PART I

THEORETICAL FRAMEWORK

CHAPTER 1. Conceptualizing and operationalizing gossip

“Definitions of gossip will always be complex and controversial.”

(Taylor, 1994: 34)

1 Introduction

“Most introductory books start with a definition, but I prefer to leave such things to the end, where they can do less damage. Definitions are useful for resolving disputes, but they can easily become intellectual straitjackets, tempting people into the mistaken belief that words have fixed or essential meanings that should be defended against the tide of cultural change and scientific progress.” (Evans, 2001: 147-148)

Evans (2001) is correct in saying that definitions fit better at the end of an analytical work. However, before I can analyze ‘gossip’, I still think some conceptualization of my research topic is required. Skipping an introductory conceptualization process of my study would be a methodological fault, or as Wert and Salovey (2004a) recently said:

“Although defining and describing gossip is an interesting enterprise in its own right, this difficulty in settling on a definition of gossip may contribute to stymied progress in its study. Indeed, it often seems that the scientific discussion of gossip has bogged down at this first step, that of defining the phenomenon. If empirical researchers are to proceed, they must grapple with this issue of exactly what behavior is being studied when one studies gossip. What ‘counts’ as gossip?” (Wert & Salovey, 2004a: 76)

In this chapter I do not want to put forward ‘the’ definition of gossip, but a framework for my analysis on gossip. A more funded and clarifying definition will be the result of my analysis, and be presented later in this dissertation. To argue why a definition is needed both at the beginning and the ending of an introductory book, or study, let me explain with reference to the work of Sapir and Whorf.

When thinking about gossip, we use language. When researching gossip and communicating this to others, we use language. It is important to note that in this process it is the language we use that will define how we think about and conduct research on gossip. I refer here to the Sapir-Whorf hypothesis that outlines the relationship between thought and language. Whorf (1956) claims that humans can only think about objects, processes, and so forth that have language linked to them (linguistic determinism). Whorf (1956) further claimed that culture is determined by language (linguistic relativity). How people look at the world is determined by the language that they use.

A classical example that is often mentioned when talking about the Sapir-Whorf hypothesis is the amount of words the Eskimo’s use to refer to ‘snow’. However, as Pinker (1994) explains: as much attention has been given to this ‘fact’, it is not true. It is a myth, Eskimos do not have four hundred words for snow, not even two hundred, one hundred or even nine. It is true that experts can come up with a dozen of words, but that is not different from English talking people who as well can come up with multiple words for snow, such as snow, blizzard, avalanche, flurry, hail, etc., therefore:

“Where did the myth come from? Not from anyone who has actually studied the Yupik and Inuit-Inupiaq families of polysynthetic languages spoken from Siberia to Greenland. The anthropologist Laura Martin has documented how the story grew like an urban legend, exaggerated with each retelling. In 1911 Boas casually mentioned that Eskimos used four unrelated word roots for snow. Whorf embellished the count to seven and implied that there were more. His article was widely reprinted, then cited in textbooks and popular books on language, which led to successively inflated estimates in other textbooks, articles, and newspaper columns of Amazing Facts.” (Pinker, 1994: 64)

Pinker (1994) further criticizes the linguistic relativity for treating other, non-literate cultures as weird compared to our own. Citing Pullum (1991), Pinker (1994) explains that if Eskimos would use various words to refer to snow, this is no different from breeders using various names for breeds, like botanists having names for shapes of leaves and interior decorators having multiple words to describe the subtle shade variations of 'blue'. It is not the words that make us look at things differently, but it is how detailed or focused you look at something that will determine the various names you will use to describe the phenomenon.

I mention this debate on the Sapir-Whorf hypothesis, because I do agree that how I will first define ‘gossip’ will determine my research on it. I mention the Eskimo-snow hoax, because as my research on gossip progresses in this dissertation, I will claim that various names should be used when looking at gossip as a researcher of this phenomenon. Now, to start off, let me first outline what ‘gossip’ means in a general sense, as a breeder would first explain what a ‘horse’ is in general, before mentioning all the various horse breeds he knows of.

To conceptualize and operationalize gossip, I will mainly draw from contributions of the Communication Studies, since as Saunders (1999) claimed that the contribution of Communication Studies to a better understanding of gossip is that they offer definitions of gossip to operationalize the study. Still, before framing gossip within the Communications Model, and outlining the various debates in this discourse, I want to start with an etymological overview. What we call ‘gossip’ today is different from what ‘gossip’ referred to some decades ago. Therefore, I begin with the historical roots of ‘gossip’, and the Dutch variant ‘roddel’. I not only discuss the etymology of the word ‘gossip’ in English, but also focus on the Dutch word ‘roddel’, since a major part of my research will be done in the Dutch speaking part of Belgium.

