Elitism on University Websites

(1)

6/29/2018

Elitism on

University

Websites

AUTHOR: Larissa Tijsterman

E-MAIL: Larissa.tijsterman@hotmail.com

STUDENT NUMBER: 10218661

FIRST READER: dhr. prof. dr. F.J.M. Huysmans

SECOND READER: dhr. dr. B. Rieder

PROGRAMME: Research Master Media Studies

TRACK: Culturele Informatiewetenschap

(2)

1. Introduction

For most prospective students the university website is one of the first places to go to find out more about the institution’s programmes. Universities spend a considerable amount of time on designing their web pages to not only suit the current users, often students, staff and those affiliated, but also to attract potential new students. Although beautiful campus pictures, high-quality video and stunning layouts are important, the contents or better said the text is fundamental to bring about the message the university which convey to the prospective students. Programme and course descriptions are carefully worded in order to not only transfer information but also to inspire and excite its readers. The website texts, often produced by the university’s communications department but also by teaching staff themselves, always have an intended audience in mind. This does not just constitute the prospective student but to a certain aspect of these students such as creativity, analytical, ambitious, exploring, factual, etc. The language the universities use on their websites can impact the decision prospective students make in their choice of university.

From first-hand experience, I can vouch that the language used to describe two similar university programs helped me to make the decision about which university I wanted to attend. As both universities were in the same town and had similar prestige and programs the language used in the description of the programs was vastly different, where one university described its program in a very formal matter, used technical jargon and boasted about famous authors and while reading it I felt pressured and unworthy to join that university. Whereas the other description had a more casual tone, used simple and emotive language and addressed the prospective student's feelings. The second description felt more welcoming and reassuring: even if you didn't have any previous knowledge on the subject but you are just interested you would fit right in. Having chosen the second university, I still to this day do not regret it one bit. My own experience sparked my interest in how language use on university websites creates a certain image.

Universities take great care in the text they put online on the website is one of the main public communication channels. Recently Yale University, one of the top universities in the world, decided to replace the words ‘freshmen’ and ‘upperclassman’ for more gender-neutral names in all their formal communication which includes the website. ‘Yale began to consider the language change last year amid growing calls for greater gender inclusivity on campus’ (Hamid). Clearly the university was concerned about the way the language they used portrayed the image of a university not being inclusive enough to all genders and they are not the first university to address this particular problem, in 2009 the University of North Carolina scratched the word ‘freshman’ from all its official documents and in 2015 the University of Emory did the same.

(4)

Research on university websites has only just started, however, the main focus often lies on how the websites comply with the Web Content Accessibility Guidelines 2.0 requirements. The WCAG 2.0 was set up by the Web Accessibility initiative which is part of the World Wide Web Consortium in order to set international guidelines on how web pages can be made accessible for everyone including those with a disability. ‘The power of the Web is in its universality. Access by everyone regardless of disability is an essential aspect’ (Berners-Lee). The WCAG 2.0 sets out 4 principles with an explanation of how a website’s accessibility can be increased. Their main focus lies in making websites accessible for people with a disability, the four principles; perceivable, operable, understandable and robust serve as an evaluation/testing tool for websites. Perceivable, operable and robust are mostly concerned about the technical aspect of a website, understandable, however, focuses especially on the text. Karhu et al. are one of the first to combine the accessibility aspect with the readability concept concluding that web text from several universities in Finland has a level of readability from hard to very hard. They used the Flesch-Kincaid readability test in order to find out for what reading level readers would have to have in order to understand that text, the scores from the Finland universities come in at understandable for college (hard) or college graduate (very hard). This is both surprising and unsurprising. The pages selected for their research are the front page and then a page containing a form and a page containing a table, both the second and third pages are pages contain information about programs, contact information and admission information, information for prospective students and current students whose reading level is in 12th

grade or developing towards a graduate level. Sadly Karhu et al. do not give any insights into why the readability level is high or how it attributes to the accessibility of the university web pages. As the readability formula has been used in the context with the accessibility of the universities website text the suggestion here is that a difficult level of readability will reduce the accessibility of the website as the text is difficult to understand.

Universities have entire departments dedicated to communications. Part of their job is the careful review of any text that is published under the university’s name, not just screening for grammar errors but also judging whether or not the texts are in line with the “house style”. Each word has been deliberately considered, each phrase carefully designed in order to get the message across. Yet these texts are found to be even ‘more “literate” than textbooks’ (Biber 189). Why is it that these texts are deemed difficult to read? Are universities showing off their language skill or do they want to attract only students who already understand such language?

Readability formulas came to live in the 1920s as a tool for teachers and later publishers to evaluate the proper reading materials for young students, as their popularity grew their use quickly expanded from textbooks to law text to public health information.

(5)

These days readability formulas are integrated into almost all popular text editors and form an important element in search engine optimization. The access to an understandable text can move entire communities and even countries. For example with Tyndale’s bible which was the first Bible to be translated into plain English from Hebrew and Greek in the 1520s. As the Bible was usually written in Latin and ancient Greek, these languages were only taught to the highly educated people connected to the church. The more common middle class that could read, read English and maybe a little Latin, would only know the contents of the Bible through their ministry. Tyndale’s bible allowed those able to read English to interpret the Bible by themselves which started a great breach within the Catholic Church. Apart from being an example of the power of translation, it is also an example of language use that limits and expands the accessibility of a text that results in a change in perception of large institutions, such as the church, that have a prominent position in society.

To make a bit of a leap, universities have a prominent position in society as well. Universities are producers of knowledge and next generation of highly skilled people. Through centuries of existence, they have developed their own register to communicate. The academic register recognized by its complex structures is difficult to master therefore its mastery is a sign of accomplishment which allows one to move up the ladder in academia.

As readability formulas can predict what level of education is necessary for someone to understand a text, they are the perfect tool to estimate how complex university web texts are. Complex text can cause certain readers to be turned off from engaging further with it, while those used to reading complex texts, often higher educated, will be less deterred. In order to attract the best and the brightest universities might purposely use complex language in their programme and course descriptions as a method to get the cream of the crop of prospective students. How then can the complexity of web texts of programme and course descriptions, as indicated by readability formulas, contribute to the prestige and elitist character of ranked universities?

