The Latin particles quidem and equidem: a discourse pragmatic description

(1)

The Latin particles Quidem and Equidem:

A discourse pragmatic description

Anna Giskes

Master Thesis

Anna Giskes, student number 5934079 University of Amsterdam

Master Linguistics Faculty of Humanities

Supervisor: Dr. R. Risselada

Second reader: Prof. Dr. C. H. M. Kroon Trondheim, August 27, 2016

(2)

(3)

The Latin particles Quidem and Equidem:

A discourse pragmatic description

Anna Giskes

August 27, 2016

(4)

(5)

Acknowledgments

An unfinished Master thesis is a reality I lived with longer than I had planned or expected. Since I first picked the subject for this thesis, I emigrated twice, learned Danish, got married, and biked over 3 000 kilometers through Europe. It fills me with joy that this thesis now finally came together.

Firstly, and most importantly, I thank my supervisor Rodie Risselada. Her relentless enthusiasm, that sparked my love for linguistics in the first place, has been inspiring in the process of writing this thesis. I have very much enjoyed the hours that we spent discussing versions of my thesis. Her comments made my thoughts and analyses sharper every time. She gave me the confidence to develop my research in the direction I wanted to. I am extremely grateful for her patience, time, support, and willingness to solve so many practical problems. Without her, this thesis would not have been finished at all.

I am also grateful to my husband Jeroen for putting up with a quite unpleasant version of me during the last period of writing. His support is invaluable to me, as is his ability to put things into perspective when I struggle to do so. I further want to thank him for taking care of the layout of this thesis: he made it look as pretty as it does.

I started my Bachelor Greek and Latin at the UvA with a group of incredible people that later formed the ‘Leesclub.’ They became my friends and had an infectiously serious attitude when it came to studying. Being certain to find so many friends in the library every day made writing this thesis much more pleasant. Also after emigrating, their friendship (and ability to look things up for me in the library of the UvA) has been important to me.

Finally, I want to thank Caroline Kroon for taking the time to read this thesis as a second reader at short notice. I hope that I did justice to her work, which has been a great inspiration for this thesis and many assignments I wrote during my Master.

(8)

(9)

Chapter 1

Introduction

When it comes to describing the meaning and functions of words, particles may be the most difficult category. Even native speakers often struggle to explain ‘the’ meaning of particles in their own language. Apart from the fact that particles do not have a referential meaning, they tend to have different meanings in different situations. Depending on one’s goals, a researcher has to determine his position between the extremes of describing each meaning in each concrete situation, and describing what unites all meanings and functions. The latter option sounds as a reasonable attempt to gain understanding of a particle’s nature; in practice, however, it proves to be complicated to pinpoint what it is that unites all meanings and functions. In the introduction of the monograph on quidem by Solodow (1978), a hint of all too familiar despair shimmers through the surface:

“The main difficulty is that the categories they [i.e., previous re-searchers] establish have no center, nothing in common, nothing that persuades us that the word is a unit. The description of the several categories might as well be of so many different words. Ordinarily we ought to employ a kind of Ockham’s razor and assume that one word or grammatical form has one basic meaning, from which the origin of others can be understood. This assumption is reasonable unless we know of syncretism or suspect it. [...] But no such suspicion attaches to quidem, and we are rightly bewildered by the unrelated categories of its use that are postulated.” (pp. 1–2)

In most dictionaries, the particle quidem indeed comes across as a multi-purpose tool that can shape-shift into sometimes almost opposite meanings, adapting to the needs of the writer. Lewis and Short Latin dictionary, for instance, tells us that quidem can be translated as ‘at least,’ just as easily as ‘however,’ ‘surely,’ ‘for instance,’ ‘for my part,’ and ‘indeed’: a remarkably differing set of translations for one word. And if one decides to consider equidem a mere variant of quidem, this leads to even more translations and possible functions.

The research of (e)quidem through the ages is characterized by a movement away from this potpourri of different meanings in the direction of one ‘basic meaning.’ Thus, unlike e.g. Solodow (1978), more recent studies by Kroon (2005, 2009) and Danckaert (2014, 2015) have a unifying approach. However,

(10)

1.1. RESEARCH QUESTION AND METHOD

formulating a basic meaning often means that a price has to be paid: a loss of specificity or a loss of all-inclusiveness. I will amply discuss the basic meanings of quidem suggested by Kroon (2005, 2009) and Danckaert (2015) in this thesis. The work of Kroon has functioned as a starting point for the research process. Her basic meaning is an elegant construction that is arguably valid for the majority of instances of quidem. However, applying this basic meaning to the instances in my corpus, I conclude that it has its limitations, and possibly does not grasp the essence of quidem. The work of Danckaert, on the other hand, provides a very specific basic meaning, but one that cannot be applied to all cases of quidem. What is more, his claim is not supported by adequate evidence. Nevertheless, Danckaert’s main claim will form a major part of my definition of quidem.

The studies addressing quidem either disregard equidem or consider it a variant of quidem. Still, some of them do note that equidem sometimes behaves differently. Apart from its preference for occurring with verbs in the first person, it is noted by Solodow (1978) and Danckaert (2015) that equidem somehow turns the attention towards the Speaker. I approach equidem as an independent particle, and suggest that it has a basic meaning of marking commitment.

1.1 Research question and method

The main question that I will investigate in this thesis is: What are the functions of the particles quidem and equidem? Being far from the first researcher to address this question, I start my thesis by investigating the findings of previous studies. Although these studies do not, in my opinion, satisfactorily answer this question, they do provide information about distribution and other observations, accompanied by a collection of allegedly typical and atypical instances of quidem and equidem. These previous studies provide an orientational frame of reference for the next step: the analysis of the functions of both particles in my own corpus. The analysis is based on the groundbreaking model of discourse analysis that Kroon (1995) developed for the description of particles. Although her work is focused on connective particles, the framework is suitable for the description of other types of particles as well, as she herself has demonstrated in her analysis of quidem (Kroon, 2005), and Schrickx (2011) has demonstrated for a number of commitment markers such as nempe, quippe, and videlicet. I have tested my corpus using a number of parameters. Some of them are suggested by Kroon’s framework, others are aimed at testing the hypotheses from previous studies, and finally, there are follow-up parameters that help me further define the particles’ discourse functions. A full list of parameters is provided in section 1.3.1.

Kroon’s research method aims at formulating a basic meaning. Applying this method, I will in chapter 3 propose a basic meaning of marking commitment for equidem. For defining quidem, however, the hierarchical model based on the concept of basic meaning proves problematic. When discussing the results of my research on quidem, I will, therefore, also address the question of how we can represent quidem. I will argue that quidem is best described in terms of a radial category, with a central meaning of preserving positive polarity, in line with the main claim of Danckaert (2014).

Finally, I will address the question of the relation between quidem and equidem. I essentially treat them as two different particles, based on the different discourse situations in which they occur, which enable pragmatic motivation

(11)

CHAPTER 1. INTRODUCTION

for different discourse functions. A comparison shows, however, that there is a certain overlap in the functions they cover.

My thesis addresses the following research questions: 1. (a) What are the discourse functions of quidem?

(b) What is the optimal model to describe quidem, in order to best represent its range of meanings and functions?

2. (a) What is the basic meaning of equidem? (b) What are the discourse functions of equidem?

(c) How does equidem relate to other particles that have similar discourse functions?

