaning “what“ in both cases, /wetin/ is an interrogative pronoun in Original Sufferhead. In general it seems to take the place “what“ holds in Standard English. Another striking feature in Original Sufferhead is the reduplication of words, especially English loan-words. This is a significant morphological process in Pidgin English, a process of word formation that can be a process of lexical modification, too.“
From this we can say that “Contents“ was rightly ranked very high. For one thing, it makes the visitors aware that apart from writers who use Nigerian Pidgin, there are also musicians who employ it in their lyrics. But above all, it contains much information on the language itself. The site is naturally of special interest to African music fans, as it lists a huge number of other African music resources online, including three newsgroups. Note that “Contents“ is a prime example of how vague and useless titles of hyperlinks can be!
Yahoo!’s matches 2 and 7 both eventually can lead to Nigerian Pidgin version the Universal Declaration of Human Rights, a particularly valuable item. However, only the search engine Google retrieved a direct link to it, so that I decided to discuss that Web page in chapter 3.2.2.
“Nigerian Languages“ gives a brief description of Igbo, Yoruba, Hausa and many more as well as of Nigerian Pidgin, providing information on for instance genetic affiliation and the regions where those languages can be heard. The reader learns that Nigerian Pidgin is partially intelligible with Krio of Sierra Leone and Cameroon Pidgin, and that the Bible was first printed in Nigerian Pidgin in 1957; that it is a creole with native speakers, as well as used as a pidgin between Africans and Europeans, and Africans from different languages; that there is no unified standard or orthography; that it is used in novels, plays, radio, poetry, advertising, and that its importance and use is increasing. This is one of the very few informative sites on Nigeria Pidgin, so it surprises me that, as mentioned previously, it has not been indexed by Yahoo! until recently.
Match 9 with the title “thank you!“ is a list of all the people and organizations that have contributed to The Four Essential Travel Phrases. The title of this project aroused my curiosity and I clicked on the “home page“- button. The site offers the four following sentences in more than 2,000 languages (including Esperanto, Greenlandic Inuktitut, Jamaican Creole English, Latin and Yoruba): 1. Where is my room?, 2. Where is the beach?, 3. Where is the bar? and 4. Don’t touch me there! The Nigerian Pidgin English is: 1. Wey my room dey?, 2. Wey de beach dey?, 3. Wey de bar dey? and 4. Make una no touch me there o!
Africa - Language/Linguistics is a link site compiled by Stanford University. It provides links to e.g. the Ethnologue and the Hausa Home Page. The one that relates to the topic of Nigerian Pidgin is “Some pidgin sayings and their translations“ by Gaga Ekeh. Yahoo!’s link numbers 3, 4, 5 and 8 were lists of publications on all kinds of linguistic aspects (not only on Nigerian Pidgin). While such lists might inform you about valuable reference books on the market, they usually do not supply any online research material.
Oddly enough, a commercial site is HotBot’s first choice: “CA Webhotel - Info: Pidgin and Creole TMA Systems, Singler (ed.)“ presents a book in which seven pidgin languages, including Eighteenth Century Nigerian Pidgin English, are analyzed. The reader is provided with an online order form, but regrettably he/she does not get to see any excerpts from the book.
Matches 4 and 5 are links to the same document, but with different URLs. It is another commercial site, introducing Nicholas G. Faraclas’ book “Nigerian Pidgin“ as “the first comprehensive grammar of Nigerian Pidgin. [It] provides basic descriptive and analytical treatment of the syntax, morphology and phonology of a language which may soon become the most widely spoken in all of Africa.“ This is everything the prospective buyer is told as far as the contents is concerned. Routledge, too, furnishes an online order form.
Match 8 on HotBot is the contents page of a 1986 issue of the Journal of Pidgin and Creole Languages. This obsolete page was retrieved, because it announces a review of the book Social and Linguistic History of Nigerian Pidgin English: as Spoken by the Yoruba with Special Reference to the English Derived Lexicon“ by Anna Barbag-Stoll.
Matches 9 and 10 are by pure accident connected. In 9th place, we find the personal home page of Shana Poplack, full professor from the University of Ottawa. In a way, such home pages serve as online visiting cards, mostly with special focus on qualifications and publications. We learn that Professor Poplack has published for instance the paper “Nothing in context: Variation, grammaticization and past time marking in Nigerian Pidgin English“. This information, however, is irrelevant for those people who resort to online material only. Match 10 with the mysterious title “put title here“ presents a concise summary of a paper by Tagliamonte, Poplack and Eze about the pluralization system of Nigerian Pidgin.
None of the links that came after match 10 focused on Nigerian Pidgin; many were only bibliographies of linguistic literature. In 21st place on HotBot we find the Nigerian Pidgin translation of an information brochure of a “non-profit, non-governmental organization, which strives for environmental solutions.“ (www.pearson-college.uwc.ca/pearson/gaia/brochureni.htm). This page is of interest insofar it supplies the second longest sample of Nigerian English I found on the Internet.