Drawing on the work of French sociology specifically that Bourdieu on power, high culture and prestige language this study will explore how academic language use contributes to complex language use within universities, how ranked universities are seen as elitist and how ranking contributes to the prestigiousness of universities. The corpus will consist of program and course descriptions from the Top 100 ranked humanities department of the Times Higher Education (THE) as it is not only a widely used ranking system for universities with a very open and clear methodology but also allows to sort on subject creating rankings specialised for each field. Because THE allows for sorting on the subjects the choice for Humanities has been made in order to avoid the technical

(6)

jargon of the STEM courses. As readability formulas have different variables concerning the complexity of text I will be looking at the average outcome of the eight most popular used; Linsear Write formula, Flesch reading ease, Flesch-Kincaid grade, Dale-Chall score, Coleman Liau index, Gunning Fog index, SMOG and the Automated Readability index.

Outcomes from this study cannot only help universities to better understand how their language use on their websites matches the reading levels of their intended audiences but also provides a method and valuable insights of the use of readability formulas on crawled texts. Ultimately this studies hopes to contribute to a better understanding of the use of readability formulas by expanding its use from structured well-formatted text to unstructured messy web-based texts.

(7)

2. Theory

2.1 The Complexity of Academic Language

The transfer from secondary education to higher education can be quite daunting. Not only do young students experience a drastic change of environment and people they are also confronted with a whole new language register. Many university programmes offer special courses, often mandatory, called academic skills or academic writing where students rapidly come to grips with the academic language they are getting confronted with in their studies. As these courses are often designed to not only teach students important research and formatting skills but they also teach how academic texts are written and should be read. By offering these courses universities acknowledge that the academic register is a language different from everyday registers and most working/industry registers.

Research into the academic language did not really kick-off until the late eighties / early nineties when the need for descriptive research on the academic register and characteristics arose for designing courses such as academic writing (Biber 6). From there on written academic language has been researched extensively yet as Biber points out ‘focused on scholar-to-scholar written communication, rather than on the types of discourse encountered by and used by students in colleges and universities’ (8). Biber’s study is one of the first large-scale studies that takes in all aspects of scholarly to student communication that looks at the entire spectrum of communication. His corpus consists of not only written communications such as textbooks and assignment instructions but also verbal communications such as lectures, discussion and tutor meetings. Biber also included online communications such syllabi, brochures and course description. Although his research is done in the early 2000s the inclusion of digital and web-based text already displays their importance as a communication tools that carry the academic language beyond the scholar to scholar communications.

‘Many of these texts are among the first material that a prospective student receives from a university, either through paper copy or on the Web: informational brochures about student services and academic programs, university catalogs (including both discussion of general requirements and specific programs), etc. Although not often considered “academic discourse”, written material of this type is ubiquitous on campus and required reading for the prospective student attempting to navigate the maze of university requirements and services’ (Biber 26).

(8)

Biber already realised the importance of those online texts by including them in his corpus. However, he also makes the remarks that they are often not considered ‘academic discourse’ but they are vital for prospective students to ‘navigate the maze of university requirements and services’ (26). He might not consider them academic discourse per se but he does indicate that they contain vital information for the prospective students. One would expect that these texts then are written clear and concise as to not confuse the prospective students on both the content of the programmes as well as their requirements. If these texts are written in academic prose then they might become an obstacle for students to read as they are not used to reading such complex text, which in turn might influence their decision for their choice of university. Some students can even be excluded from the language use as they potential fail to follow the correct instructions for the application or even understand the requirements resulting in their inability to complete a full application. In a potential extreme case, such difficult and complex language can be used in such a way that its purpose is not the provide information but as a façade creating an illusion of information being provided while in fact, the correct information has to be obtained through another source.

When Biber breaks down the characteristics of his corpus he finds that the institutional writing which contains texts such as university catalogues and brochures share more characteristics with the academic textbooks than anticipated. Biber expected that these texts would be ‘highly accessible’ yet had to conclude that ‘institutional writing is more complex than any other university register’ (50). He found that Institutional writings mostly consisted of noun phrases with very little verbs and clauses which is quite difficult to read. Unfortunately, Biber does not go into greater detail on why institutional writing is more complex than any other university registers, in fact, he spends very few words on describing which characteristics make these text so complex. Further on Biber comments specifically on institutional writings when he is analysing expressions of stance in chapter 5 (127-31). Here he concludes that institutional writing often switches between personal and impersonal style which he contributes to the fact that often these texts have multiple authors before they are being approved which can clearly be felt while reading the texts as they feel somewhat ‘schizophrenic’ (Biber 130). Then he comes back to institutional writings one more time when he compares the different registers on several linguistics features in order to see how far or how close the registers are alike. This type of research, also known as multi-dimensional, is a quantitative linguistic method that takes two types of opposite linguistic features by which different registers can be measured based on a score of various linguistic characteristics of which each have been assigned either a positive or negative number. Based on the outcome for each register, these registers can then be laid out in a line where either end makes up the opposite linguistic feature. Biber’s first dimension is the

(9)

distinction between oral or written discourse, both opposites of each other, is the last time he makes explicit remarks on institutional writings being more complex. Although it was not a big surprise that oral registers scored high on oral discourse and the written registers scored high for written discourse, within the written discourse institutional writing scores the highest of all written registers. As the institutional writings contain mostly university catalogues and program brochures it is remarkable that these as Biber puts it ‘are even more “literate” than textbooks’ (189). He found that programme brochures contained a ‘dense amount of nouns and nominal modifiers … which occur closely together to build very complex noun phrase structures’ (Biber 189). Biber does not provide any explanation why these texts are complex as his research is more focused on describing and comparing these different university registers. However, he does leave his reader with the following statement:

‘In fact, the linguistic style found in many university catalogs and program brochures is often more reminiscent of highly technical academic prose than textbooks written for novices in an academic discipline’ (ibid).