3. Is there a semantic-functional relation between quidem and equidem?

1.2 Corpus

Table 1 presents an overview of the corpus. The majority of the corpus is drawn from Plautus and Cicero. Half of the passages from Cicero come from the speeches; the other half from the letters to Atticus and the philosophical works. The corpus is complemented by a smaller number of instances of the particles by other authors.1 _{For a full list of instances, see appendices A.2 and A.3.}

Author Quidem Equidem

Plautus 77 71 Cicero Speeches 25 50 Letters to Atticus 39 10 Philosophy 11 0 Other 21 40 Total 173 171

Table 1: Composition of the corpus.

Cicero and Plautus complement each other in several respects: they cover different genres and stem from a different period. The choice for drawing my corpus predominantly from two authors is mainly a practical one. My analysis of quidem and equidem takes into account a broad textual context for each instance of the particles: In order to study the exact communicative situation and relation between Speaker and Addressee at the moment of speech, extensive knowledge of the surrounding text is required. Limiting the number of authors and texts made it possible to investigate this number of instances, within the limited scope of a master thesis. The group of other authors functions as a control group, and has the aim to minimize the possibility that certain observations are only to be ascribed to an author’s personal style, and not to the more general characteristics of the particles.

I have conducted the corpus research using the Brepols Library of Latin Texts. The examples displayed in this thesis are taken from their database. The translations that accompany the examples come from the Loeb Classical Library Series, unless stated otherwise.

(12)

1.3. CONCEPTS

1.3 Concepts

Before describing the results of the corpus research in more detail, I will discuss a number of theories and concepts that play a central role in my analysis of quidem and equidem. Firstly, I will give an overview of the discourse model developed by Kroon (1995), which is a crucial pillar my research is built on. Secondly, I will treat the conversational model by Verhagen (2005), which treats the concept intersubjectivity. Finally, I will briefly recall the classic study on conversational implicature by Grice (1975) and politeness theory by Brown and Levinson (1987).

1.3.1 Discourse units

My research is very much indebted to the work of Kroon (1995, 2005). Apart from her sharp observations regarding the behavior of quidem, I use her general framework for the description of particles as outlined in Kroon (1995). Rooted in the Geneva discourse model, this framework provides an insightful categorization of particles based on the discourse structure of text.

Kroon chooses to divide discourse in hierarchically organized units based on their communicative status, instead of on syntax. A syntactically oriented division of a text takes the sentence as a meaningful unit. Taking the commu-nicative organization of a text into account, Kroon does not divide discourse in sentences, but in Acts. The Act is defined as “the smallest identifiable unit of communicative behavior” (Kroon, 1995, p. 65). It is the smallest unit in communication. The unit Act regularly coincides with one whole sentence; it is a convenient vehicle for an Act. However, Acts can also be larger or smaller than one sentence. As Kroon (2005) points out, quidem is often placed in Acts that are larger or smaller than one sentence. I will discuss this characteristic of quidem in chapter 2.

On the level above the Act we find the Move. The Move is “the minimal free unit of discourse that is able to enter into an exchange structure” (Kroon, 1995, p. 66). By means of a Move a Speaker achieves a communicative goal, which can be reacted on as a whole. The qualification ‘free’ does not mean that it is unrelated to the preceding discourse; it rather stresses that it should be seen as a complete element that has an internal logic, and that does not require any addition at that particular point in the discourse.

A Move may consist of one or more Acts. When a Move consists of several Acts, there is usually an internal hierarchy between them: there may be a Central Act, in which the ‘point’ of the Move is made, and one or more Subsidiary Acts, which provide an introduction with background information or another form of support that makes the communication of the Central Act successful. In the example shown in figure 1, drawn from Kroon (1995),2 _{the first Move consists of}

respectively a Subsidiary Act and a Central Act.

In discourse containing more than one speaking participant, a Move often coincides with a turn in the conversation. Moves may invite the interlocutor to react. In discourse a more monological character, Moves typically coincide with a division in section. In persuasive genres, a Move may concern the elaboration of one argument, consisting of Acts that provide an introduction, a step-by-step

(13)

Figure 1: Example of a discourse structure, involving Act, Move and Exchange.

development of the argument and a conclusive statement. In narrative texts, a Move may cover a description, or a chain of states of affairs that together build up a step in the development of a story.

The advantage of considering the communicative structure of a text is that it enables us to identify the type of discourse unit a particle occurs in. For example, the particle equidem has, unlike quidem, a preference for placement at the beginning of the reactive Move in an Exchange. This is a linguistic clue that, as we will see in chapter 3, supports my definition of equidem.

1.3.2 Coherence and levels of discourse

Halliday and Hasan (1976) address the question of what distinguishes a stretch of discourse from a collection of words. To begin with, they observe that discourse is cohesive through co-referentiality; by using, e.g., pronouns, demonstratives, and definite articles, elements of a text are tied together. Furthermore, they note that discourse is always ‘about’ something: it depicts a story world that has coherence by containing a certain consistence in characters and events. Apart from this story world, discourse also deals with the discourse situation, in which a Speaker addresses an Addressee. These observations form the basis of the three levels of discourse that Kroon (1995) distinguishes: the representative level, the presentative level, and the interactional level.

Connections on the representational level of discourse are representations of connections in the story world that a stretch of discourse describes. This layer contains the manifestation of a Speaker’s interpretation of the real world or, in the case of fiction, an imagined world. Causality and chronology are the most straightforward types of connections on the representational level of discourse.

The presentational level of discourse deals with how the represented proposi-tional content is presented. Here, we find a Speaker’s choices as to the discourse structure. It is the Speaker who adds internal relations between Acts and Moves and decides to present certain information as part of the foreground or background, or how subsequent Moves are related to one another. In terms of discourse units, on the presentational level of discourse the function of an Act in a complex Move is considered. Kroon claims that the particle nam has an important function on the presentational level, signaling that the information in its host unit should be considered subsidiary.

Finally, the interactional level of discourse concerns the relations between Speaker and Addressee, and the situation the discourse takes place in. On this level, conversation management takes place. The Participants take turns,

(14)

1.3. CONCEPTS

reacting on each other’s Moves, and handling each other’s expectations. An example of a predominantly interactional particle from Kroon’s research is enim. With a basic meaning of ‘consensus,’ enim signals that a Speaker expects the Addressee to agree with respect to the content of the host unit.

Unlike Halliday and Hasan, Kroon does not consider textual cohesion brought about by means of co-referentiality to be a system that causes coherence; Kroon views the system as the result of coherence in the story world and discourse situation, see Kroon (1995, p. 60). Her presentational level of discourse does not concern textual cohesion as Halliday and Hasan describe it, but with a Speaker’s choices as to how to translate a concept into a stretch of discourse.

Both quidem and equidem have functions on the interactional level of discourse. In order to see where the differences between the two particles are situated, I propose to use a more concrete picture of the processes that take place on the interactional level of discourse. For that purpose I have used the concepts introduced by various other authors and theoretic approaches, which I will discuss in the following sections.These concepts pertain in particular to the interactional level of discourse.

