Table of contents
2. Facebook – the user’s perspective
3. Communicating online: the basics of CMC
3.1. CMC on Facebook
3.3. Facebook as a hypertext
4. How does speech work on Facebook?
5. Quoting according to Bublitz
5.1. The forms of quoting on Facebook
5.2. The functions of quoting on Facebook
6. Quoting on Facebook
6.1. Common text actions on Facebook’s Timeline
6.2. User-authored text actions
6.3. Software authored-text actions
7. Quotations on Facebook – look who’s speaking!
7.1. Quotations in user-authored posts
7.2. Quotations in software-authored posts
9.1. Table of Screenshots
According to Boyd and Ellison (2007:2), “Social network sites [are] web-based services that allow individuals to (1) construct a public or semi-public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) view and traverse their list of connections and those made by others within the system.” People use such services to represent their personality online, thus facilitating communication with friends and also the maintenance of social ties. Clearly, the most successful (and influential) of those networks is Facebook. Established in 2004 as an exclusive community for Harvard students, it opened for the general public in 2006 and has since then gathered over 1 billion members all across the globe, currently holding second place of the most visited websites worldwide.1
Human communication has always been characterised by the recycling of language. When people talk, a great extent of the content of their discussion is devoted to what other people have already said. In doing so, speakers refer to utterances made by themselves or others inter alia by repetition and paraphrasing. They do so simply to report other people’s words but also to evaluate what has been said. Quoting has also found its way to communication media online, due to the technical properties of computer-supported environments facilitating the act of accurately reproducing a previously made statement.
Characteristically, Facebook employs mechanisms of automated text creation and distribution to support the users in making quotations. These mechanisms though, can be very complex and impenetrable. This is why I strive to provide insight into the process of quoting on Facebook in this paper. To do so, I will provide insight into the basics of the scientific field of computer-mediated communication as well as the structural principle of hypertexts in online environments. Following that, I am going to employ a newly shaped definition of the act of quoting by Bublitz in an examination of the most common text actions on the social network site Facebook. Thereby, I plan to come up with a framework of quotations on Facebook in both user-created and software-created texts.
2. Facebook – the user’s perspective
First, we have to examine the general features of the Facebook user interface. Any potential user has to create an account to log into the social network. The account structure rests on two main pillars: the home page, which acts as a starting point for all of the user’s endeavours in the Facebook environment and the personal profile, the place where any user may upload personal information, photos and videos to create the representation of their identity online.
The homepage2 provides access to all important functions of the network, the Facebook chat (a communication device similar to instant messaging services) as well as all the necessary controls a user needs to operate the website. The essential part of the home page however, are the newsfeed, located in the centre of the page, as well as the Ticker, located in the top right corner below some of the control elements. The newsfeed shows updates posted by the user’s network contacts, thus acts like the main information hub to investigate what is new in one’s network of friends. The Ticker provides real-time information about the network-actions of befriended members. I will describe these features in more detail later in this paper.
The second part of the user account is the personal profile3. It consists mainly of user submitted data and thus serves mainly his or her self-presentation. The profile data is divided into several layers. At the top, there is the Information layer, which provides access to all important parts of the profile and information about the user. Below that, on the left of the page are the Photo and Interests Layers, which serve as access points to the user’s uploaded images and videos as well as an overview over his or her established network connections with various items (i.e. pages of films, books, apps). Next to this layer, on the right of the profile, there is the Timeline layer, which shows the user’s posts in chronological order with the newest post on top.
In the following, I will now provide an overview of the basics of computer-mediated communication with a specific focus on those practices most applicable on Facebook.
3. Communicating online: the basics of CMC
The worldwide introduction of the internet into people’s homes, along with the hypertext system it is based on, kicked off an enormous change in human life. Evidently, this change especially affected the way people communicate, as can be seen by the numerous programmes and websites which are specifically designed to enhance human interaction online. This led to a change in existing linguistic patterns as well as the development of entirely new ones. As a consequence, linguistic pragmatics (among others scientific fields) began to take great interest in so-called computer-mediated communication or CMC.