If programme brochures are already more complex than academic textbooks how then can prospective students get a good idea what a programme is about let alone compare the same types of programmes of different universities? Unfortunately, since Biber’s study, specific research into the complexity of programme descriptions is non-existing. Biber seems to be the only one engaging with these specific texts whereas other research into scholar/university to student communications seem to focus on the acquisition of academic language by students or look at language use of students from different backgrounds. Especially in scholar to student communication language studies seem to have the same goal as Biber that of describing how academic discourse comes into being and how it is used but never why.

Perhaps Bourdieu’s Homo Academicus and The State of Nobility can provide us with the why. Published in 1988 and 1984 Homo Academicus describes the intricate structure of higher education as a field. Bourdieu found strong links between language, identity and worldview. He showed that people in ‘tenured’ positions were more likely to participate in the election compared to those of more wide-ranging professional careers ‘a participation which implied an already established pattern of social conformity’ (Grenfell 41). In The State of Nobility Bourdieu looks specifically at these connections within the higher education setting and how language plays a vital role in it all (ibid). He argues that language reinforces social selectivity, however, this is not necessarily done by explicit

(10)

identity statements such as “you belong to this group” but more subtle by using “euphemization”: ‘the same judgement is delivered, but it is done so in ways and phrases which fane objectivity’ (ibid). So certain words will carry values that enables them to judge without seeming to be. However, the taxonomy of these value words are different for different groups and different circumstances, in fact, they ‘are based on cognitive relations which are formed in childhood, reinforced in adulthood, and heavily laden with the values and interests of particular factions in the social hierarchy’ (ibid). Bourdieu was not alone in this thought, Basil Bernstein a contemporary of Bourdieu researched language acquisition with children of different classes and found that the register and manner in which one talks determines his or her place in society and also the place one finds itself in society dictates the register and manner of speech (Bernstein 165). The family life is the most important place where children learn their language but it is also the place whereas Bernstein puts it: the place of the most restricted code.

Language at home is very context driven and draws extensively on shared knowledge, therefore, fewer words and sentences are needed in order to transfer knowledge. However, this is not restricted to one class, in fact, it is universal (Bernstein 173). Restricted code is also not limited to the home either, almost all people will use it in various places with various people but all will have in common is the familiarity in the group with a strong shared knowledge such as a group of employees that have worked together for a long time or a group of good friends.

The counterpart of restricted code, elaborate code, is the key to what divides the classes or better said the amount of exposure. Elaborate code as opposite of restricted code uses more words and sentences in order to communicate, it does not presuppose any shared knowledge or context and therefore, everything has to be elaborated in order to successfully communicate. For example, law texts or in education elaborated code is used extensively as the target audience is often unknown or deemed to have no prior knowledge on the subject. The elaborate code just like restricted code can be learned by anyone, however, the access to elaborate is much more restricted. ‘Elaborated codes give access to alternative realities yet they carry the potential of alienation of feeling from thought, of self from other, of private belief from role obligation’ (Bernstein 176). Because elaborate codes need to be taught they cannot be assimilated solely in a familiar setting such as at home, they require contact across different language registers and fields. Bernstein found that when asked questions lower class children hardly use any elaborate code whereas middle-class children of the same age use quite some elaborate code in their answers (167). Indicated by longer sentences and more words elaborate codes are already a first boundary that distinguishes higher classes from the lower classes. Bourdieu however, goes further to not just look at the complexity of the language but how boundaries are drawn up within the language (Grenfell 42). But where Bernstein spots the issue

(11)

as a vicious circle where class determines language and language determines class, Bourdieu holds the power relations responsible for the upkeep of class through language. Through the education system elaborate language is being reproduced from the top down, those that are in power, the elite, are also the ones that have the most interest and the most to gain from the education system as a place for reproduction of high culture.

Academic language has been researched extensively the past couple of decades in order to provide better training for starting students as well as confirming long-held believe that academic language is complex and difficult. Biber’s study showed that academic language is very prevalent not just in scholar to scholar communications but also scholar to student and university to student communications. Most interestingly is the fact that within his corpus the written texts where the most complex forms of academic language resided were those of programme catalogues and brochures were taken partially from the universities websites. Although Biber does not provide an answer Bourdieu and Bernstein agree that complex difficult language serves as a boundary for the class difference with here, in particular, the boundary of the higher class. Considering that the target audience of these text are prospective students who have little to no experience with reading academic language the strong presence of academic language in these texts seem to function as a gate-keeper for the university, an early preselection where those that are not familiar with such complex language are discouraged while the university keeps up the appearance of openness as the description are available for all.

2.2 Construction of Elitism – Bourdieu, Cultural and Academic Capital

and the Rigid Hierarchical Education System

If complex language use serves as a boundary for the higher classes and it being present in university texts intended for prospective students then what does that say about universities? For a better understanding of why universities use complex language in their programme and course descriptions, one has to turn to Bourdieu. Bourdieu has done extensive research on the French higher education system and how they contribute to the construction and maintenance of elitism and class divide (Martin 20). He showed that class inequalities were reinforced by the institutions themselves. Students coming from a lower class background are at a disadvantage compared to students from a higher class background because they often lack the support at home, often hold part-time jobs next to their education, often had less pre-academic training and often engage less in high cultural events (Martin 28).

According to Bourdieu, our whole society is made up by a multitude of social structures and fields in which individuals ‘are productive agents who produce the structures they need to safeguard the

(12)

originating social condition’ (Robbins 61). Each individual has a habitus, which holds the knowledge of the social structures such as norms and expected behaviour, instilled on them partially by inheritance and partially by investment in social relations. Inspired by Marx’s ideas of capital, one of the most important parts of the habitus which is also responsible for class differences is the cultural capital.

Cultural capital is not capital in the physical sense such as money or material, rather it is the cultural tastes, styles and traits which are reinforced by shared conventions (Lareau and Weininger 574). Cultural capital is literary the culture that an individual carries with him and uses it to navigate the social structures. Acquiring and accumulating cultural capital, however, is much more elusive; part of it is internalised knowledge of social structures which is inherited and a part is externalised which is something you pursue, nothing is therefore set in stone but is rather always on the move (Lareu and Weininger 578). The external accumulation of cultural capital is often being linked to the participation of cultural activities which for example for the higher classes, visits to museums or attending theatre making them quantifiably unites which researchers can easily measure. However, it is not specifically the act itself but the interaction which occurs. Cultural activities allow an individual to accumulate the know-how of a social structure by studying its traits, tastes and styles (Robbins 32-7). It is these traits, tastes and styles which define status as social groups reject some while adopting others. For the boundary of the higher class the rejection of tastes because they are deemed inappropriate is just as important as the shared tastes (Peterson 87). The type of traits, taste and styles of the cultural capital are different for each field which is ‘a space of positions, a network of relations, and especially an arena of struggle for conflict overvalued forms of capital and symbolic power’ (Martin 44). As tastes of the cultural capital for the higher classes are taught in higher education the academic field, especially those in the humanities, is undoubtedly connected to that of the higher classes.