1.3.3 Intersubjectivity

Verhagen (2005) proposes a discourse model3_{in which the concepts subjectivity}

and intersubjectivity play a central role. Subjectivity concerns the process of a speaker’s conceptualization that precedes an utterance. When a Speaker makes a statement about the world, he uses an internal model of the world that is based on his own perception, knowledge, and experience. A new piece of information about the world will be integrated in the internal model as it is perceived by the Speaker. It is not only registered as a certain state of affairs (‘it is raining’ or ‘Peter said he plays for Manchester United’), but also analyzed and linked to existing information in the internal model, resulting in an interpretation of the state of affairs and how it relates to other events and knowledge about the world (e.g. ‘it is still raining,’ or ‘Peter is lying’). Insofar as an utterance is a representation of the world, it is a conceptualized representation, having been provided with relations such as chronology, causality, probability, desirability, etc. Verhagen coins this process of conceptualization the viewing arrangement, which he visualizes as shown in figure 2.

Figure 2: The viewing arrangement, cf. Verhagen (2005, p. 5).

(15)

The vertical line represents the relation between the Viewer and states of affairs in the world (objects of conceptualization). On this line, we find for example deixis, and the Viewer’s point of view in a broad sense. The horizontal line concerns the relation that a viewer establishes between states of affairs, for example causality. Both the horizontal and the vertical line are part of the Viewer’s internal model, and do not necessarily depict ‘real’ relations between states of affairs.

In order to describe interaction, the Addressee needs to be added to the viewing arrangement as a second conceptualizer.4 _{This results in the ‘construal}

configuration,’ which Verhagen suggests as a model for human interaction. This configuration is based on the claim that all human linguistic interaction is essentially argumentative. Although language gives us the ability to be merely informative, that is, stating (interpreted) facts about the world, interaction is always oriented towards the wish to influence others and their internal model of the world. In order to do so, a Speaker takes into account the Addressee’s internal model of the world, as is depicted in figure 3.

Figure 3: The construal configuration, cf. Verhagen (2005, p. 7). The lower horizontal line represents the Ground: a shared version of the internal model of the world, containing knowledge that is assumed to be mutual, about the world, each other, and the situation at the moment of conversation. The Ground is the basis for a conversation. As Verhagen puts it, “the point of a linguistic utterance [...] is that the first conceptualizer invites the second to jointly attend to an object of conceptualization in some specific way, and to update the common ground by doing so” (Verhagen, 2005, p. 7). The term intersubjectivity, as defined by Verhagen, concerns a Speaker’s awareness and use of the Ground during conversation.

The argumentative nature of interaction is a part of the Ground. That means that both Participants are aware of the fact that a linguistic utterance comprises an argumentative intention: the Addressee will try to infer the argumentative intention. In other words, an Addressee assumes that an utterance never comes ‘out of the blue.’ Verhagen supports this claim by a discussion of possible

argumentative implications of the following sentence: (1) There are chairs in the room.

4_{I assume that language is always, and primarily, intended to reach, and be understood}

by a (fictive or real) Addressee. For an overview of the discussion on this question, see e.g. Mosegaard Hansen (1998, pp. 176–180).

(16)

1.3. CONCEPTS

These words would only be uttered if the presence of chairs in the room would be relevant for achieving a communicative goal. Depending on the situation, this utterance may bring about the suggestion to use the chairs, as becomes clear in the following two possible additions:

(2) There are chairs in the room. Moreover, they are comfortable. (3) There are chairs in the room. However, they are comfortable.

Although one could imagine a situation in which addition (3) would be sensible, option (2) sounds more likely to occur. The point made by Verhagen is that lan-guage use is never purely informative. The argumentative aspect of conversation is not a possible addition to the informative layer; it is rather an inherent part of linguistic interaction.

1.3.4 Conversational implicature

Verhagen’s claim that all language is argumentative finds its origin in the phenomenon conversational implicature, which was developed by Grice (1975, 1978). These studies have played a fundamental role in the development of the field of discourse pragmatics.5 _{Conversational implicature deals with the relation}

between what we (literally) say and what we mean, in other words, that which is implied by our words.

(4) john: How are your cooking skills? mary: Well, my children are still alive.

Here, Mary does not literally answer the question that is posed by John. Instead, she implies the answer by a comment. The comment is only relevant as an answer if the state of Mary’s children provides information about her cooking skills, which is the case if (1) Mary usually prepares the food her children eat, and (2) the fact that her food is edible is distinctive. The latter is only true if Mary’s cooking skills are dreadful, and, on a scale containing all possible direct answers to John’s question, her food is close to inedible (see figure 4).

Figure 4: Scale of possible direct answers to John’s question.

In conversation, we are very skilled in this ‘reading between the lines,’ and understanding the implied message that an utterance brings about. We are successful as interpreters, because conversation follows a set of ‘rules,’ known as the four Maxims of Grice:

1. Maxim of Quantity: make your contribution as informative as is required. Do not give too little or too much information.

(17)

2. Maxim of Quality: try to make your contribution one that is true. 3. Maxim of Relevance: be relevant.

4. Maxim of Manner: be perspicuous.

Together, these maxims constitute the cooperative principle, which states that the contributions Participants make in a conversation are actual contributions that help the conversation move in a certain direction or towards a certain goal. Utterances are assumed not to be isolated events; they are interpreted in the light of the situation, the conversation, and the stage of the conversation. In the example above, the interpretation on Mary’s words is based on the assumption that she is answering the question. It is important to note that this is a principle, and not a law. It predicts how people generally behave in conversation, and therefore how Participants expect their interlocutors to behave: it forms a guideline for interpretation by an Addressee.

Intersubjectivity plays an important role in the execution of the cooperative principle. In order to behave according to the maxims of Grice, a Speaker needs to make use of the Ground and his knowledge of the Addressee’s internal model of the world. On the other side, the Addressee uses the Ground in order to interpret utterances as behaving according to the maxims.

Conversational implicature consolidates coherence, adding an element to the concept of cohesion and coherence. Discourse is characterized by coherence and cohesion. The role of conversational implicature is to fill in any coherence that is not explicitly presented in the text, making it fit the Ground and assigning it a (role in a) communicative goal. Our expectation of coherence causes us to create coherence even if it is not (explicitly) there, or if it is simply not there.

The implications of the pro-active attitude of the Addressee pose a com-plication for research of the discourse function of particles. In some cases, it is difficult to find falsifiable evidence for a discourse function. Since a specific type of connection between two Acts, Moves, or other units may be implicit, its presence is, in principle, always a possibility, even though it might be more probable in some situations than in others. Part of these difficulties can be overcome by using linguistic clues as additional evidence, as Kroon convincingly demonstrates in her dissertation (Kroon, 1995). Furthermore, the problem can be minimized by defining the context in which a certain function may occur as specifically as possible. I will discuss this problem of evidence more extensively in the treatment of quidem in chapter 2. Furthermore, I will zoom in on certain aspects of Verhagen’s model, which will illustrate how this model can be applied to specific discourse situations.

1.3.5 Politeness theory

Politeness is an influential mechanism for the course of a conversation. Polite-ness is the central phenomenon in politePolite-ness theory, which was developed by Brown and Levinson (1987, 1978). This fundamental theory is based on the assumption that we do our best in conversation (and otherwise) to acknowledge and respect the interlocutor’s privacy and integrity. Brown and Levinson use the anthropological term face, which they define as ‘the right to be unimpeded in one’s actions (negative face) and the right to be approved of (positive face)’

(18)

1.3. CONCEPTS

(Brown and Levinson, 1987, p. 13). In conversation, we try not to violate these rights, and expect the same from our interlocutors.