2 The words ‘gossip’ and ‘roddel’ in a historical perspective

2.1 The word ‘gossip’: from godfather to bad talking mother

“Gossip has a bad reputation around town – if not around the world. Among the West African Ashanti, nasty or scandalous gossip about a tribal leader is punished by cutting of the gossiper’s lips.” (Levin & Arluke, 1987: 3)

With this awful fact, Levin and Arluke (1987) start to sum up some examples that show that the English word ‘gossip’ clearly has a negative connotation today. In 1936 Henry Lanz published an article in which he compares ‘gossip’ with ‘the devil’:

Philologically the word “Devil”, Old English deofol , is derived from Greek diabolos (dia, “through,” and ballein, “‘to cast”), meaning “one who casts words,” “arguer,” “accuser,” “calumniator.” Thus the Devil is originally associated with the “sins of the tongue.” (Lanz, 1936: 492)

Lanz then describes how gossip, as the modern evil of our societies is all around, in our daily talk, in written texts, and even in the arts. However, he must admit that he does not know how to ban this evil from our daily lives. At the end of his argument against gossip he concludes:

“Strangely enough, the word gossip, be it paradoxically pointed put in conclusion, points in the direction of heaven rather than the Devil.” (Lanz, 1936: 499).

This is because, when looking at the etymological history, the word ‘gossip’ hasn’t always had a negative reputation. Up until 1811, according to the Oxford English Dictionary, the noun referred to a mode of conversation in a friendly sense, and not focusing on a person in a bad sense, as it is today (Spacks, 1985). ‘Gossip’ is derived from the old English word godsibb, meaning ‘godparent’ or ‘godfather’. But, as Rysman (1977) explains, the term ‘godfather’ in this context has a broader meaning than just referring to one family member having a close relation with a related child, as today. In times when people lived in more rural settings, all had large social networks of family, friends and acquaintances that all were potential godparents:

“In small villages, the god-parents were also neighbors and there must have been considerable talk among them, especially during the idle winter months. The god-parent was not merely a formality; he or she was involved in a pseudo-kin relation with the whole family in a society saturated with kin and pseudo-kin relations. Just as feudalism and chivalry prescribed a set of pseudo-kin relations for the nobility through the institution of vassalage, so the institution of the god-parent relation allowed peasants to develop imaginary familial relations with their peers. For some analytical purpose, one can treat the family and associated god-parents as a single, unitary group.” (Rysman, 1977: 176-177)

During the Elizabethan period (16^th – 17^th centuries) gossip referred to individual relationships, typically masculine, men’s drinking, gathering of male friends in bars, raising the glass. The female variant of gossip pointed to the gathering of family and friends during childbirth. When a woman gave birth, her female family and friends came together to give her support, and that is what gossip was about in the 19th century; meeting with family and friends, socializing with each other:

“Jan Steen’s painting dating from 1664 carried the title “Gossiping” and depicted such a domestic scene with the mother lying in bed attended by two women while others surround the kitchen table and heat water by the hearth, gesturing to one another as they engage in conversation.” (Romaine, 1999: 161)

By the end of the 19th century the connotation changed, gossip was redefined as ‘idle talk’ and “tattling’, an action that did no longer refer to the social act of gathering, but to a an ordinary, rather negative form of communication (Bergmann, 1993; Brison, 1992; Fox, 2002; Gelles, 1989; Levin & Arluke, 1987; Rosnow & Fine, 1976; Rosnow & Georgoudi, 1985; Rysman, 1977; Spacks, 1985). And remarkably: “The terms for ‘godmother’ in French (commère) and Italian (commare) also took a negative semantic turn, unlike the terms for ‘godfather’.” (Romaine, 1999: 161).

Levin and Arluke (1987), Rysman (1977) and Romaine (1999) suppose that during the 19th century men began to fear that their spouses would no longer remain humble. People then were convinced that women should stay at home to cook and take care of their husbands. When they did gathered together with other women what were they doing, and more importantly, what did they discuss? The word gossip not only changed from something good to something bad, but was also no longer a habit of both men and women, but ‘privileged’ to women only; women talk. The fear of men towards their wives became even more apparent when demonstrations were held in England in the 19th century. Pamphlets were distributed to warn people about ‘women talk’, and these conversations were ridiculed and even forbidden:

“Gossip poses a threat to the social order because it may give rise to slander and intrigue and expose illicit and secret liaisons. Professions such as midwifery allowed women passage between households, largely free of male control, to exchange mother wit-knowledge of intimate matters such as contraception and abortion. In 17th-century England, broadsheets denounced women’s gossip and its dangerous powers, among them curses and spells.” (Romaine, 1999: 163).