One possible view on why academic language is used for programme catalogues and brochures is that university may be used it (un)intentionally to display prestige. As Biber already stated these texts are often written by committees within the university which most likely consist of highly educated staff who use academic discourse on a regular basis. The use of complex language that is perhaps too difficult for the majority of prospective students and thus can serve as a boundary or even a pre-selection as students can be turned off by language that is hard to understand or as in my own experience makes students insecure about their level of intelligence. The brightest students might already have had encounters with academic language through either their advanced studies or through their social surroundings with for example highly educated parents or private tutors, they

(13)

will be less put off by the complex language. By using complex language as gate-keeper universities create exclusiveness without explicitly stating it.

Historically universities have had a special place in society, as producers of knowledge, keepers of information while educating the next generation and because of their intellectual production in most communities universities and those associated with it are given high status (Peterson 76). Universities are the places where the next leaders are being groomed, where social ladders can be climbed and where world-changing ideas are being born. However, the education system originally did seem to serve the elite. However, with the education reforms of the 19th_{and 20}th_{-century citizens}

in the western world almost all children from all backgrounds gained the equal right to education which opened doors for many lower class people. Although primary education had levelled the playing field inequality remains in the form of private schools, private tutors even homeschooling contributes to inequality also advanced education contributed to the inequality as it had not changed altogether much.

‘The educational system serves to eliminate most working class students before the postsecondary level. University admission is only the most recent of a series of eliminations and selections, and represents the cumulative effect of many small, nearly imperceptible disadvantages and inequalities’ (Martin 28-5).

Most education systems in the Western world thrive on continually testing children and selecting only the best to continue. Until 18 years of age, most children in advanced countries are obligated to follow some form of education where they are constantly being tested. They also have to make decisions that will determine their paths through life. Bourdieu argues that having a higher class upbringing will work as an advantage in the education system as to those of a lower class upbringing simply because they have been more exposed to a higher class culture which is being taught in universities. The cultural capital as Bourdieu calls it, is largely inherited by the lifestyle of the parents and further crafted by education (Martin 28). As higher class people engage more often in higher class activities, they will most likely take their children along and teach them from early on what higher class culture and taste constitutes. The same counts for lower class people who primarily engage in lower class activities with which they will involve their children. Because universities are part of a high culture where high culture is being taught and even expected in the conduct of the students, coming from a lower class background will give a student a disadvantage as the system already supposes certain knowledge and skill that they will not have acquired yet. As Biber showed with his research academic discourse is the main register of communicating within the university, it is the means to express and defining the world it, therefore, becomes very much part of the cultural

(14)

capital as it rejects other registers because they are inappropriate to use in a university setting. Academic discourse can thus be viewed as a boundary that distinguishes the higher class from other classes.

Current research in Bourdieu’s exclusivity of high culture spells some concern on whether or not high-class people only engage in high culture activities (Peterson 87, Peterson and Kern 904, Lizardo and Skiles 2). Peterson and Kern are quite adamant in their research that high class does not exclusively engage in high-class activities and especially do not shy away from popular culture but rather appropriate them (902). They believe that the higher class people no longer engage exclusively in high-class culture but have developed a more omnivorous taste: a taste for both popular/ low-class tastes and high-class tastes (Peterson 903). Language then would also be more varied as it adopts different register through the exposure of popular culture. Changes in our society the past few decades have made it harder for higher class people to remain close and exclusive:

‘Rising levels of living, broader education, and presentation of the arts via the media have made elite aesthetic state more accessible to wider segments of the population, devaluing the arts as markers of exclusion. At the same time, geographic migration and social class mobility have mixed people holding different tastes […] Thus the diverse folkways of the rest of the world’s population are ever more difficult to exclude’ (Peterson and Kern 905).

And indeed in their research Peterson and Kern were successful in demonstrating that people which they classified as high-class had an omnivorous taste for music genres (902-3). Yet what they fail to research or even mention is the manner in which the higher class engages with the popular music genre. As tastes are part of Bourdieu’s cultural capital it can be viewed as more than merely an object with which can be engaged with or not. It is not simply an on/off switch. The high-class distinction is not of material or clear participation but it is the manner of participation, the actual interaction that makes the distinction between classes (Lizardo and Skiles 23). ‘[I]t is precisely the

ability and competence required to separate content from form that is at the basis of the set of

class-distributed cognitive schemes that have become valuable’ (ibid). So cultural capital is not just knowing and participating in high culture but also a certain skillset and mindset with which high class is able to distinguish themselves. And those skills are partially inherited through family life and also be acquired by education, especially higher education. The use of academic language in combination with popular culture makes their interaction vastly different than lower class people. Within the

(15)

university higher class people learn how to interact with cultural and how to appreciate the arts if they then apply this skill set to popular culture their experience is still different than that of lower class people. Universities, therefore, play a key role in distributing cultural capital but also in upholding the class differences.

Since universities have become accessible through scholarships and state support class variation within the classroom has become more common. However, those of the higher class still have an advantage through their inherited cultural capital.

‘Teachers, it is argued, communicate more easily with students who participate in elite status cultures, give them more attention and special assistance, and perceive them as more intelligent or gifted than students who lack cultural capital’ (DiMaggio 190).

Thus students who have inherited high-class cultural capital will have an advantage over those that do not have that luxury. This might not, however, be a conscious action teachers take as it is part of the education system to transfer the cultural capital to the students, those students who are already advance simply benefit from their competence by caring out their cultural capital.