However, violating someone’s face to some degree is unavoidable in conver-sation; one could even go as far as saying that it is inherent in conversation. Speaking to someone means demanding his or her attention, which mildly vi-olates the right to be unimpeded. So-called face threatening acts (FTAs) are thus impossible to completely avoid, even though they disagree with the aim to respect one’s face.

In order to have a successful conversation and maintain the relationship with the interlocutor, we use politeness strategies to redress FTAs. There is a wide range of strategies, which have in common that they signal that the Speaker is aware of performing an FTA, and does not take it for granted that the Addressee will accept it. This is the reason why politeness strategies are part of the Ground: they require the Speaker to take into account the perspective of the Addressee. In example (5), both Speakers use politeness strategies to redress the FTAs they commit. This is not a complete list of politeness strategies, but an illustration of how they work.

(5) a: Could you print the documents for me, if it is not too much trouble?

b: I’m sorry, I don’t have a printer. Shall I ask my neighbor whether he can do it for you?

The FTA here is a request to do something. This threatens the Addressee’s negative face, since it violates his right to be unimpeded. The politeness strategies that are applied by A are:

1. using an interrogative illocution instead of an imperative, which gives B the possibility to opt out;

2. using could instead of can, which presents the question as hypothetical, and not as a direct request;

3. minimizing the FTA by stating the condition if you have time. At the same time, this addition provides B with a reason (excuse) not to fulfill the request.

B’s answer contains also an FTA. By not granting A’s request, B neglects A’s positive face. He makes use of the following politeness strategies:

1. giving motivation for turning down the request;

2. signaling that he is unable to grant the request, not unwilling; 3. apologizing;

4. offering an alternative action (showing good will).

In some cases, marking one’s commitment to the truth of a statement may be a politeness strategy. When turning down a request or invitation, a politeness strategy of signaling that one is nevertheless enthusiastic about the proposal can be strengthened by using commitment markers: “I REALLY would have loved to join, but...” In chapter 3, I will show how equidem may be used as such.

(19)

1.4 Heuristics

In the description of particles, Kroon (1995) strives for explaining all behavior of a particle as stemming from one basic meaning. The basic meaning may be applied on different levels of discourse, and be utilized for different communicative goals. With this model, a polymorphic landscape of functions and meanings can be neatly presented as different appearances that branch out from one basic meaning. Kroon observes that apart from the actual uses inside the branches, a basic meaning may bring about side effects. Side effects are all kinds of features with which a particle frequently occurs because its basic meaning forms a good combination with these features; not because the feature is a manifestation of the basic meaning. An example of a side-effect is discussed in Kroon’s analysis of enim. Enim often occurs in Acts that have a connection of causality with the preceding Act. However, enim does not express causality; its basic meaning of ‘consensus’ causes enim often to be used in Subsidiary Acts that give a reason or motivation. The particles that are treated by Kroon (1995) are suitable for the top-down model ruled by one basic meaning; for each particle in her research, the model results in a clear picture. However, it is important to realize that the basic meaning is an abstract construction. It might be a useful construction to help us understand the behavior of a particle, and perhaps how actual uses came into existence, but it does not convey any ‘real’ meaning of a particle at a certain moment in time. Diachronically, the ‘actual use’ branches can become so far removed from the basic meaning that they do not take part in the basic meaning any more at least, synchronically speaking. It is, for example, possible that what originally was a side effect becomes reinterpreted by speakers as part of a particle’s meaning. Another possible scenario that would disturb the strictly hierarchical model of the analysis of a particle is that it takes over functions from another particle with which it has a certain overlap. Thus, looking for a basic meaning is an important step in forming an idea of what a particle does, but it is not necessarily the most insightful way to map a particle in all its manifestations. The particle quidem proves to be problematic to be meaningfully described within Kroon’s model. Therefore, I will suggest a less hierarchical representa-tion for quidem, although the process of attempting to define a basic meaning remains an important analytical tool. Inspired by Schrickx (2011) and Simon-Vandenberghe and Aijmer (2008), I will suggest a representation of quidem as a semantic map. The advantage of a semantic map is that it allows radial cate-gories: categories that have central and less central elements. Elements can be linked like a chain: an element is linked to another element, but not necessarily directly to the most central element.6 _{In anticipation of this representation of}

quidem using radial categories, I will be referring to quidem as having a central meaning, instead of the basic meaning from Kroon’s hierarchical model.

1.4.1 Linguistic clues

As Kroon (1995) points out, defining a particle’s basic meaning is a process of induction. The texts in the corpus present us with the actual use of the particle, at first sight indistinguishable from side effects and coincidental patterns. In

6_{For an elaborate description of radial categories, see (Lakoff, 1999, pp. 91–96). His case}

study concerns a noun; Mosegaard Hansen (1998) applies radial categories to discourse particles in French.

(20)

1.4. HEURISTICS

order to separate the side effects, I looked for common features in all instances of a particle in many different aspects of the particle’s occurrence, on the basis of a list of possible ‘linguistic clues’ which may point us in the right direction by sketching the outlines of a particle’s behavior. The parameters I have applied on the instances in my corpus contain the linguistic clues (“heuristische Indikatoren”) that are suggested by Schrickx (2011, p. 62). Her list is an adaptation of the list of linguistic clues that Kroon proposes. The list is complemented with a number of more specific parameters that are relevant for quidem and/or equidem. I will discuss these linguistic clues one by one, and point out how they could provide relevant information.

The parameters I will use in my corpus research are (parameters 1–6 after the linguistic clues by Schrickx, originally taken from Kroon (1995)7_):

1. Discourse type

2. Collocation with other particles 3. Sentence type

4. Communicative structure 5. Tense

6. Person

7. Position in the sentence 8. Narrative layer

9. Pragmatic motivation for commitment

10. Pragmatic motivation for preserving positive polarity 1. Discourse type

Discourse types8 _{are defined by two parameters, resulting in four groups. The}

first parameter concerns the distinction between monologal and dialogal texts, referring to the number of participants that take part in the discourse. In monologal texts, the Speaker is the only (active) participant. Since we have only written Latin at our disposition, most texts I deal with are essentially monologal: there is one author, who narrates his story as a single Speaker. This does not mean that monologal text has no Addressee; as I stated before, in my view, all

7_{Schrickx also proposes ‘genre and text type,’ ‘content,’ and ‘modality’ as linguistic clues. I}

do not use these as parameters in my research for the following reasons:

Genre and text type: in order to draw conclusions from a particle’s distribution among genre and text type, one requires a large and miscellaneous database. My corpus is too small and unbalanced regarding genre and text type, consisting of only few authors.

Modality: I do not use this utterly complex parameter because I do not need it for the description of quidem and equidem. The only aspect of modality that is relevant for these particles concerns commitment, which I adopt as a separate parameter in my research.

Content: the communicated content of a particle occurs with is an important linguistic clue, but it is too broad to create sensible categories that would enable an organized grouping of the corpus. There are, however, interesting tendencies as to content in the corpus, which I will discuss in the results sections for both particles.

(21)

text is intended to reach, and be understood by, an Addressee. This means that all discourse can be considered as part of a dialogue, although the addressee may not be present as a participant, and the rest of the large-scale implicit dialogue may never be realized. However, in monologal texts, there is no turn-taking between the Participants: the Addressee does not join the ‘conversation’ as a Speaker. In a dialogal text, there is more than one participant that produces the text. Plautus’ plays are an example of a written dialogal text.