The fight against gossip grew in some places. For example, in Texas until the beginning of the 20th century women were no longer allowed to talk to other women outside their own family. Whoever infringed upon the law was punished (Rysman, 1977).

To conclude this overview of the English word ‘gossip’, I end by giving the current definition according to Merriam Webster’s Dictionary (www.m-w.com):

gos·sip

1 a dialect British: GODPARENT b: COMPANION, CRONY c: a person who habitually reveals personal or sensational facts about others
2 a: rumor or report of an intimate nature b: a chatty talk c: the subject matter of gossip

2.2 The history of the word ‘roddel’

Compared to the English word ‘gossip’, the Dutch or Flemish variant ‘roddel’ has a more recent history. In Belgium the word first appeared at the end of the nineteenth century, somewhere between 1865 and 1870. When I searched for the meaning of the word ‘roddelen’ in the “Woordenboek der Nederlandsche Taal” of 1924 (Van Der Meulen, 1924) I found a description that does not fit at all in the nowadays use of the word roddel. The word “roddel” back then was used to refer to the rumbling noise of streetcars or the rumbling noise of hungry stomachs:

“RODDELEN, onz. Zw. Ww. Klanknabootsend ww., in Z-Nederl. gebruikelijk.

Al schuddend, schokkend, hobbelend een dof geluid geven. Verg. Robbelen (I). // Een roddelend rijtuig, SCHUERM. Aan heur roddelen ken ik mijne kar, RUTTEN. De käär roddelt oëp de stiëwäg, TUERL.

Rommelen. Verg. Robbelen (II), 2). // Mijn buik roddelt, SCHEURM. (zie ook TUERL.; RUTTEN).” (Van Der Meulen, 1924: 657).

According to van der Sijs (2001) we inherited the word ‘roddel’ from the Yiddish vocabulary, a language that appeared in German ghettos in late Middle Ages, derived from German, Hebrew and Aramese. The language infiltrated the Dutch language when Jews fled from Germany to the Netherlands in the beginning of the seventeenth century. Since most of these Jews were poor, they came in touch mostly with lower class Dutch people, which caused that their Yiddish language mixed up with the folk language of Dutchmen. Yiddish became more popular over time, but during the nineteenth century Yiddish was banned from the official Dutch language. Nevertheless some words did not disappear, and even appeared in official dictionaries, like for instance the word ‘roddel’, meaning ‘kwaadsprekerij’ or ‘bad talk’ (van der Sijs, 2001: 269-273; de Vries & de Tollenaere, 2002: 308).

Of course the act of gossip in Belgium and the Netherlands has a tradition that goes way back much further than the word ‘roddel’. But how did people refer to the act before they called it ‘roddel’? The etymological dictionary of the Dutch language (Het Groot Etymologisch Woordenboek van de Nederlandse Taal, 2003) refers to four main words that can be historically linked to the word ‘roddel’: ‘babbelen’, ‘baker’, ‘anekdote’ and ‘broddelen’. The last word ‘broddelen’ doesn’t explain much, it refers to ‘screw up’ ‘making a mess of something’, and is only related to the word ‘roddelen’ because of structural reasons (b-roddelen) and not because of semantic reasons. The other three words though, all have a semantic connection to ‘roddelen’.

The first word ‘babbelen’ has the clearest semantic connection with ‘roddelen’. Current ‘babbelen’ means ‘non-stop talking’, ‘chatting’, a meaning derived from two older connotations of the word. First, ‘babbelen’ used to refer to ‘moving the jaws’, ‘biting’ during the 16th century, and maintained this meaning in some dialects. The movement of the jaws refers to the talking behavior that is related to ‘babbelen’ today. Presently the jaw movement can also be found in the word ‘babbelut’ that has a quite similar structure as ‘babbelen’, and stands for a specific kind of candy that requires a lot of jaw movement. The second ancient meaning of ‘babbelen’ goes back to the Latin word babulus, which means ‘chatting person’. This Latin word was incorporated in many other languages and in Dutch it referred to ‘stamelen’ or ‘stotteren’ which can be translated as ‘to stammer’. ‘Babbelen’ is actually onomatopoeia for unclear speech, like a barbarian. It is also related to the word ‘babbe’ that refers to the proto-language of young children.

Now what links the above definitions with the word ‘roddelen’? A variant of the word ‘babbelen’ is ‘blab’ which literally means ‘babbelen, roddelen, zijn mond voorbij praten’, which can be translated as ‘babbling, gossiping’ (Het Groot Etymologisch Woordenboek van de Nederlandse Taal, 2003). In this etymological context I notice no negative connotation of the word ‘roddelen’, but a rather innocent meaning of ‘talking constantly, babbling’.