In The State Nobility Bourdieu defines academic capital which has similar traits as that of cultural capital but it is set in an institutionalized form and is therefore specific for the universities as a field (Naidoo 458). The traits, tastes and styles of academic capital are all linked to the reproduction of the university structure and to ‘intellectual renown’ in the form of achievements and mastery of academic discourse (ibid). Even though universities have become more accessible their internal hierarchy has hardly shifted. Students have to study very successfully for almost a decade before they can acquire an entry position and then it takes several decades before they come into a senior position all the while they are constantly measured. There are only very few who are in such a privileged position that they do not require the constant measuring, often they are the one who are in power within the institution and thus responsible for the reproduction of academic system. Because of this top-down system change happens extremely slowly as generations pass by before the entire hierarchical system has been renewed.

The validation through measuring of both students and staff is not just one enforced from within but also carries weight outside of the academic field for example when it comes to raising funds for the university or finding support for cutting-edge research.

‘The educational system not only propagates a system of common academic classifications, but also reinforces social classifications due to a correspondence

(16)

between the hierarchy of academic disciplines and the unequal distribution of cultural capital’ (Martin 41).

Liaising with other social economic powers not only validates the internal power structures but also upholds them as the structure is on “display” to those outside it as communication channels are regulated often by social conventions. For example, when a journalist requires an expert view they will ask a senior scholar over a recently graduated PhD student even though the PhD student might have done excellent research on the exact subject the journalist is asking about. However, due to the social conventions set by the academic field PhD students are not deemed knowledgeable enough to answer questions from those outside of the academic field because those outside of the field have some understanding of its hierarchical structure. Academic language becomes, therefore, part of the identity creation not just a mechanism for knowledge transfer. As Heller and Morek point out, the socio-symbolic function of academic language is very much under-researched even though previous research showed strong evidence for the perception of academic language as ‘a “better” or “higher” language’ and academic language functioning as a ‘visiting card’ shows one’s identity and to which social circles they belong to (178-9).

The hierarchical organisation within the university is a product and a structure of the academic capital, it accumulates through academic accomplishments such as degrees and publications but also through the mastery of the academic disposition in manners, speech and writing (Naidoo 458). Thus the mastery of academic discourse at an early stage can help climb the hierarchical ladder in the academic field faster as the amount of academic capital is larger compared to the peers which in return gives more prestige. As academic language is part of the academic capital it, therefore, does not only give an advantage within the academic career path but also shows a competence that excludes others who do not have the same level of competence. One has to not only walk the walk but also talk the talk in order to belong the academic field. Even those outside the academic field have some understanding of its cultural and academic capital to know that it is hierarchically structured and that power is distributed from the top down.

Complex language use in universities is a marker of both academic capital as cultural capital as it is the means to accumulate tastes as well as a manner to demonstrate competence within a social group. Higher classes further develop their cultural capital at universities making these institutions part of the high-class culture. However, within these institutions advancement is strictly coordinated through the hierarchy which measures the academic capital which already starts at the beginning of the education system. Through constant testing of children and students on not only knowledge but also on social intelligence such as a knowing and acting the required behaviour, manner, writing and speech the academic field excludes all those who cannot acquire enough academic capital. But even

(17)

making it through years of proofing to be worthy of being part of the academic world young researchers have to continually validate their position within the academic field by for example publishing their research in notable journals, participating in conferences, lecturing.

Although language and specifically academic language is not the only way to accumulate academic capital it is a means to an end. Academic language allows individuals to identify themselves, helps to navigate the social structure within academia and is the language in which academic capital can be accumulated. All those in the higher classes accumulate their cultural capital through the education system and therefore through the academic language making its competence in this complex language a marker of high class. Those who fail to keep up with the discourse cannot fully participate in the same field and are thus excluded from participation.

2.3 Under the looking glass: Time Higher Education Ranking a

production of the elite for the elite

As we have read in the previous chapter the academic field within institutions are hierarchically organized from within with those with the most academic capital on top. This hierarchy is however no longer just confined from within. Universities have always competed with each other however this was often at a local or regional scale. It is not just important to simply be the best but one has to be the best of the best this was made easier by globalization and greater social mobility. The choice of which university one wants to be associated with has become larger which has made universities more the determined to distinguish themselves from one another. Although listings and comparisons of universities existed they were often at a regional level such as the Ivy League universities in the United States. Through globalization and digitization, it has become somewhat easier to compare universities across the globe and since the early 2000s yearly global ranking lists have been the talk within the field of academia.

The Times Higher Education ranking (THE) first appeared back in 2004 and has grown to one of the leading ranking lists for universities worldwide. THE was to provide a different look on university ranking as its counterpart the Academic Ranking of World Universities often referred to as the Shanghai Ranking as this list used to be produced by the Shanghai Jiao Tong University. The Shanghai Ranking originated back in 2003 as one of the first well-researched ranking of universities worldwide yet it was as much as a success as heavily criticized. Shanghai Ranking set out to find ‘world-class universities’ but are somewhat doomed in their quest due to ‘the difficulty to obtain “internationally comparable data”, they decided to rank order universities based on academic or research performance’ (Billaut, Bouyssou and Vincke 3). Billaut, Bouyssou and Vincke in their critique on the

(18)

Shanghai Ranking go as far as saying ‘the Shanghai rankings a poorly conceived quick and dirty exercise with no value whatsoever’ even stating that would it have been a Master Thesis it would have ‘surely failed’ (34). Their critique might be hard but is well founded and well researched. Their critique included the major focus on counting criteria of research output (citations, prizes and medals) and the lack of a mitigating factor on size which results in larger institutions being automatically ranked higher than smaller institutions simply because their research output and capacity are greater (Billaut, Bouyssou and Vincke 18). Other criteria include: lack of access to raw data, measuring criteria does not line up with the intended purpose, many smaller decisions are not well documented or not documented at all and that the data collections seems to be based mainly on what was already available resulting in a ranking list that cannot be reproduced (ibid 17).

