In his article “In Their Own Words: On Writing in Second Person,” Joshua Parker reflects on second–person narration and looks at the issue from the perspective of authors who use such narration in their works. In Parker's view, authors' self–commentaries may help us understand better the possible functions of second–person narration in fictional texts. Parker's main claim is that these authors are men and women “with professional experience as writers, who are capable of speaking quite eloquently on their own reasons for writing in second person” (167). One argument that seems to follow from this, although it is not expressly mentioned in the text, is that authors' viewpoints ought to be favored over narratological or other literary–theoretical approaches or ought at least to be taken more seriously than has hitherto been the case. As Parker puts it, there is “a surprising dissonance between what theorists often tend to assume about the form and what authors themselves experience in creating it” (167). He even proposes, somewhat tongue–in–cheek, a “writer response theory” in analogy to reader response theories (167). Parker presents authors' self–reflexive comments, quoting writers such as, among others, Chuck Palahniuk, Denis Johnson, David Foster Wallace, Pam Houston, Lolo Houbein, Peter Bibby, and John Encarnacao, who talked in interviews or wrote in non–fictional writing about their use of second–person narration. The main result of Parker's survey of these comments [→page 64] and of a number of texts written in second person is the following: “Seeing the self as 'other' often only takes place during descriptions of certain events or over periods of text. This self, like its experiences, is unstable. What is inscribed in second person, then, is the author's relationship to this self, a relationship often in flux” (171). Before I address Parker's main claims in more detail, I will outline four aspects that, to my mind, need to inform any research on writing in second person not only because they already appear individually or in combination in most scholarly work addressing this type of narration (e.g., Fludernik, “The Category of 'Person'”; Kacandes; Richardson) but also because they allow for interdisciplinary approaches to the topic (see Mildorf): 1. the anthropological dimension; 2. generic distinctions; 3. structural typologies; 4. functions and effects. Parker mixes up these aspects or does not follow them up assiduously enough, which explains why some of his claims are essentially flawed.

1. The Anthropological Dimension

Parker begins his article with the example of the cave paintings at Lascaux, arguing that “their author conceived of an experiencing point of view other than his own” and that he created these paintings “with the consciousness of designing images […] for an Other” (165). This is then linked by Parker to “what any writer working today might likewise pursue” (165). One can object to this associative connection by quoting Denis Dutton, who said that “[t]he state of the arts today can no more be inferred from looking inside prehistoric caves than today's weather can be predicted from the last Ice Age” (Dutton 3). It is also not unproblematic to link painting and writing without paying due attention to their respective medial expressivity. And one may question the underlying presupposition that art is always created for an “other.” Could I not simply paint or write for my own pleasure, without having any specific audience other than myself in mind?

Leaving these points of criticism aside, however, one can see in Parker's argument an attempt to bring into sharper relief something more fundamental concerning the relationship of human beings to [→page 65] fellow human beings, which is also expressed in the very use of the personal pronouns “I” and “you”: namely, that we are ultimately “relational beings,” as psychologist Kenneth J. Gergen has it. Gergen argues that “there is no isolated self or fully private experience” and that instead “we exist in a world of co–constitution” (xv). Somewhat paradoxically Gergen further claims that “even in our most private moments we are never alone” (xv) because we orient ourselves to what others say, presumably think or expect. The only trouble is (and here Gergen's argument also becomes political) that we are taught to think of ourselves as “bounded beings,” as isolated individuals. Applied to narrative theory, Gergen's approach seems to support the idea that literary as well as non–literary stories are always directed at (real or imagined) recipients, thus confirming the relationality of human beings.¹⁾ In this context, we might expect you–narration to be(come) more of a norm rather than the exception. The fact that this is not the case perhaps points to the predominance of what Gergen calls the “boundedness” of selves in Western thought and indeed to the predominance of self–centredness.²⁾