However, a negative connotation is noticed when looking at the etymological connection between the words ‘roddel’ and ‘baker’. This last word's meaning is derived from the 17th century word ‘bakermoe(de)r’, meaning ‘dry-nurse’, which in turn is derived from the 15th century word ‘bakeren’, referring to ‘koesteren, warmhouden’ which can be translated as ‘to coddle, petting’. The second part of the word ‘bakermoeder’, ‘moeder’ simply means ‘mother’, and has dropped out from the word. The connection between the words ‘baker’ and ‘roddel’ can again be found in a derivation form ‘baker’, ‘bakerpraatjes’, meaning in 1898 ‘kletspraat, roddel’, translated as ‘gossip’, but with a negative connotation. It refers to the kind of ‘foolish negative talk’ that was common among dry-nurses (Het Groot Etymologisch Woordenboek van de Nederlandse Taal, 2003). Here we clearly find a negative connotation of the word ‘roddel’, but we have to comment that this etymological record is much more recent (end of 19th century) than the more positive connotation of ‘chatting’, which we found in the word ‘babbelen’. This context of negative talk during childbirth is parallel to the etymological history of the English word ‘gossip’, as I mentioned earlier.

The last word that appeared in our etymological search of the word ‘roddel’, ‘anekdote’ first appeared in the 18th century and is derived from the Latin word anecdotum, which means ‘unpublished’. Current the word ‘anecdote’ refers to a ‘short amusing story’. The link with the word ‘roddel’ is an anecdote itself; the Byzantine historicist Prokopius (500-600) wrote seven books that glorified the life of emperor JustinianusI, but the eighth and last book about him was a book full of scandalous stories and gossip about the emperor’s life. Therefore it could not be published before JustinianusI died, and it was called Ane’kdota (Het Groot Etymologisch Woordenboek van de Nederlandse Taal, 2003). The connection between the words ‘roddel’ and ‘anekdote’ lies simply in the fact that the book full of gossip-like stories about JustinianusI was called Anekdota. But looking at the current meanings of both words I see the parallel that both are short amusing stories.

Again, to end this historical overview of the Dutch word ‘roddel’, I present the current definition of this word as a noun and as a verb, according to Van Dale’s dictionary (www.vandale.be):

rod·del (gossip as a noun)

1 kwaadsprekerij => achterklap, gekonkel, gekonkelfoes, gossip, kletspraat, roddelpraat (evil talk)

2 onwaar bericht, geval van roddel => kletspraatje, roddelpraatje (untrue message)

rod·de·len (gossip as a verb)

1 met genoegen praten over anderen, m.n. in ongunstige zin => iem. over de tong halen, kletsen, konkelen (talking about others with joy, but in negative sense)

2.3 From innocent, bonding behavior to negative, scandalous content

From this historical overview I conclude that both the words ‘gossip’ and ‘roddel’ have shifted from a focus on innocent social behavior to negative scandalous content. The earlier definitions of gossip stressed a behavioral strategy, the modern definitions focus rather on gossip as a noun than a verb. Still gossip is nowadays used to refer to both an act as well as the content of some of our conversations. In the following section I will outline which current discussions exist on how to define both gossip as a noun and gossip as a verb.

3 Conceptualizing gossip today

3.1 Gossip as a message and an action in current conceptual debates

“We all ‘know’ what gossip is, but defining, identifying, and measuring it is a complex enterprise for practical investigation. At the very least, the everyday understanding of the term gossip is included in, but insufficient to encompass the construct as used by researchers. Defining the content, circumstances and functions of gossip will help to put the research methods used to study it into perspective.” (Foster, 2004: 80).

“Whether or not experts have been able to concur on a definition of gossip, it is not very difficult to intuitively decide whether something is or isn’t gossip.” (Rosnow & Fine, 1976: 84). Everyone knows what gossip is, until one asks to give a definition of gossip. Thomas (1994) once compared gossip with embroidery; both can be eye-catching and hard to ignore, and both can be covering a large space, but be almost unnoticeable. It is very difficult to define what gossip is and what it is not, conversations are most often a mixture of both.

Gossip refers to stories, and gossip means action: “[…] gossip is about something and is something in itself; any study of gossip should consider it as both text and social activity.” (Brenneis, 1989: 226). When defining gossip, it is indeed necessary to differentiate between gossip as an act (‘gossiping’, or ‘to gossip’) and gossip as a noun, ‘a gossip’. Post (1994) also stresses the need to distinguish gossip as a noun and gossip as a verb. Since, as he explains: as an action, gossip unites people and is less threatening than gossip as a noun, which can be malicious and even group-threatening.