THE provided an alternative to Shanghai and along with the QS ranking, they make up the three most popular world university ranking lists. THE has continually evolved their methodology over the years yet is still very much criticized. In its earlier days THE, similar as to the Shanghai Ranking relied heavily on research output for measurements with citations being the highest contributing factor as it makes up almost a third of the total score (Boulton 76). THE, however, also makes an attempt to incorporate teaching as a criterion that weighs in at a total of 30% of the overall score (Times Higher Education). More than half of the teaching score is made up by university reputation that is acquired by a survey which at first sight seems like a breath of fresh air compared to all the hard data on research output however, as Boulton critically points out ‘[s]uch approaches are most likely merely to reinforce existing, conventional stereotypes’ (76). He has a point, for how do you capture a reputation score in a quantitative manner when you are asking thousands of academia who they think is the best, does this not sound just like a popularity contest? And will not those universities with the biggest marketing budget be on the minds of most scholars? Will newer younger universities be able to compete fairly with the ancient giants such as Oxford and Harvard? Maybe the most important question to ask regarding the reputation score; what has a reputation to do with teaching? THE explains on their website that the survey examines ‘the perceived prestige of institutions in teaching’ (Times Higher Education). But how can you measure the perceived prestige of institution in teaching as academic usually only have an overview of their local specialised area of expertise (Boulton 76). However, it must be said that quantifying teaching excellence on such a large but also global scale each year cannot happen without some sort of comprise on the data collection.

So if university rankings are subjected to so much critique by the academic world for whom are they made and who benefits from them? For students, especially undergrads, ranking might not be an

(19)

ideal factor to compare different universities as Nello Angerilli explains in an interview in Forbes ‘The overall rank of a university, however, will not always provide that kind of detail nor will it reveal how many of the most accomplished researchers are teaching undergraduate classes’ (Cairns 2). As the rankings focus so much on research output undergrad students at the start of their academic adventure might value the quality of teaching, amount of contact hours, atmosphere and location of the university and other factors more than how well the university does research and how many citations they can get. Nonetheless, Angerilli does state a few lines further on that high ranked universities have ‘very competitive entry requirements’ which will result in the best and the brightest students sharing the same classroom, which in turn can ‘form an important part of a professional network after graduation’

(Cairns 2). So although ranking is not a big factor for students they do matter for the best, brightest and most of all the most ambitious students whose choice of the university can impact their professional careers.

Although it is hard to say whether university world ranking lists have been made for governments and policymakers, they certainly do make a big impact. Universities’ main production is that of knowledge and highly skilled workforce it is no surprise that governments have vested interest in having a high ranked university as they can boost the economy (Luxbacher). For example, governments in upcoming economies might want to invest in their own education by providing scholarships to higher ranked universities abroad (Luxbacher). Such a large movement of international students also has its drawback for example in the Netherlands international students has put a serious strain of the already tight housing market while Dutch students dread that more and more programmes will have limited places due to the high demand from abroad fearing that they will be left out (van Gelder).

Immigration and immigration policies have also been affected by university rankings. Having a degree from a high ranked well-known university can help open doors all over the world the opposite unfortunately is also true. Most countries each have their own education system with different requirements for higher education this often results in the same types of degrees completed in different countries being not recognized everywhere. For example, a Master’s degree received from the University of Aberystwyth in the UK is not recognized as a university grade Master’s degree in the Netherlands. This may seriously limit career options in the Netherlands. Many refugees in Europe have found themselves in a similar situation where highly skilled well educated Syrians are unable to continue their careers in Europe as their skills and education are not recognized by the industries resulting in many refugees working in unskilled or low skilled jobs when they finally find employment (Betts et. al 21-5).

(20)

Building profiles of universities on the basis of common factors can help students and researcher to compare universities over a greater spectrum but it can also move policymakers and even governments to take definitive actions. As there are so many stakeholders who use university rankings it is not surprising that with every release universities like to boast about their positions after all having a good position in a large accepted ranking list might not only result in extra funds for the universities but also the possibility to attract the best and the brightest students and researchers. The lists are similar to the hierarchical structure within universities the amount of academic capital of all the individuals in one university is measured against that of other universities making it an ultimate tool of exclusion. All universities strife to be on the list and those on the list strive to be on the top as those on the top strive to stay there. Similar to the organization of high class the ranking lists determine whether the academic capital of a university is large enough to be included.

2.4 Concluding Remarks: Academic: Language and Prestige

Young students who are just starting out at the university are being submerged in academic discourse, the language of the academic field. Recognizing the struggles of young students adjusting to the new language researchers have turned around and examined their own discourse in order to improve classes in academic writing and academic skills. They found that academic written language consisted of many complex noun phrases however, this was often in scholar to scholar communications. One of the most extensive scholar to student communication studies was that of Biber. Published in 2006 Biber described linguistics features of a large scale oral and written communications of several American based universities. Within his corpus, he not only examined textbooks but also assignment descriptions, syllabi, course catalogues and programme brochures including those published on the university’s website. One of the most interesting and unexpected results from his research was that the institutional writings such as catalogues and brochures were such complex text that they exceeded the academic textbooks on use of academic language. This was much unexpected as the intended audience of these texts are prospective students who have not yet mastered the academic language.

Although Biber does not provide any insights into why these texts are so complex one possible explanation is that of class. Bernstein found in his research that class inequality manifests itself in language as children from lower classes are less exposed to elaborate codes compared to children from higher classes. As elaborated codes require more words and sentences because the context has to be described, those more exposed to elaborate codes will find it easier to engage with more complex language. Bourdieu concluded similarly but expanded the thought with his concept of

(21)

cultural capital. Cultural capital defines traits, tastes and style of a field which is inherited and developed through education, the amount of knowledge and know-how of a field defines an individual’s place in the field. The higher classes have developed high-class tastes by engaging in high-class cultural activities such as visiting a museum or viewing a play and deeming popular culture inappropriate. However, more recent research shows that high-class people have developed a more omnivorous taste as they participate in both high class and popular class activities. Yet their participation is different as they have developed a different mind- and skill set compared to those of the lower classes. This can be attributed to their development of cultural capital while at university, where they have been taught how to appreciate culture. As this is done through the language they might express their tastes in popular culture in a similar manner as they would with high-class culture making the academic discourse an important marker of the higher class.