The second parameter, monological/dialogical, does not have to do with the number of actual participants, but with the type of coherence in discourse, i.e., how Acts are connected. Monological discourse has the traits of a monologue, whereas dialogical discourse behave as a dialogue typically would. In dialogical passages, the discourse contains of Moves and Reactive Moves, uttered by different (embedded) voices; the coherence between the Acts has an interpersonal character. In monological discourse, being more inclined to narratives, the Acts have a connection on the Presentational and Representational level of Discourse. The focus is not on the interactional element of discourse.

Piecing together the two parameters results in four discourse types: monologal monological, monologal dialogical, dialogal monological, and dialogal dialogical. Monologal monological discourse and dialogal dialogical discourse are straight-forward: the discourse structure and number of voices matches the number of actual speakers. Monologal dialogical discourse occurs when a Speaker reports an embedded dialogue in his monologue: although there is one actual speaker, the discourse will consist of several voices uttering Moves and Reactive Moves. The other mixed discourse type, dialogal monological discourse, is found in co-narration. When two participants together tell one story, taking turns or finishing each other’s sentences, the discourse structure is not interactional, but presentational/representational, in spite of the presence of two alternately speaking participants. As Kroon (2009) observes, quidem is relatively often found in dialogal monological texts (see section 2.1.2).

In most discourse we find switches between discourse types. I let the local discourse type prevail over the overall discourse type in case of a discrepancy. For supporting my claims, the distinction monological/dialogical is more informative than the distinction monologal/dialogal, because it provides information about the levels of discourse a particle acts on.

2. Collocation with other particles

Collocation with other particles, adverbs or conjunctions should be distinguished from fixed particle combinations. Frequent collocation with other particles might make a certain interpretation of a particle more probable or improbable. For example, a modal particle marking uncertainty will not frequently be found in combination with particles marking commitment or consensus. However, fixed combinations should be excluded from the research, at least initially. The combination might have a meaning that differs from the sum of both elements. In my research, I excluded si quidem (also written as siquidem) and ne quidem. 3. Sentence type and illocution

Pinkster (2015, pp. 306–307) distinguishes four sentence types in Latin: declar-ative, interrogdeclar-ative, imperdeclar-ative, and exclamative sentences, each bearing a

(22)

1.4. HEURISTICS

corresponding ‘basic’ illocutionary force. In some cases, the sentence type does not match the actual illocutionary force. Sentences of the interrogative type may, for instance, be interpreted as a directive. Pinkster calls these cases ‘indirect illocutions,’ as opposed to ‘basic illocutions,’ in which the illocutionary force matches the sentence type. I will only distinguish basic illocutions.

4. Communicative structure

The recursive character of the Act-Move structure makes it difficult to pinpoint immediately on which level a particle acts: an Act is also part of a Move and of an Exchange, so that a particle is always positioned somewhere in an Act, Move and Exchange. Nevertheless, studying the position of a particle in the communicative structure can be helpful.

To begin with, it is useful to work with a more specific description of the type of Act or Move a particle occurs in. In the case of Acts, there is the distinction between central Acts and subsidiary Acts; for Moves, I distinguish initial and reactive Moves in dialogical texts.

Another important aspect is the exact position of a particle within a unit, paying special attention to the positions at the beginning and the end of a unit. I assume that the best position for a particle to somehow modify a longer Move, creating a relation between that Move and some element in the text, story world, or Ground, is at the beginning or end of the Move.

5. Tense

Tense gives us not only information about the relation between states of affairs, but also about the relation between the Speaker and the communicated content. With a choice of tense, a Speaker places himself in a certain position, assuming a certain relation to the communicated content. Especially in narrative discourse, a choice of tense influences the presentation of a story. Adema (2007) developed a model of narrative discourse modes, in which tense is approached as a device to manipulate the (temporal) distance between the Speaker and the communicated content.

6. Person

As Schrickx (2011) notes, this parameter is interesting when studying commit-ment markers. Since commitcommit-ment markers operate on the interactional level of discourse, modifying the Ground, they are likely to occur with the first and sec-ond person. Equidem displays extreme preferences concerning Person, occurring almost exclusively with the first person: a characteristic that very concretely distinguishes equidem from quidem.

7. Position in the sentence

Although the position in the communicative structure is a much more interesting one for a discourse-oriented research, the position in the sentence is relevant because of the specific behavior of quidem and equidem: whereas equidem appears mostly as the first or second word of a sentence, quidem is never found in sentence-initial position. This has caused discussion about the differences between the two (see section 3.3).

(23)

8. Narrative layer

This parameter comes from Kroon (1995, pp. 111–113). Discourse consists of a recursive network of speakers and voices; it contains several layers. The first layer, where the speaker directly addresses the (distant) addressee in their own hic et nunc is the communicative layer. The narrated world is represented in the reportive layer. The content of the reportive layer refers to an illusionary world or situation, embedded in the communicative layer. The communicative layer is, in terms of time, location, and participants, confined by the reality of the com-municative situation. The embedded reportive layer on the other hand, may deal with states of affairs on different locations, times, involving different participants, and is not bound by reality. In some monological discourse, mostly narrative discourse, the communicative layer is completely implicit; in other, the Speaker in the hic et nunc makes himself known to some degree. By doing so, the Speaker can deliberately disturb the present of the presented world on the reportive layer, for example in order to give his opinion or directly address the Addressee.

A Speaker can choose to virtually position himself in the reportive layer as if it were the communicative layer, creating the illusion of a ‘live report.’ For the purposes of my research, it is irrelevant whether we are dealing with an actual reportive layer or an embedded reportive layer; in both cases, the Speaker behaves the same.9

9. Pragmatic motivation for commitment

Particles that express complete certainty are grouped together as commitment markers. Commitment markers express the Speaker’s commitment to the actual-ity of a proposition. Examples of commitment markers in English are definitely and undoubtedly.

Commitment markers are redundant in a default situation. This is a result of Grice’s maxim of quality, which states that we assume that a speaker (in normal interaction) is always committed to his utterances: that he believes in their actuality and has some evidence to support them. The speaker’s commitment is taken for granted, and expressions of certainty are in this respect superfluous. They consequently must mark something else, or the maxim of quality must somehow be invalid. Halliday and Mathiesen (2004) even go further, and propose that commitment markers are generally used to camouflage a lack of commitment: “we only say we are certain when we are not.” In this view, commitment markers,

paradoxically, mark lack of commitment.

Although this is a possible actual use of marking commitment, it is not its most frequently occurring pragmatic motivation. Simon-Vandenberghe and Aijmer (2008) and Schrickx (2011) propose challengeability as the most important prag-matic motivation: we stress our commitment when we have a reason to assume that the Addressee disagrees. We further find commitment markers when it is crucial for a Speaker to be believed, even though there is no direct evidence for the truth of a statement. This is the case in threats, promises, and pleas. Finally, as

9_{See also Adema (2007) on discourse modes in Latin and the base of a narrator. Allan (2009)}

also elaborates on this notion, distinguishing different narrative modes instead of narrative layers. The immediate diegetic mode is used by Speakers in (the illusion of) a hic et nunc; the displaced diegetic mode is generally used to display the reportive layer. The modes are characterized by a number of features that are helpful to discern the narrative layers.