Other theorists in various disciplines have reflected on the relationship between “you” and “I,” and a comprehensive study of second–person narration would have to take these theories and positions into account. In this regard, linguistic approaches promise to be fruitful for analyzing the forms and functions of address terms and audience design (cf. Coupland 54–81; Mildorf “Second–Person Narration”), while psychological, anthropological and philosophical accounts may be helpful for the actual interpretation of these texts. However, one needs to be careful not to further muddy the waters, as it were. Parker himself refers to philosopher Martin Buber (who is also mentioned by Gergen when he reflects on writing as relationship, Gergen 221). This is not surprising as Buber's treatise “Ich und Du” (“I and You”) ponders on the inextricable coexistence, interdependence and reciprocity of “you” and “I.” Parker's reading of Buber leads him to describe the relationship between “you” and “I” in second–person narration as follows:

[→page 66] To write “I” in a text is not necessarily to underline a narrator's own existence as an enunciating source (an “I” can just as easily refer to the historical character of the first person narrator in the diegesis, described—at a distance—by the narrating voice). But when an author writes “you,” he insists on both a reader's existence and on his own, putting his narrator in relation to an Other, and defining his position as narrator by this relationship. (172)

This and other passages in Parker's article are difficult to understand not so much because of the complex theoretical issues implied but because Parker sometimes apparently uses technical terms such as “narrator” and “author,” “narratee” and “reader” interchangeably, or because he disentangles “narrator” from “narrative voice” (does he mean the distinction between “narrating and experiencing persona” here?). This conflation of categories is most visible in Parker's repeated claim that author and narrator are or become identical. For example:

A “you” addressed to the self creates alterity between a described situation and the enunciating voice, fortifying the author's identification with an extradiegetic narrator, helping to guide the story along during descriptions of trauma. (172; bold type my emphasis)

Of course the conflation of “author” and “narrator” matches Parker's quotations from various writers, in which they claim that they use second–person narration in order to distance themselves from what they write. However, Parker's argument along these lines not only suggests slippages in the use of theoretical terminology but also a lack of differentiation among generic categories. What kinds of primary text does Parker take into focus? This leads me to the next point.

2. Generic Distinctions

Another scholar Parker draws upon is Philippe Lejeune. This is surprising since Lejeune is mostly known for his work on autobiography. In his groundbreaking study Le pacte autobiographique, Lejeune argues that an absolute criterion for autobiographical writing, whatever form [→page 67] it takes, is complete identity of author, narrator and the person whose life is told.³⁾ This criterion separates autobiographical writing from, say, biographies or novels. Of course there are texts which play with generic categorization, e.g., when novels are cast as autobiographies or autobiographies turn out to be fictitious. The main difficulty in analyzing such texts, Lejeune says, is that one has to be careful to differentiate between “identity / identité” and “resemblance / ressemblance” (the latter is captured by Lejeune under the term “copie conforme,” 35). Interestingly, the primary texts Parker presents are not autobiographies per se. Some of them may contain autobiographical material if one can trust the self–reflexive statements made by Peter Bibby and John Encarnacao (Parker 171), for example. This does not make the texts autobiographical, though, and therefore to argue that writers use the technique of second–person writing to distance themselves from what they wrote (rather than saying that they distance their narrators from what is depicted in the storyworld) is imprecise, if not incorrect.

Parker refers to Lejeune's book Je est un autre in order to mount his argument about this distancing function of the second–person pronoun:

For many authors writing in second person seems to provide a middle ground, as Philippe Lejeune has conceived it (36–37), between the “owning” of an experience by writing in first person, and the stance of complete alterity from it implied by third person. […] Authors may use second person to treat subjects closely drawn from personal experience simply because second person allows themselves [sic] to hold an experience at a certain distance. (170)