In what follows I will frame the current aspects of discussion about ‘how to define gossip’ in the discourse of Communication Studies. Before presenting the different aspects of discussion about both gossip as a noun and gossip as a verb, I will indicate how both are related to each other. I frame both aspects of gossip in an information model that links both and highlights which features I should look at.

In this chapter I will sometimes briefly touch on Media Gossip as well. However for a more detailed discussion on the definition of media gossip I refer to the last chapter (chapter 7) on Media Gossip.

3.2 Framing gossip in communication models

“The elements that determine whether information is gossip are the subject, the producer, the recipient, and the relationship between them.” (Smith et al, 1999: 122). Gossip concerns the transmission (gossip as an act) of information (gossip as a noun). Gossip is an act of communication, and as Lasswell (1948: 37, see also figure I.1) said, when studying communication we need to look at 'Who said what to whom, through which channel and with what effect?' I frame my overview of all points of discussion about gossip, in Laswell’s (1948) classical model of information transmission. Although Lasswell focused on mass communication and propaganda, his model is applicable to gossip, both in an interpersonal and mass media context, which both concern the transmit of gossip information.

Figure I.1. The Lasswell formula:

Of course criticisms have been given to Lasswell’s model, and other models of information transmission. Relevant to my discussion, are following critiques; first of all the context wherein the transmission takes place is not considered in these models. Secondly, this model looks at the effects of the transmission, but overlooks the intentions of the communicator. Braddock (1958) adjusted Laswell’s formula, to a more extended version where those critics are taken into account (Braddock, 1958, as cited in McQuail & Windahl, 1981). Braddock’s adjustment can be represented as follows:

Figure I.2. Braddock’s extension of the Laswell Formula (copied from McQuail & Windahl, 1981: 11)

An extra comment, I add, is that even in this extended model of Braddock, the relation between the sender and receiver is not considered. How the sender and receiver are related to each other, will most certainly influence the information transmission between both actors. I therefore opt to add the extra question ‘And how are sender and receiver related to each other?’ to Braddock’s adjusted version of the Lasswell formula.

Altogether, I will frame the conceptualisation of gossip according following characteristics; first of all I differentiate for gossip as a noun, which is the message in the transmission model. As I will outline, some of the most severe debates about gossip focus exactly on this content. Next, looking at gossip as an action I will look at disagreements about:

The channel used to dismiss the gossip,

The context wherein this happens,

The actors of gossip (communicator and sender); who gossips?

The relation between those actors,

The motivations of the sender, and,

The effect of the gossip transaction.

3.3 Gossip as a message: debates on what gossip content is about

Looking at gossip as the message of the information model is looking at gossip as a noun. I do not discuss here the action, but purely the content of what is being transmitted. As much as can be debated on aspects of the act of gossip, the biggest discussions rise when it comes to the content.

3.3.1 Gossip: universal themes about human subjects

Merry (1984) and Gilmore (1978) say we can gossip about humans or events. But, the first major distinction that differentiates gossip from other forms of communication, such as rumors, is that at least one human subject must be present. Therefore, others (e.g. Foster, 2004; Gelles, 1989; Rosnow & Fine, 1976; Rosnow & Georgoudi, 1985) classify talk about events as ‘rumors’. Gossip concerns personal matters (Nevo & Nevo, 1993) “Have you heard two dogs were fighting in the park yesterday?” is not gossip, whereas “Have you heard John’s dog was fighting with another dog in the park yesterday and he did not interfere?” is gossip. From the moment a human subject is involved in the content, the message is ready to be potentially classified as gossip.

About these human subjects, gossip topics cover a broad range of social matters. Almirol (1981: 295) collected data on gossip among the central Californian Filipinos and concluded that “The most common topics are immorality, love affairs, family quarrels, bankruptcy and other economic matters, inhospitality and rudeness.” Love affairs were also the most common gossip topic in the 18th century French documents that Darnton (1997) analysed. Saunders (1999), concentrating her research on gossip among older women in a support group, concluded that the most discussed topics were (1) fashion, (2) dating, and (3) problems at the Center they were part of. In general we can say that gossip ranges from love issues, (Gelles, 1989; Levin & Kimmel, 1977; Spacks, 1985), to health (Percival, 2000; Smith et al, 1999) and many other person-related aspects.

McGill (1968) says that universal themes, that appeal to a broad audience score best in information transmission. Divale and Seda (1999) did a cross-cultural analysis of gossip topics, among 136 societies, that are listed in both the Standard Cross-Cultural Sample and the Human Relations Area Files. They came up with a list of 24 gossip topics that occur cross-culturally. Using a factor analysis they could cluster these topics to three primary factors, and five secondary factors. An overview of their 24 topics and how these can be clustered can be found in table I.1.