The academic field uses the academic discourse as well as an identity marker. The cultural capital of the academic field; academic capital consist of academic excellence in the form of certificates, degrees and awards and out of the academic disposition of which language is an important part. The education system is built the eliminate anyone who has not acquired enough academic capital, through rigorous testing of students from the moment they enter the system only those with enough academic capital are deemed worthy enough to continue their studies. Coming from a higher class will have an advanced over those from a lower class as their family has most likely been through the same education system and therefore have the know-how of the field which they have passed on to their children. For those that wish to pursue a career in academia will find that even after accumulating academic capital for almost a decade at a university this is still not enough. The internal social structure of the university is highly hierarchical where the most senior members, those with the largest amount of academic capital are on top. When graduated from a PhD young researchers like all other staff members are still continually being tested for example by their publication and citations rate. Although academic discourse is not the main means for measuring having a high level of competence in is vital in order to establish oneself and move up the ladder of academia.

With the coming of the global university ranking lists in the early 2000s universities themselves compete with each other to be the best of best. Even though these lists are under constant scrutiny because they favour large Anglo-American institutions over smaller and give far more weight on research output then teaching quality they do have the ability to move policymakers and even entire governments.

Attracting the best students and researchers in order to rise on the ranking lists can be a motive for universities to use academic language in texts for prospective students. It will help as a preselection

(22)

as students who are not competent enough in the academic language are deterred whereas those who are more competent are empowered as it is easier for them to navigate. Its use can also be a sign of prestige as the university display their own academic capital through texts. Nonetheless, academic language is both a marker as a boundary which distinguishes those affiliated with the field of academia by rejecting all other registers as they are deemed inappropriate. Its competence is vital for navigating and advancing in the higher class and specifically the field of academia.

(23)

2.5 Capturing Complexity

Since Biber’s study in 2006, no studies have looked specifically into the presence of complex language in texts aimed at prospective students. It has been more than 10 years since Biber published his work on institutional writings although I can still remember that I ordered brochures of different university programmes seven years ago these days all information can be found on the university web pages. These texts are now not only available for prospective students, parents and secondary school but for anyone with an internet connection. Although the medium has shifted somewhat it would be interesting to see whether or not universities use complex language in their programme and course descriptions. As academic language is linked to the higher classes and a sign of prestige within the field of academia it could be expected that ranked universities would perhaps be more inclined to use more complex and difficult language in order to display or even validate their leading position. However, such an in-depth descriptive linguistic study as Biber’s although encouraged is out of the question as I lack both knowledge and time to explore linguistics features of texts intended for prospective students. Nonetheless, there are other methods that can be used to give an indication. Academic language is largely made up of elaborate codes and carries many linguistic features that when put together makes its writings perceived as difficult and complex. Readability formulas would suit perfectly to give an indication of a complexity/difficulty large bodies of digital texts. Because they measure what age or level within the education system a person must be to adequately read and understand the measured text, they can be used to see whether or not these texts are accessible for the prospective students. By using readability formulas on course and programme descriptions of ranked universities an indication can be given of the complexity of these texts as well as some insights of the performance of these formulas on texts from university web pages. Although readability formulas have been well researched they have not yet been applied on web-based text targeted at intelligent adults which will bring some challenges but also interesting observations with them.

What makes a text easy to read? Capturing complexity of a text is no mean feat. Over the past century, many researchers have searched for methods and clues in texts that enable them to identify why certain texts are perceived as difficult or complex while others seem easy. To measure a text's complexity many variables play an important part in the outcome; environment, text subject, fonts, text placement, device, background reader, length of the text, the purpose of reading, tone, style and ultimately language. Qualitative research methods can give an accurate picture of the complexity of a text. However, being time-consuming and often small scale with sample passages, they are not ideal for long text or a large corpus. Quantitative methods are more suitable for large text and corpora, yet they are less accurate and can only give a prediction of the complexity of a text. Readability formulas

(24)

have been the favourite quantitative method for predicting the complexity of a single text or a whole corpus and have stood the test of time and technological innovation. They are widely used educational tool which helps teachers in finding texts that matched the reading levels of their pupils. Readability formulas have gone up and down in popularity over the years but a handful has stood the test of time. Their history tells a story of four markers in the English language that constantly reappear in the formulas as indicators of complexity: length of words, length of the sentence, amount of words with two or more syllables and the number of words that are not present in a predefined list of easily understandable words.

2.6 The History of Readability Formulas; From Counting Words by

Hand to Teaching a Robot the Secrets of a Wordsmith.

Readability formulas have been around for almost a hundred years but their foundations were laid back in the late 19th century when modern statistics were used to understand literature. One of the first published works on this topic was a little book called Analytics of literature, a manual for the

objective study of English prose and poetry by Lucius Adelno Sherman published in 1893. As a

professor of English Literature Sherman, like many of his peers, was on a quest to understand how literature could be used for the moral and spiritual enlightenment of citizens. But unlike the rest, he used statistics for a quantitative approach to literature rather than qualitative methods such as aesthetics which was popular at the time. Sherman had noticed that the average sentence length had significantly changed over the years: in pre-Elizabethan times sentence lengths were on average 50 words long whereas in Sherman's time the average sentence length had dropped to 23 words (right now it is about 20 words per sentence; DuBay 10). As the average sentence length was getting shorter, Sherman discovered that they also became simpler and more abstract; sentences became shorter and more concise and only contained the necessary information which makes them easier to read. He also noted that individual writers remained consistent in their average sentence lengths which allows the use of samples rather than the entire text.

Moving on a few decades we find Edward Thorndike, a psychologist at the University of Columbia, who wrote The Teacher's Word Book in 1921. Unlike Sherman, Thorndike focused on word length rather than sentence length. He had noticed that German and Russian teachers were counting word lengths in order to determine how difficult or how easy a text is. As we engage with the same word more often, it becomes easier to read and as our vocabulary grows it becomes easier to master longer words. Thorndike's book or rather list consisted of 10.000 English words ordered by frequency which he later expanded to 20.000 in 1932 and in 1944 with help of Irving Lorge to 30.000 words (DuBay 19).