(24)

1.4. HEURISTICS

is noted by Schrickx (2011), commitment markers are also used to underline irony. I will show that almost all instances of equidem are positioned in such contexts. In addition to the pragmatic motivation for marking commitment, I inves-tigated a number of formal features of commitment markers. Pinkster (2004, p. 192) notes the following features for ‘subjective modal adverbs, in which the speaker [...] takes personal responsibility for the content of the proposition and signals how certain he is about its truth’:

• they may be used as one word sentences in answer to an open question; • they do not occur in imperative sentences;

• they are found in affirmative and negative statements, and [...] the truth value of the statement is the same if the adverb is omitted;

• they cannot be under the scope of negation themselves;

• they are found in combination with other evaluative adverbs and adverbial phrases of the subjective type such as recte (‘rightly’) and falso (‘wrongly’); • unlike all other evaluative adverbs and many adverbs occurring on the first and second level, truth value adverbs are not found in the second member of a coordination pair of the type et feci X (‘and I did it in X manner’) or atque id X.

I will show that these features are valid for equidem.

10. Pragmatic motivation for preserving positive polarity

Danckaert (2014) claims that quidem (and possibly equidem) is a marker of what he calls verum focus: he considers quidem to be a particular type of focus particle, expressing preserving positive polarity. He uses the definitions from H¨ohle (1992, pp. 112–113), which I will briefly discuss here.

Polarity has to do with the question whether something is or is not the case. When polarity is marked, the actuality of a state of affairs is being specifically affirmed or negated; positive polarity affirms, as in (6), and negative polarity negates, as in (7).

(6) _a: Mary does not have a Facebook account.

b: (You are wrong,) Mary DOES have a Facebook account! (7) _a: Mary took a selfie on the beach.

b: No, Mary DIDN’T take a selfie on the beach.

Markers of positive polarity can be further divided in polarity reversing and polarity preserving. In (6), speaker B reverses the negative statement of speaker A. The reason for speaker B to make his statement is straightforward: he believes A’s statement to be false and states the opposite. Now, consider the following example, from H¨ohle (1992, p. 112, example 2):

(8) _a: Ich habe Hanna gefragt, was Karl gerade macht, und sie hat die alberne Behauptung aufgestellt, dass er ein Drehbuch schreibt. b: (Das stimmt,) Karl SCHREIBT ein Drehbuch.

(25)

Parameter Values

Discourse type monological/dialogical Collocation

Sentence Type declarative/interrogative/imperative/ exclamative

Communicative structure beginning of a Move/beginning of an Act/ end of a Move/end of an Act/other

Tense praesens/imperfectum/perfectum/

plusquam perfectum/futurum/ futurum exactum/ellips

Person 1st/2nd/3rd

Position in the sentence 1/2/3/4+

Narrative layer communicative layer/reportive layer

Commitment yes/no, + pragmatic motivation

Preserving positive polarity yes/no, + pragmatic motivation

Table 2: The data gathered for each instance of quidem and equidem.

The statement of speaker A is confirmed by speaker B: the positive polarity is preserved. Danckaert states that quidem signals this: quidem “takes as its argument a proposition p, and asserts that p is indeed true” (Danckaert, 2014, p. 127). I will discuss Danckaert’s claim and argumentation in chapter 2.

1.5 Concluding remarks

In conclusion, the data that I gathered for each instance of quidem and equi-dem are listed in table 2. In addition to these data, I paid attention to the communicated content.

In this chapter, I have presented the research question and described the research methods I have used. The corpus research has been set up following the framework from Kroon (1995). Further, I gave an overview of the concepts and theories that I use to support my claims. They provide models that I used to define the possible pragmatic motivation for quidem and equidem to occur. In the following two chapters, I will discuss the results of the corpus research; in chapter 2, I will address quidem, and in chapter 3 equidem. Whereas chapter 2 is solely concerned with defining quidem, chapter 3 will also address the comparison between quidem and equidem.

(26)

(27)

Chapter 2

Quidem

2.1 Previous research

In 1880, Guilielmus Grossmann published De Particula Quidem, an extensive description of quidem as it is used in Early and Classical Latin. He concludes that quidem has five different meanings: explanatory, concessive, continuative, affirmative or adversative. Not much changed in the next century: Kühner and Stegmann (1914) give a similar account of quidem in their Grammatik der Lateinische Sprache: they treat quidem as a ‘modal adverb,’ in the modal category “Gewissheit, Versicherung, Bekräftigung, Beteuerung, Bejahung.” They ascribe the following functions to quidem: emphasizing, restricting, adversative, con-cessive, motivating/confirming, and complementing/continuative. At first sight, this list seems even less coherent than Grossmann’s categories: the multitude and in some cases vagueness of the meanings makes quidem look almost too versatile not to cause confusion in communication. However, Kühner and Stegmann make one big step forward with respect to Grossmann, by proposing a common feature for all the functions of quidem: they ascribe a mainly emphasizing function to quidem. They specify that the expressed emphasis is always related to another, unemphasized element, which may be expressed or left implicit: “Es [i.e., quidem] dient zunächst zur Bekräftigung und Auszeichnung eines Begriffes im Verhältnis zu einem anderen, entweder ausgedrückten oder zu ergänzenden Begriffe. [...] Alle verschiedenen Bedeutungen liegen [...] an sich nicht in dem Worte, das immer seine hervorgebende Kraft behält, sondern ergeben sich jedesmal aus dem Zusammenhange.” (Kühner and Stegmann, 1914). Thus, they observe that part of the different meanings are contextually determined, rather than being an inherent semantic feature. This is a crucial observation for the development of the description of particles. Kühner and Stegmann assume a basic function of quidem, which all meanings have in common. Further, they mention the importance of the context: the context influences how quidem manifests itself, and which specific meaning it takes on. These observations form the basis for the model for the description of particles by Kroon (1995): in her terminology, there is a distinction to be made between basic meaning and actual use. I will now discuss the more recent studies on quidem by Solodow (1978), Kroon (2005, 2009), and Danckaert (2014, 2015).

(28)

2.1. PREVIOUS RESEARCH

2.1.1 Solodow’s account of quidem

Solodow (1978) methodologically builds on K¨uhner and Stegmann’s observations, although he criticizes them for leaning too much on Cicero’s texts, and for not displaying a satisfactory ‘theory’ (Solodow, 1978, pp. 10–11). What Solodow means by ‘theory’ only becomes clear when he presents his own research goals in the following: he aims for finding the “core of the word, the essential meaning that informs its uses.” He characterizes his work “centripetal rather than centrifugal.” His goal is to create as little categories as possible, and to connect the categories by the ‘real’ meaning of quidem.

The ‘real’ meaning attributed to quidem by Solodow is an elaboration of K¨uhner and Stegmann’s definition: “Quidem essentially emphasizes, as has often been recognized, but it does so in a special way, always with reference to something else. Quidem emphasizes one statement (or phrase or word) while directing attention to another which contrasts with the first, supplementing or modifying it” (Solodow, 1978, p. 13). In other words, quidem emphasizes, and it signals a connection between two elements.

Solodow distinguishes five different functions of the particle, which are in fact the types of connection established between the two elements:

1. contrasting quidem, which sets up the first half of a contrast;

2. adversative quidem, which occurs in the second member of a contrast; 3. emphasizing quidem, which emphasizes a word without a contrast being

explicitly mentioned in the text;

4. limiting quidem, which is close to the English ‘at least’;

5. extending quidem, which extends or intensifies a previous statement. I will give one example from each category. The examples are drawn from Solodow’s data.