Their three primary factors, which are ‘general gossip’, ‘political gossip’, and ‘machismo gossip’, make sense concerning classification of specific topics. The five secondary ones might look a little confusing, such as gossip about ‘cattle’ in combination with ‘marriage’. The authors explain that cattle are often used as a bride price and this explains the combination. Also combining ‘religion’ with ‘farming’ seems to make little sense, but Divale and Seda (1999) argue that a lot of farming involves religious rituals.

Their research is a nice overview of what can be considered as cross-cultural occurring gossip topics. However, the presentation of their analysis misses more detailed descriptions of the 24 categories. For instance, gossip about ‘cattle’, can only be called gossip if it is talk about a certain person’s property of cattle. As I just explained, gossip requires human subjects, we do not gossip about non-human things, such as animals. Therefore the study of Divale and Seda (1999) should be looked at with a certain scepticism. Their relevance for the discussion here is that their analysis confirms the existence of a cross-cultural broad range of topics that are present in gossip conversations.

Since gossip is about social matters it is often denigrated as being ‘small talk’, not delving into deep issues. (Ayim, 1994; Morreall, 1994). In later chapters, I will contradict this claim and argue that gossip even plays a crucial part in everyone’s life. To finish here, I suggest that, when defining gossip, a clear restriction about the presence of at least one human subject (or gossipee) should be made, but no restriction should be made about what is being said about this gossipee. Gossip is about all traits and behaviors of a certain person, or multiple persons.

Table I.1. Divale and Seda’s 24 topics of gossip across 136 societies classified into 8 factors

3.3.2 The subject of gossip: who do we gossip about

Looking in more detail to the gossiped subjects, or gossipees, I noticed some critical points of discussions between researchers that can be relevant to take notice of in my conceptualisation process. First of all, most gossip is about living human beings, but we cannot exclude the dead from gossip conversations. People talk, and gossip about dead people as well (Gilmore, 1978). We do not gossip if we talk about Little Red Riding Hood, therefore fictional characters can be excluded. However, people who once existed can be subject of gossip.

Next, a more elaborated discussion focuses on how known the gossipees should be to the gossipers. Noon & Delbridge (1993) think gossip is restricted to talk about members of one’s social setting “… thus speculation about the private lives of royalty, for example, would fall outside of our definition and could be considered ‘idle chatter’.” (Noon & Delbridge, 1993: 25). Hannerz (1967) is a little less strict, leaving open some option to gossip about celebrities as well. He says that gossipers need some minimal awareness about the gossiped subject. This is, the gossipee should be identifiable to the gossipers. And:

“Thus it is possible to gossip about movie stars and royalty, although one does not interact with them in face-to-face relationships. On the other hand, talking about an unknown man whom one passed in the street once is not gossip according to this view.” (Hannerz, 1967: 39)

So, according to Hannerz we need not to restrict gossip as strictly, as Noon and Delbridge suggest, to only gossipees that are part of the social network of the gossipers, but a little restriction has to be made. Gossip about non-celebrity people who are not part of the gossipers’ social network is excluded from Hannerz’ definition, since these people are in no sense identifiable to the gossipers.

Still, looking at Ben-Ze’ev’s (1994) classification of gossipees: “The objects of gossip fall into three major groups: (a) people in our immediate surroundings, (b) famous people, and (c) people whose intimate and personal lives are unique.” (Ben-Ze’ev, 1994: 17). In this statement I find an argument to incorporate non-celebrity and non identifiable gossipees in the general definition of gossip. Subjects can become gossipees if they have a unique feature about them. I agree with Ben-Ze’ev and also say that we can gossip about people we do not know. To clarify my statement, consider the following conversation:

Anne: “Have you heard the story about that Kung Fu chick?!”

Maggie: “No, tell me!”

Anne: “Well, she was walking her dog in a park, in the afternoon, when a guy jumped on her back, trying to cut her throat! Luckily she was able to react very fast; it seems that she was a skilled Kung Fu fighter or something like that! Anyway, she seriously attacked the guy, even managed to tie him up with her dog’s leach, and brought him to the police station.”

Maggie: “Oh, what serendipity!”

Anne: “Yes, sure, since that guy seemed to be the one who killed another girl on campus last month!”