Online documents about Nigerian Pidgin itself are hardly available and, provided you find some, not very detailed. “Nigerian Languages“, for example, comprises no more than seven lines. Some items are retrieved because they contain sample sentences; many matches are shopping sites, others tables of contents, but irrelevant to the online searcher.
Please note that unlike with Nigerian Pidgin, a topic which is obviously not widely-covered on the Internet, it is possible and necessary to specify your search for more popular languages such as German: While the simple search on AltaVista found more than 4 million matches for “german“, the advanced search for the key word “german“ and the Boolean expression “morphology AND inflection NEAR noun“ significantly narrowed down the results to 48 matches, the first one entitled “Morphology of German nouns“!
What about the 66 Web pages that AltaVista recommends us? How many new relevant will a third search tool return? Will the results be listed in a more logical order? Based on my experience, I expect this long “hit list“ to be loaded with irrelevant pages, many of which we have already found on Yahoo! and HotBot. Let us now examine whether this assumption is biased or justified. Here is AltaVista’s “top 10“ as of March 30, 1999:
It is worth noticing that the Web page “CA Webhotel - Info: Nigerian Pidgin, Nicholas G. Faraclas“, which occupied the highest position on AltaVista’s list in February, has disappeared entirely from the list. Many other pages from the Creolist Archives Web site, however, are still included, e.g. match 5, which is about the Creolist Archives’ netiquette policy. I will discuss the Creolist Archives Web site in chapter 3.3.1.
Match 6 was a paper on the intercultural nature of modern English, in which the term “Nigerian Pidgin“ appeared only once as an example for a variety of English. Neither did match 7 deal with Nigerian Pidgin itself. It was retrieved because the author cites Poplack’s and Tagliamonte’s paper “ Nothing in context: Variation, grammaticization and past time marking in Nigerian Pidgin English“. Match 8 happened to be a list of Tagliamonte’s publications. Match 12 showed the table of contents of the April 1997 Journal of Pidgin and Creole Languages issue. Unfortunately, the online version of this magazine does not offer papers in their entirety, but only provides brief summaries.
With some frustration we notice that many links do not relate to Nigerian Pidgin at all. Moreover, it seems that Yahoo! specializes in retrieving several Web pages from the same Web site, which accounts for the length of its list! Best example: Various pages from the Journal of African Music and Popular Culture site are to be found in the places 9, 10, 11, 13, 15, 16, 17, 18, 41, 55 and 59. Remember that Yahoo! returned only the “Contents“ page which lists all relevant links. As for this query, AltaVista outdoes the other search engines in quantity, but at the price of quality and clearness.
Not only is the presentation unclear, but also the order of links gives rise to criticism. Many people tend to look at the first ten or so matches, especially when querying several search engines. Due to the listing of various items from one site, the Nigeria-specific page from the Ethnologue site comes as low as on position 24, despite the fact that it contains more relevant information than almost all other documents that are ranked higher. Since AltaVista’s top ten did not exactly yield much usable material, I guess few people are tempted to browse any further and will thus miss this page. Astonishingly, this page was in first place when using the advanced search (cf. next chapter).
Some Web pages are useless not primarily due to their contents, but simply because of technical problems. “IRIS sound database“, AltaVista’s top choice and HotBot’s third match, claims to have some Nigerian Pidgin texts in stock, but when clicking on those links, the computer informed me that “The requested URL could not be retrieved“. Another point is that it has not been modified since December 1996, which is an unusual long period for Web sites. Most dead links inform you about the infamous Error 404: “Not found. The Web server cannot find the file or script you asked for. Please check the URL to ensure that the path is correct. Please contact the server’s administrator if this problem persists.“
3.2.1.2 Phrase Search “Nigerian Pidgin English“
In this chapter, I will examine whether you are likely to miss relevant matches when using the less general key word “nigerian pidgin english“, as in many texts only the term Nigerian Pidgin is used. The query “anglo-nigerian pidgin“ retrieved hardly any documents and are therefore not dealt with in this study. On Yahoo!, the number of results were reduced to these five pages:
Matches 1 and 4 are pages about the Declaration of the Human Rights and match 2 is one of the pages that is maintained by the Journal of African Music and Popular Culture of the African Music Archive. Like the others, matches 3 and 5 were analyzed above and considered quite useful. So finally I was returned a list that contained only useful links!
HotBot retrieved the following list of 15 matches:
As expected, HotBot’s list was longer and less clear. In fact, it contained many documents of no relevance for this study. In contrast to Yahoo!’s list, however, it included the second longest sample of Nigerian Pidgin that I found on the Internet: “brochurenig“ (cf. chapter 3.2.1.1). Consequently, it depends on the purpose of your research whether you are satisfied by a concise quality list or a more extensive list that includes irrelevant pages. Note that it makes no sense to submit the query “Nigerian Pidgin AND sample“, as the documents in question usually do not contain the word “sample“!
AltaVista retrieved no less than 30 matches. Of the first ten results, match 1 (&#