(9) de loco nunc quidem iam abiit pestilentia, sed quamdiu fuit me non attigit. (Cic. Fam. 14.1.3)

As to my place of residence, the epidemic has now quidem disappeared, but while it lasted it did not touch me.

(10) benigne edepol facis. —immo tu quidem hercle vero. (Pl. Rud. 1569)

You are truly acting kindly. —No, it’s really yourself quidem. (11) restat Aurelia. hic quidem etiam praesidia habeo. (Cic. Phil. 12.23)

(after having discussed various roads): The Aurelian road remains. Here quidem I have even (etiam) protection.

(12) non me quidem faciet auctore hodie ut illum decipiat. (Pl. St. 206)

(after advice from the addressee to a third party): He will not by my advice quidem act so as to deceive him today.

(13) emissus aliquis e carcere. et quidem emissus per imprudentiam, ...

(Cic. Planc. 31)

Someone has been released from prison. And quidem he has been released out of ignorance, ...

(29)

CHAPTER 2. QUIDEM

That quidem has a preference for positions where a connective particle is expected, is well argued. The five categories form a good overview of the different types of connections quidem occurs with. The emphasizing quality of quidem is, however, less convincingly underpinned. Admittedly, proving emphatic force in written discourse is much less straightforward than proving a connective function; the expression of emphasis is generally associated with intonation, and it has no formal restrictions. Emphasis can be expressed for any word or larger unit, and it is not tied to formal features such as illocution.10 _{Thus, proving emphasis}

in written discourse is not obvious. Another complication is that a Speaker, by emphasizing one element, essentially evokes another element, which automatically becomes an unemphasized or even denied counterpart. Emphasis can therefore be expected to often occur in pairs of sentences, or elements, which are closely connected. Consequently, the frequent co-occurrence of a connective particle and emphasized elements does not necessarily prove emphatic force of the connective particle: emphatic elements and connective particles just feel at home in the same environments. If quidem is indeed a connective particle, the emphasis might be just a side-effect, and vice versa: if quidem has an emphasizing quality, the occurrence in closely connected pairs of clauses or ideas might be a side-effect.

In Solodow’s third and fourth category, where the contrasted element is not explicitly present in the text, we find a related problem regarding evidence: the emphasis lies necessarily with the quidem-clause, since the other clause remains unexpressed. We could even say that the implicit second element only exists by virtue of a certain form of emphasis. In some cases, if we do not assume emphasis—and emphasis is difficult to prove—there is no implicit second element. Solodow’s claim that quidem emphasizes is also weakened by its contradictory behavior in the different categories. In the category Adversative quidem, it appears in the second element, giving it emphasis; in the category Contrastive quidem, it appears in the first element, which “serves as a foil” to the following element, which is emphasized.11 _{This implies that emphasis has more to do with}

the order of the two elements than with the presence of quidem.

In conclusion, Solodow’s account of quidem is problematic in several respects. To begin with, the characteristics that Solodow ascribes to quidem are difficult to prove or disprove. What is more, Solodow does not address the problem of evidence. He does, however, give a clear overview and useful categorization of the contexts in which quidem occurs. Before continuing to the treatment of quidem by Kroon, I would like to draw attention to one more observation by Solodow. He claims that the truth of the quidem-clause is always insisted on (p. 13). Whereas concessive relations such as “He is not tall, but he is strong”

frequently occur with quidem in the first or second element, we should not find sentences displaying an opposition, such as “He is not tall, but short.”

As I will discuss in section 3.2, Solodow regards equidem as a variant of quidem, and argues that his description of quidem is fully applicable to equidem.

10_{Hengeveld and Mackenzie (2008, pp. 66–67) describe emphasis, for the above mentioned}

reasons, as a speaker’s intensification of a Discourse Act, functioning predominantly on the ‘interpersonal level,’ i.e., being part of the communicative process, more than part of the

discourse organization and the morphosyntactic structure.

(30)

2.1.2 Kroon’s account of quidem

Kroon (2005, 2009) investigates quidem from a functional point of view. She uses the framework that she created for the analysis and description of Latin particles in Kroon (1995). Kroon’s main goal here is not to give a complete definition of quidem, but to support a broader claim about grammar and discourse by the behavior of quidem in certain situations. She consequently only describes part of the behavior of quidem. She does, nevertheless, suggest a basic meaning. Kroon rejects Solodow’s assertion that quidem has emphatic force, and proposes a discourse organizational function: quidem marks a deviation from the default relation between grammar and discourse. Normally, Kroon states, grammar and discourse agree with each other: a discourse unit (Act) corresponds with a semantico-syntactic unit (sentence or clause). In some cases, though, these two layers do not show the same division. These cases, ‘mismatches’ between discourse and grammar, are marked by quidem. The host of quidem is on the one hand an independent Discourse act, but on the other hand, this Discourse Act forms a ‘conceptual unity’ with another Discourse Act that is semantically and/or syntactically closely connected. I will illustrate Kroon’s hypothesis with example (14) and (15), drawn from my corpus.

(14) tum das eas poenas quae solae sunt hominum sceleri a dis immortalibus constitutae. nam corporis quidem nostri infirmitas multos subit casus per se, denique ipsum corpus tenuissima de causa saepe conficitur: deorum tela in impiorum mentibus figuntur. qua re miserior es cum in omnem fraudem raperis oculis quam si omnino oculos non haberes.

(Cic. Har. 39)

You are then suffering that punishment which is the only one appointed by the immortal gods for the wickedness of men. For the infirmity of our bodies quidem is of itself liable to many accidents; moreover, the body itself is often destroyed by some very trivial cause; and the darts of the gods are fixed in the minds of impious men. Wherefore you are more miserable while you are hurried into every sort of wickedness by your eyes, than you would be if you had no eyes at all.

(15) _a: “Diabolus Glauci filius Clearetae lenae dedit dono argenti viginti minas, Philaenium ut secum esset noctes et dies hunc annum totum.” (Pl. As. 751–754)

b: neque cum quiquam alio quidem

a: (reading an agreement): “Diabolus, the son of Glaucus, has made a present to Cleareta, the procuress, of twenty silver minae, that Philenium may be with him night and day for this whole year.”

b: And not with any other person quidem.

The mismatch in the first example becomes visible when we display the underlying discourse structure of the passage, shown in figure 5. It conveys a Move, consisting of several Acts. Act 1 is a statement. It is supported by a Subsidiary Act introduced by nam, presenting an argument. Act 3 resumes the main storyline, and concludes the Move. Where Act 1 and 3 both consist of a single semantico-syntactic unit, Act 2 has two clauses with two foci. To

(31)

CHAPTER 2. QUIDEM

Figure 5: Discourse structure of Cic. Har. 39.

understand the message conveyed in Act 2, the reader needs both clauses. According to Kroon, quidem signals this ‘mismatch,’ pointing out to the reader that both clauses form a conceptual unity.

Example (15) demonstrates a different context in which a semantico-syntactic mismatch tends to occur: in monological dialogal discourse. The second speaker adds an extra specification to the utterance of the first speaker, extending the clause. Kroon describes the actual use of quidem in more detail by creating subcategories: local vs. global functioning quidem, and backward-linking vs. forward-linking quidem. The local quidem, operating on an Act, is found in (14) and (15), and in concessive environments, such as (16):

(16) est istud quidem Laeli aliquid, sed nequaquam in isto sunt omnia.