Computer-mediated communication has been defined by Herring (2007: 1) as “predominantly text-based human-human interaction mediated by networked computers or mobile telephony”4. The use of an electronic medium for communication calls for new ways of scientific analysis and interpretation. Due to physical separation of the conversational partners, electronic interaction lacks the interactivity and immediacy of face-to-face conversation (Georgakopoulou 2011: 95). The rapid development of internet technologies also brought forward various means of spoken online communication such as computer-aided telephony through services like Skype or other Voice-over-IP-services, which may impede clear distinctions for linguistic research. The majority of online communication on the Internet today, however, still occurs in written form “by transmitting messages via networked computers” (Herring 2001: 612). This so-called computer-mediated discourse (ibid.) takes place on an abundance of various discussion fora, message boards as well as instant messaging services and social network sites. Since one of the most popular social network today – Facebook – is the corpus deliciti of this paper, I will focus on written forms of CMC or rather CMD from now on.
Compared to other written forms of communication, online discourse (mostly) is rather spoken-like and has therefore often been called ‘interactive written discourse’ or ‘written speech’ (Georgakopoulou 2011: 96). In general, the users of CMC employ various elements of either spoken or written discourse depending on the relationship between the participants, their specific communicative aims and also on the medium in use, i.e. what kind of computer service is used to talk to each other. Due to the mixture of characteristics of both written and spoken discourse found in CMC, computer-mediated communication takes place at the intersection between written and spoken dialogue (Georgakopoulou 2011: 96) and thus can be seen as a hybrid form of conversation.5
Therefore, different forms of CMC can be located on different points of the written-spoken continuum. It follows that written online discourse can either be located near the ‘spoken’-end or the ‘written’-end respectively. More spoken-like communication is called synchronous, where the interlocutors converse more or less instantly, as it is the case for chats or instant text messaging. In contrast, computer-mediated communication may be asynchronous, such as in e-mail or discussion fora, where participants “have more time for producing, editing and interpreting messages” which may result in “better opportunities to reflect and comment on their own and their fellow participants’ contributions” for the users. (Tanskanen 2007: 88). Although those distinctions may not always be as clear cut as presented above – there may be long breaks in conversations via instant messaging systems as well as e-mail conversations may be almost instantaneous – they still apply nonetheless.
3.1. CMC on Facebook
Being the highly frequented social network site that it is, Facebook combines a variety of communicative forms. The service provides means of communication reaching from (theoretically6 ) asynchronous modes like comments on uploaded content (i.e. status updates, photos) and posts to Timeline, to highly synchronous modes such as instant messaging with the Facebook-Chat, which was implemented in 2008 (URL 1). The enormous popularity of smartphones led to the introduction of the Facebook-App and more recently the Facebook ‘Messenger’-App, both of which provide even more ways to communicate, for instance by enabling the user to call contacts directly from the app-interface.
The primary way people interact on the Internet is through the reception of text on a computer screen and the input of text with the help of a keyboard. What most people lack, however, is an understanding of the actual mechanisms behind what they are seeing on their screens. The internet is a network of numerous interconnected computers that enables data transfer across any distance. This data transfer is being facilitated even further through the World Wide Web, “a system of computer servers which are connected through the Internet to support the exchange of files which are formatted mostly in HTML (hypertext markup language7 )” (Arendholz 2013: 7). In contrast to traditional text, which is generally seen as written language8, hypertext eludes such a clear definition. The Media and Communication Dictionary (2011: 93) describes a hypertext document as follows:
A document with embedded links called hyperlinks on images and/or highlighted words or phrases that when clicked on take the reader to other parts of the document. Hypertext allows information to be retrieved in a flexible, non-linear way. (Kleinman 2011: 93)
Although this delineation provides a fundamental understanding of hypertext, it misses some of its characteristics. Kleinman in fact describes what Storrer (2000: 236) calls Closed Hypertexts (namely “Geschlossene Hypertexte”), which feature a set number of linked modules in contrast to an opposing concept of open hypertexts (namely “Offene Hypertexte”), which stand out through “open ends” for users to add new modules. In the same paper, Storrer provides a more thorough definition for hypertext in general, originally drawn up by Berk/Devlin in 1991:
“Hypertext: The technology of non-sequential reading and writing. Hypertext is technique, data structure, and user interface. (…) A hypertext (or hyperdocument) is an assemblage of texts, images, and sounds – nodes – connected by electronic links as to form a system, whose existence is contingent upon the computer. The user/reader moves from node to node either by following established links or by creating new ones.” (Berk/Devlin, in: Storrer 2000: 227)
The definition above – despite being a lot older – names the essential criteria for hypertexts. They are not just a set item, but instead are created dynamically through interaction between the author and the reader. The definition also indicates that hypertexts generally lack a fixed linear structure. It follows that different readers may chose different paths when reading a hypertext. Moreover, Hypertexts consist not just of text, but also images and other semiotic signs and are inconsistent in a way that they can be altered or deleted. It is important to note that hypertexts can only be accessed using a computer (Storrer 2000: 230) or a similar device such as smartphones, tablets and the like. Following Eisenlauer (2013: 64) among others, hypertexts are based on four key characteristics, namely multilinearity, fragmentation, interactivity and multimodality.