Two girls discussing what happened to another girl and a guy they both don’t know. If we restrict gossip to information about known gossipees, than this conversation is not gossip. Still, what we call ‘gossip magazines’ are covered with such stories. I think these conversations are very much the same as any other gossip conversation about known gossipees. Overall I therefore propose not to make any restrictions concerning the gossipees in my general conceptual definition. We gossip about everyone, known people and unknowns. As Almirol (1981) says: “Gossip is viewed as a social equalizer; everybody gossips and everybody can be subjected to gossip.” (298). There is a ‘but’ to this, as he argues, though: “But there are persons who are more likely to be gossiped about than not. Persons who are regarded as constant violators of social values are dealt with promptly and vigorously through gossip.” (Almirol, 1981: 298).

As I will argue later in this and other chapters, I think a classification of gossip is necessary to resolve all discussions and make future research on gossip more clear and comparable. Almirol (1981) mentions that violators of social norms will be more likely to become gossipees. My chapters on the social uses and evolutionary explanations for gossip (chapters 3 and 4) will delve deeper into this matter. Here I want to outline a general definition which I will use for my contribution to a better understanding of gossip.

3.3.3 Can a gossiper be a gossipee as well

Rosnow & Fine (1976) define gossip as: “news about the affairs of another, one’s own memoirs or confessions, or any hearsay of a personal nature, be it positive or negative, spoken or in print.” (Rosnow & Fine, 1976: 87). And Almirol says that: “[…] gossip is always done behind the subject’s back. An important rule in gossip is never to confront the person face-to-face.” (Almirol, 1981: 297). I use these citations to open a next point of discussion; do we restrict gossip to talk about absent others, or do we also include talk about the behavior of present persons?

Most researchers (e.g. Arno, 1980; Ayim, 1994; Bergmann, 1993; de Vries, 1990; Eder & Enke, 1991; Foster, 2004; Gelles, 1989; Hannerz, 1967; Kuttler, et al, 2002; Merry, 1984; Morreall, 1994; Nevo & Nevo, 1993; Spacks, 1982, 1985; Wert & Salovey, 2004b; Young, 2001) claim that gossip is the telling of tales about absent others. Foster (2004) even argues this to be the only restriction necessary to classify a conversational message as gossip. In his recent overview of the most important studies of gossip, he says: “So the absence of the third party seems to be a minimal and, for many, sufficient requirement.” (Foster, 2004: 81).

Other researchers (Rosnow and Fine 1976, 1978; Dunbar, 1998a; Fox, 2001; Gilmore, 1978; Medini & Rosenberg, 1976) clearly disagree with Foster and others, and even take the totally opposite positions in this discussion, by saying that people can even gossip about themselves.

I opt for a middle-road position. I agree with the comments of Foster (2004) and Kuttler et al (2002) who argue to call gossip about yourself self-disclosure. Gossiping about yourself is a contradiction in terms, I think. Gossip is talk about other persons. Talking about yourself might be very similar to gossip and can even be used to reach similar goals, but I do agree with most researchers to restrict the definition of gossip to talk about others, and not the self. However, I am not as strict as most researchers, who declare that gossipees can never be present in conversations. It happens that, when talking in a group, someone suddenly starts talking about one of the participants to all others. “Hey pardon me, in case you might not have noticed, I am standing next to you!” Have you never uttered this expression, when you where made fun of in front of a whole group?

Foster (2004) and Kuttler et al (2002) suggest to call information about present others ‘disclosure’, but I do not see why this kind of talk about the traits and behavior of a third person cannot be labelled as gossip. The only further restriction I make is that the gossip cannot be addressed directly and solely to the recipient if this person is the gossipee as well. This is, we can gossip about present others, but those others should be third persons. “Did you know you are such a clumsy person?!” uttered in a dyadic interaction is not gossip. It is directly transmitted to the receiver, who here is a gossipee as well. I agree to call this disclosure. However, if Ann tells Suzy ‘Did you know Margareth is such a clumsy person!?” while Margaret is present at the moment of transmission, I do consider this to be gossip.

I therefore restrict the gossipee(s) to ‘third’ persons, with no restriction of absence; the gossipee as third person can either be absent or present at the time of the information transmission. But with a restriction of being an active member of the conversation: gossipees cannot gossip about themselves, they cannot be senders of gossip. Nor can they be receivers of gossip; the gossip message cannot be addressed to them personally. They can be physically present when two or more gossipers gossip about him or her, but cannot be an active participant of the conversation.

3.3.4 Is gossip good, bad or both

“My favorite letter of those I received when newspapers proclaimed my belief that “gossip is good for you” came from a woman who assured me I would sell more books if I alleged instead that gossip is bad for you.” (Spacks, 1985: 258)

In her article ‘In praise of gossip’, Patricia Meyer Spacks (1982) has mixed feelings concerning the goodness or badness of gossip. Gossip can be healing for the teller, but can seriously damage the reputation of those gossiped about, and: “[t]he trouble is that no clear line divides what I’ve called “good” gossip from what I’ve called “bad”: they inhabit a continuum. One can slide from one to the other, and then slide back again.” (Spacks, 1982: 36).