(25)

Both Thorndike and Sherman were on the right path but it was not until Lively and Pressey put two and two together while they were tackling the problem of selecting science textbooks for junior high school students. The books were filled with technical jargon which made the science class more like a vocabulary class. They developed three different methods and tested them on 14 different textbooks and one newspaper. The first method looked at the vocabulary range, the second looked at the number of words that were not in Thorndike's list and the third used the median of the index numbers of the words that were present in Thorndike's list. The last method was the most successful leading the authors towards the following statement: ‘The fundamental value of Thorndike's contribution is obvious; the “Word Book” has opened up a whole new field for investigation’ (DuBay 40).

Continuing from Lively-Pressey formula, the early formulas tend to focus on average sentence length and average word length, building on the idea that long sentences and long words are perceived as more difficult to read (Begeny and Greene 189, Reed and Kershaw-Herrera 76-7).

As more research was done in linguistics the readability formulas became more sophisticated but this resulted into more elaborate mathematical equations making it harder for the users to understand how certain formulas are actually measuring the readability of the text leading some to think that using one or two formula is enough to evaluate readings (Begeny and Greene 201).

Since the 1980s the validity of readability formulas was put to the test. With the help of modern-day computing, it became easier to analyse large and diverse corpora against multiple formulas. Along with an increased interest into cognitive science, this resulted in a steep drop of the number of popular formulas and those that stuck around were measured against qualitative methods in order to get a view on their predictive aspect (DuBay 25-6).

Going into the 21st_{-century, readability formulas are widely available especially on the web which}

brings us a new problem. With the online readability tools, anyone can either type or simply copy some text into the text box, press a button and voilà - it tells you the readability of the submitted text. Although, it often refers to which readability formula it uses the problem lies not in the formula that is being used but in the way the text is being analysed. When we write code to parse a text we tell it what to do when it encounters certain characters such as periods, exclamation point and commas.

For example, the period: In general the period is used as a full stop, marking the end of the sentence, however, the period is also used to indicate an abbreviation as in the following sentence: ‘She was, as she said, suffering from la grippe; grippe being then a new word in St. Petersburg, used only by the elite’ (Tolstoy 1).

(26)

Saint Petersburg here is abbreviated however when a program parses this sentence it will interpret the abbreviation as a full stop resulting in an extra sentence. This might seem insignificant but when all readability formulas are based on average sentence length getting the number of sentences wrong will result in an incorrect score. This does not limit itself to abbreviations, each readability formula has its Achilles heel both in the formula itself as in getting the raw statics from the text wrong.

To continue the thought on how to define a sentence Gunning's Fog index also relies on average sentence length however in Gunning's original formula each clause was considered a sentence resulting in this sentence no longer being seen as 1 whole sentence but in fact 5. A simple parser cannot make a distinction between clauses especially when clauses are not separated by clear punctuation. Even the most sophisticated natural language processors can have problems with identifying clauses.

However luckily for Gunning in 1985, Judith Bogert was successful in demonstrating that when dealing with literature and large pieces of text taking sentences from the capital letter to the period worked just as well.

Another problem with the fog index when using an automated text analyser is the definition of complex words. Gunning defines complex words as words that contain 3 or more syllables not counting any suffixes. When counting by hand this is not a difficult distinction to make however when a program needs to interpret whether or not a word is three syllables and whether or not the word contains any suffixes that should not be counted it becomes a lot harder. Within the information retrieval field, programmers face a similar problem with suffixes and they prefer a hard lemmatization, however, this method does not help with the accuracy of the Fog index as nouns ending in –ed or -ing who manually would be counted, now might be considered two separate syllables and therefore not a complex word. Of course, nobody is obligated to use an automated text analyser to get the raw statistics from a text, one can always count everything manually. However, with the Fog index performs better with larger text as it makes a more accurate prediction compared to smaller texts.

In 1975 McLaughlin critiqued that most readability formulas including the Fog index were unnecessarily complicated as well as laborious as automation was still a costly option. He came with an easy to use time-saving formula building on de Gunning's polysyllable principle. The SMOG index allows you only to calculate the polysyllables of 30 sentences which most often come at about 600 words. This makes to SMOG index also limited when you have a text with less than 30 sentences and

(27)

it restricts you to the use of just 30 sentences. Although you will have to count the syllables manually calculating the readability is a lot more accessible as well as accurate.

However, when you are dealing with many different texts having to manually calculate the syllables of 30 sentences of each text is a time-consuming task. Living in a fast digital society you simply cannot afford to do handwork.

When trying to avoid the lemmatization and syllable counting the Dale-Chall formula may provide an answer. Rather than looking at the syllable count, the Dale-Chall formula looks at the number of difficult words in relation to the total amount of words. The difficult words are determined by checking them against a predefined list of ‘easy’ words. Those words that are not present on the list are considered difficult. Originally in 1948, the list consisted out of 763 words which on average 80% of fourth- grade students (nine to ten years old) are familiar with but in 1995 this list was updated and now contains 3000 words.

Combining the average words per sentence with an average amount of difficult words makes an easier way to automatically calculate the reading grade and less likely to fail because on the basis that the program cannot handle language the same way as a human does. However, the Dale-Chall formula is not ideal when searching for reading materials for younger children and neither for multilingual text or text with dialect or slang such as Faulkner's As I Lay Dying but also works of Dickens as the word list is focused on American English.

2.7 Modern Application of Readability Formulas

When dealing with readability formulas it is important to keep in mind that all formulas are a quantitative method for measuring the readability of a text and cannot provide a definite answer on the readability. They are only a mere indicator of a possible age or grade from which the reader would be adequately able to read the text. Nonetheless, they can provide a useful insight into to structure of texts and their ease of use allows large quantities of text to be analysed in a matter of seconds. Because of this, readability formulas have long been used to find appropriate reading material for young students to practice their literary skills but their application has expanded to all sorts of reading material but their application has since been expanded. The formulas have been one of the favourite methods to evaluate public health information as Meade and Smith point out:

‘if the written directions or warnings for a medication are disregarded because they cannot be read, both patient and practitioner are put in a precarious position. While this position is similar to that of the child who is mismatched with reading material, it holds the potential for far graver consequences for the healthcare consumer’(155).

This statement is also true for prospective students: if they cannot adequately read and understand the programme and course information how then are they supposed to know what the programme