(Cic. Sen. 8)

There is something quidem in that objection, Laelius, but not everything. Kroon (2009, pp. 155–156) illustrates the global use of quidem, arguing that quidem may also act on the level of the Move.

Apart from being local or global, quidem may be forward-linking or backward-linking. Example (14) is forward-linking, and example (15) is backward-backward-linking. As also observed by Solodow, quidem may occur in the first or the second element of the connected pair of statements. Kroon notes that the forward-linking quidem is more common in written text than in (the illusion of) spontaneous speech, since it requires a degree of planning that is more suitable for written discourse. Kroon (2005, pp. 584–585) gives two linguistic clues that support her definition of quidem: (1) in dialogue quidem has a preference for monological dialogal discourse; and (2) embedded predications hosting quidem have a relatively low degree of syntactic integration. These clues are both focused on the specific, additive, use of backward-linking quidem that Kroon (2009) highlights, and do not represent all the facets of the behavior of quidem.

Kroon’s basic meaning of quidem as a discourse organizational particle agrees with a large number of occurrences of quidem. The problem with the definition is, however, that in many cases, it rather does not disagree; it is difficult to falsify. The crucial—perhaps unsolvable—point here is the absence of a concrete definition of the conceptual unity. Discourse is pragmatically coherent: every utterance must be interpreted in the light of the preceding discourse. As

(32)

Addressees, we are not only very good in perceiving coherence; we even expect and—if necessary—create coherence, and go to great lengths to interpret a text as meaningful and representing a coherent world. Grice’s maxims cause the Addressee to expect that the speaker’s utterance represents an element from a real or imagined world, and that the Speaker conveys his message in a way that the addressee will understand. The phenomenon of ‘six-word novels’ is a good illustration of how much we effortlessly infer during conversation. In a six-word novel, an author is challenged to tell a whole story in six words:

(17) Sorry soldier, I only sell pairs.12

A reader will assume that this sentence makes sense, and will fill in a context in which it does: the speaker sells things so we can assume (for the time being) that he is an employee of a shop; the most common thing that comes in pairs is shoes, so we could guess that it is a shoe shop; the soldier’s request is turned down, and the explanation of the rejection implies that the soldier’s request involves buying shoes not as a pair; a reason for wanting half a pair of shoes, is missing a leg. Since soldiers have a relatively high risk of losing a leg, it is a plausible scenario to explain the text. A reader automatically constructs a scenario such as this one, and will change it only when proven wrong.

We apparently have a strong tendency to interpret discourse as being cohesive, and representing a coherent world. Apart from the fact that discourse generally is cohesive and coherent, readers assume an extremely cooperative attitude, consolidating the text as a sensible whole.

The problem of the conceptual unity is that it does not have a sufficiently restricted definition. Without a more specific definition, it is difficult to imagine a situation in which one can argue that a Move or Act does not have a close connection to the previous or following element. Ruling out the presence of a conceptual unity is made even more difficult by the fact that we can choose from global, local, backward-linking, forward-linking, and linking-to-an-implicit-element. With so many possible second halves, we are bound to find one pair that we could imagine to form a conceptual unity.

A related problem when applying Kroon’s definition has to do with the question when we can qualify something as a mismatch. We would need more or less objective criteria that can qualify pieces of propositional content as suitable for being conveyed in an Act or Move; only then can we judge whether we are dealing with a mismatch, compared to what is ‘normal’ or ‘expected.’ I will illustrate these problems with a number of examples (18–22):

(18) qui si verus fuit ex dolore hominum et fame, certe consules causam suscipere, certe senatus aliquid consili capere potuit; sin causa fuit annona, seditionis quidem instimulator et concitator tu fuisti, nonne id agendum nobis omnibus fuit ut materiem subtraheremus furori tuo?

(Cic. Dom. 5.11)

And if that was the genuine effect of the grief of men suffering under famine, certainly the consuls had good reason to undertake the affair, certainly the senate had good reason to adopt some determination or other. But if the scarcity was the pretext, and if you quidem were the exciter and kindler of sedition, ought we not all to have striven to take away all shadow of pretext for your madness?

(33)

CHAPTER 2. QUIDEM

(19) solebat enim me pungere ne Sampsicerami merita in patriam ad an-nos sescentos maiora viderentur quam an-nostra. hac quidem cura certe iam vacuus sum; iacet enim ille sic ut phocis Curiana stare videatur.

(Cic. Att. 2.17.3)

(after claiming that the attitude of ‘indifference’ causes Cicero to be not troubled by feelings of pride or vanity): For it used to sting me to the heart to think that centuries hence the services of Sampsiceramus to the state would loom larger than my own. That anxiety quidem is now put to rest. For he is so utterly fallen that, in comparison with him, Curius might seem to be standing erect after his fall.

(20) _a: quaeso, quin tu istanc iubes pro cerrita circumferri? b: edepol qui facto est opus;

nam haec quidem edepol laruarum plenast. (end of the Move)

(Pl. Am. 775–776)

a: Why don’t you order her to be purified as a frantic person, please?

b: Really, somehow there’s need for it, for quidem she’s certainly filled with sprites.

(21) veniat quando volt, atque ita ne mihi sit morae. domist: non metuo nec ego quoiquam supplico, dum quidem hoc valebit pectus perfidia meum. abi intro, ego hic curabo. Tu intus dicito Mnesilochum adesse Bacchidi. (Pl. Ba. 224–227)

Let him come when he pleases, and so there be no delay. The money’s at home; I fear not for myself, nor do I go begging to any man; so long quidem as this heart of mine shall be armed with its inventiveness. Go in; I’ll manage here. Do you tell them in-doors, that Mnesilochus is coming to Bacchis.

(22) a: eccere, iam tuatim

facis tu, ut tuis nulla apud te fides sit.

b: quid est? quo modo? iam quidem hercle ego tibi istam scelestam, scelus, linguam abscidam. (Pl. Am. 554–556)

a: Look at that; you are now acting according to your usual fashion, to be putting no trust in your servants.

b: Why is it so? For what reason? By the powersquidem, I’ll cut out that villainous tongue of yours, you villain.

Example (18) is a fragment from one of Cicero’s speeches. It contains several possible conceptual unities: the fragment starts with a relative pronoun, creating cohesion with the preceding Act; there are two juxtaposed conditional clauses; and within the second conditional clause, there is a contrast between causa/annona and instimulator/tu. With Cicero’s preference for symmetry and solid argumentative structure, the possible conceptual unities are numerous, both with and without quidem.

In (19), the host Act of quidem is a comment on the previous Act. There is a high degree of cohesion between the two Acts, caused by the words hac cura

The Latin particles quidem and equidem: a discourse pragmatic description

The Latin particles Quidem and Equidem:

A discourse pragmatic description

Anna Giskes

The Latin particles Quidem and Equidem:

A discourse pragmatic description

Anna Giskes

August 27, 2016

Contents

Acknowledgments

Chapter 1

Introduction

1.1

Research question and method

1.2

Corpus

1.3

Concepts

1.3.1

Discourse units

1.3.2

Coherence and levels of discourse

1.3.3

Intersubjectivity

1.3.4

Conversational implicature

1.3.5

Politeness theory

1.4

Heuristics

1.4.1

Linguistic clues

1.5

Concluding remarks

Chapter 2

Quidem

2.1

Previous research

2.1.1

Solodow’s account of quidem

2.1.2

Kroon’s account of quidem