Multilinearity means that readers are not given a set, linear order in which to peruse the text on their screen or click the presented links, but instead chose their own paths when reading a hypertext. In other words, Multilinearity can be described as how a hypertext influences its user’s choices in terms of how they may create their own reading paths through the various given hyperlinks. Depending on how strictly the given hypertext follows a hierarchical structure, it determines the way users interact with it, hence if they follow a ‘pre-established multilinearity’ (Eisenlauer 2013: 104) or move freely through the various nodes.
The structural requirement for multilinear reading of hypertexts is their fragmentation into text clusters. Those clusters can be intranodaly (i.e. texts arranged within one node), internodaly (i.e. texts arranged across different nodes of the same “base-hypertext”) or even extranodaly fragmented (e.g. through hyperlinks leading to content on different databases) (Eisenlauer 2013: 64)9.
The hypertext’s interactivity is divided into three components: first, there is cognitive interactivity, meaning that the users interact with the hypertext based on their individual background knowledge. Secondly, there is structural interactivity, which overlaps with the concept of multilinearity, meaning that create their personal reading matter within many possible reading paths. Thirdly, there is productive interactivity, which covers the user’s options to manipulate the existing hypertext through the clicking of links, the typing of website addresses or the use of a search engine (Eisenlauer 2013: ibid.).
Last but not least, multimodality means that hypertexts consist of various semiotic modes like texts, videos, audio-clips or pictures. From now on, I will refer to the various semiotic modes with the hypernym “text”, as a distinction between them is not always necessary in hypertext environments, due to the lack of differentiation made between them on the recipient user’s side. Again, it is crucial, that an electronic hypertext can only meet all those features if accessed on a computer or similar apparatus (Smartphone, Tablet PC, etc.).
3.3. Facebook as a hypertext
I will now go on to examine in what ways the website Facebook meets the criteria of hypertexts presented above, as hypertext status is essential for the way communication and most types of quotations on that particular social network.
Facebook’s rather static layout offers different levels of multilinearity to different users, depending on what they plan to do on the website. An “authoring” user, i.e. someone who wants to update their profile data is given fairly limited options of interaction with the hypertext and has to work with Facebook’s provided templates. On the other hand, users who browse the site for purely reading purposes, i.e. people who want to read news about their friends etc., are – despite having to follow the pre-set overall structure of Facebook – relatively unrestricted in how they may peruse the available texts. So, in terms of multilinearity of Facebook one can say that the service, due to its more or less hierarchical structure, generally limits user’s choices of potential interaction, although this may not always become apparent. (cf. Eisenlauer 2013: 105).