The debate about whether gossip is ‘good’, ‘bad’, or both is maybe the biggest issue researchers and laypeople argue about. de Vries (1990) defines gossip clearly as talk with a negative connotation. Baumeister, Zhang and Vohs (2004) say that psychologists in general show little respect for gossip, by attaching a negative connotation to it: “Psychology has not generally had much respect for gossip. The traditional and prevailing view has regarded it as an indirect form of aggression, akin to teasing.” (Baumeister, Zhang and Vohs, 2004: 112).

When asking the elderly people he observed to describe ‘gossip’, Percival (2000) noted that most of his respondents would say it is ‘bad talk’, even though his analysis showed that among these elderly both negative and positive gossip flourished. The everyday meaning of gossip has a pejorative sense of an undesirable action, engaged by those who don’t have better things to do, as I explained in the etymological overview. Still, as Arno (1980) comments, the anthropological definition of gossip is broader than this, defining gossip as “the informal circulation of information about and evaluation of people’s behavior” (Arno, 1980: 344), both positive and negative talk are included.

And indeed, in spite of its negative reputation in the common sense use, researchers of gossip (e.g. Ben-Ze’ev, 1994; Fine & Rosnow, 1978; Goodman & Ben-Ze’ev, 1994; Rosnow & Fine, 1976; Spacks, 1985) agree that gossip cannot always be described as malicious. Gossip is not necessarily negative talk. It is the exchange of social information about who is doing what to whom (Dunbar, 1998a; Fox, 2001; Rosnow & Fine, 1976). Studies, such as the above mentioned study of Percival (2000) among elderly people, or studies among children (Kuttler et al, 2002) all conclude that both admiring and pejorative statements are made in gossip. The study that came up with the most convincing evidence to argue this debate so far, is the study of Levin and Arluke (1985), who eavesdropped on the conversations of male and female college students. These researchers concluded that gossip could both have clearly positive (27%) and negative (25%) references towards other people, but that most gossip did not even have a clear connotation.

To rule out misunderstandings, should we use different words to refer to ‘good’ and ‘bad’ gossip? Some have suggested this and do so. For instance Gelles (1989) who talks about ‘scandal’, when referring to bad gossip. Or Guendouzi (2001), who differentiates for ‘harmless’ and ‘malicious’ gossip. The first she calls ‘peer group news-giving’ and the latter ‘bitching’. Collins (1994) keeps the word ‘gossip’ for bad talk and invents the new word ‘quossip’ to refer to the discourse that is similar to gossip, but lacks the malicious aspects. Also Spacks (1985) labels them differently: naming them ‘serious gossip’, and ‘malicious gossip’. The trouble with redefining for two different kinds of gossip is that this puts gossip in a clear black-and-white frame, while in reality different shades of grey lie in-between both extremes (Spacks, 1985).

Researchers have tried to explain why gossip has such a negative connotation. All of the arguments used are related to other points of discussion, as I will discuss later, such as Ben-Ze’ev’s (1994) claim that the ‘bad’ attached to gossip is due to the consequences, or effects gossip can have. de Sousa (1994) suggests that the bad connotation of gossip comes from the power relation between the actors. Gossip is a very powerful tool, and just like other kinds of power, it is feared and condemned. According to Sulls (1977) the recent negative association with gossip is because of the inherent unreliability of gossip. And last, Almirol (1981) states that gossip is bad talk; with bad intentions of the gossiper. This shows that all points of discussion I outline here are interrelated. In chapter 5 I will further argue that negative loaded gossip might be recalled more often, because of the so-called negativity bias (see chapter 5, section 4.1.5).

To end the good/bad discussion I conclude that gossip can both be good or bad, dependent on how you define ‘gossip’. The good sense of gossip is incorporated in the broader definition of gossip, the bad connotation characterizes the narrower, more common definition (Taylor, 1994), and for a scientific approach to study gossip it is best to take a neutral position, letting both options remain open (Kurland & Pelled, 2000). Therefore I make no restrictions on the connotation of gossip in my general conceptual definition.

Yet, ending this part of discussion, I also want to mention a practical solution to decide whether gossip is good, bad or both. In order to gossip ‘good’, or as Westacott (2000) calls it ‘ethical’, he outlines us some easy guidelines. He summed up a few questions one has to answer for him- or herself, that indicate whether a gossip story is ethical to spread around or not. A schematic overview of these decisions can be found in figure I.3.

Figure I.3. A guide to ethical gossip (copied from Kreiter, 2003: 56)