This leads to the question of Facebook’s fragmentation. As mentioned before, hypertexts can be fragmented intranodaly, internodaly or even extranodaly and all of those types of fragmentation are present on the site at hand. Pages on Facebook consists of various elements that can be seen as autonomous text clusters, e.g. the Timeline, Photos or About layers of a profile page or the News Feed, Messages, Events or Notifications layers of the Home Page. Consequently, the dominant webpage (Profile or Home page) is intranodaly fragmented. Additionally, the whole website of Facebook itself is internodaly fragmented, since it consists of numerous member profiles, group sites, pages, all of them being nodes themselves. Eisenlauer (2013: 105) points out, that text production on Facebook always “involves intranodal and internodal fragmentation simultaneously”, as every text action is automatically published on the user’s profile and at the same tame distributed through the News Feed. Extranodal fragmentation also applies, since Facebook offers the possibility to embed external content (such as YouTube-videos) into any user’s posts or profiles through the use of hyperlinks. The content, although stored on databases outside the Facebook-environment, is automatically aligned to fit the Facebook layout and may also be included in other member’s text actions, e.g. if they chose to write a comment or simply press the Share button. Additionally, various ‘social plugins’, such as a Like-Button on a non-Facebook website enable the users to “share their experiences off Facebook with [their] friends and others on Facebook” (URL 2), which again intensify Facebook’s extranodal fragmentation.10
The third characteristic of hypertexts, namely that of cognitive, structural and productive interactivity is also met by Facebook. Cognitive interactivity is given as users chose their readings based on their background knowledge and interests. Likewise, structural interactivity is given as those different texts are more than likely not presented in a specific order. Finally, productive interactivity is present due to the various (automatic) text creation processes initiated by the users. Eisenlauer notes that the act of clicking links, which was “for years connected with more receptive practices of perusing hypertexts, has transformed into a performative act” (2013: 109) due to the various automatic text creation processes explained later in this paper. In this way, Facebook’s software algorithms further unify the concepts of author and reader for this particular hypertext.
The hypertext feature of multimodality is very prominently visible on Facebook. User’s profile pages as well as posts are rife with different semiotic modes, such as pictures, music clips or videos. Additionally, Eisenlauer gives a distinction between “genuine multimodal acts [i.e.] the upload and distribution of one’s own data [and] remixed multimodal acts [i.e.] the integration of a YouTube video” (2013: 110).
4. How does speech work on Facebook?
Since users produce (sometimes speech-like) text on a social network site like Facebook (or computer-mediated communication in general), the speech acts of in everyday speech have to be slightly modified to better fit the computerised environment. The term text action, frequently used by Eisenlauer throughout his study to refer to “the general action of users generating and distributing text” (2013: 143) serves this purpose adequately. Regarding Facebook, it is important to examine how much control the users have over the text creation process and the text’s intended meaning as well as the text’s dispersion and contextualisation when using Facebook’s automatic text generation and distribution elements.
Communication on Facebook is a complex combination of user intentions and an assemblage of different semiotic forms which may or may not be organised by Facebook’s software elements. Following Searle, Eisenlauer explains that a speech act consists of the following parts:
1) The “utterance act” that refers to the uttering (animation) of the semiotic forms
2) The “propositional act” that refers to the selection of a text’s proposition
3) The “illocutionary act” that alludes to the individual communicative aims
4) The “perlocutionary act” that involves the consequential effects of the speech act (text action) upon feelings, thoughts or actions of the participants.
2 See Screenshot “Homepage”
3 See Screenshot “Profile”
4 See Herring (2010: 1) of Thurlow et al. (2007: 15) for a summary of the history of CMC research.
5 Herring (2010: 2) provides an extensive overview of scientific voices that deem CMC conversation-like because of various reasons.
6 I say theoretically, because the communication through comments may be instantaneously as well, given that both users are online at the same time and chose to converse through comments instead of messages or chat. This may be the case if a specific content leads to some kind of argument among users.
7 For a definition of HTML, see Kleinman (2011: 93f).
8 Eisenlauer (2013: 56 - 65) provides an extensive discussion of text and hypertext and their specific features and differences in textuality.
9 Storrer’s concept of hypertext-webs facilitates the understanding of extranodal fragmentation. According to her, hypertext-webs link different hypertexts and other entities (“Ganzheiten”) through hyperlinks. It follows that the World Wide Web is a hypertext-web, consisting of an increasing number of sub-webs (i.e. the various websites). (cf. Storrer 2000: 236)
10 Eisenlauer (2013: 108) points out that “member’s perception of extranodal fragmentation can be reduced, [because] contents from external websites are contextualized as illocutions serving self-presentation and/or friendship maintenance”.