Lejeune describes in the mentioned pages the various effects created through the use of different personal pronouns in autobiographies. He maintains that the first–person pronoun, like the second–person pronoun (!), seemingly glosses over the gap between “narrating I” and “narrated I” (for this terminology, see Smith and Watson): “Le 'je' (comme le 'tu') masque d'autre part l'écart qui existe entre le sujet de l'énonciation et celui de l'énoncé” (Lejeune, Je est un autre 37). Nevertheless, [→page 68] he continues to argue, we are not really duped by this play of pronoun use and still recognize author/narrator and character for who they are: “Naturellement nous ne sommes pas vraiment dupes de cette unité, pas plus que nous ne le sommes de l' 'altérité' dans le cas de la narration auto–diégétique à la troisième personne” (37). It is significant that Lejeune uses the term “alterity” with quotation marks, thus indicating that the distancing created through third–person pronoun use in autobiographies only functions as a mask but does not change the ultimate personal union among author, narrator and narrated persona. It becomes obvious that Parker's argument is flawed not only because he partially misconstrues what Lejeune writes but because he does not attend to the fundamental generic differences between the texts he discusses and the ones analyzed by Lejeune. Fiction written in the second person is a far cry from autobiographical texts and even from (fictional) texts employing (semi–) autobiographical material and must not be confused in a narrative–theoretical account.

So what can one take away from this discussion? A more comprehensive study of second–person narration has to flesh out the ways in which functions and effects of second–person narration are related to genre conventions and expectations. Second–person narration can be found in fictional and non–fictional texts; arguably, it also exists in conversational storytelling (Mildorf, “Second–Person Narration”). These different text types require attention to their specific design and to the ways in which they employ second–person narration, whether as a sustained mode of telling or only in passages. First and foremost, however, a study engaging in this subject matter would have to demarcate the kinds of text that are taken into focus and at least attempt to give a definition. Parker quotes Helmut Bonheim in order to delineate his object of study:

This article will not take up the traditional field of full–length “second–person fiction” texts […] but instead deal with cases falling under Helmut Bonheim's more open definition of second–person narration: narration in which “the 'you' is frequent enough in a section of text that the narrative effect is essentially modified” (168)

[→page 69] So when is the “you” frequent enough in a text to “essentially modify” it? And modify in what way? This is rather vague, and so is the range of historical textual examples Parker offers as predecessors of second–person narration at the beginning of his article: Beowulf is mentioned as well as Sterne and Fielding and, in a footnote, Francis Kirkman's The Unlucky Citizen (1673), Nathaniel Hawthorne's “The Haunted Mind” (1835), and Herman Melville's Moby Dick (1851). It is not difficult to recognize that Parker here lumps together kinds of you–narration which, if subjected to a rigorous typology, would not necessarily be all in one category. Not only does Parker ignore generic differences in his discussion but he also fails to base his examples on a stringent typology (let alone offer a conclusive typology himself), which, I think, is of the essence in a survey of second–person narration.

3. Structural Typologies

When Parker describes the above–mentioned historical examples as literature containing narrators that “have underlined both their own and readers' participation in texts by addressing us through apostrophe” (165), one is reminded of Irene Kacandes's concept of “Talk fiction” (see n1). Kacandes begins her book with the observation that there are literary texts which seem to engage their readers in much the same way as speakers in conversations interact⁴⁾: they ask their readers to “listen” to them and somehow expect or generate a “response.” Kacandes then investigates four modes of Talk in a range of (contemporary) literary texts and in relation to formats known as “talk radio” and “television talk shows” (12) as well as computer hypertexts and interactive video: “storytelling,” “testimony,” “apostrophe” and “interactivity.” It seems to me that the textual examples Parker provides throughout his contribution could be distributed across at least the first three of these categories, with the historical examples fitting the “storytelling” mode. This or any such typology would have made Parker's discussion more systematic and would potentially have strengthened his claim that writers implicate themselves in their texts [→page 70] and that they address themselves to an implied audience by means of the second–person pronoun.

In a footnote, Parker refers to David Herman's (“Textual You”) notion of “double deixis.” However, he presents an extremely simplified account when he explains the term as meaning that “the pronoun 'you' simultaneously refers to both a character and the narratee” (174n2). In fact, Herman assigns five possible functions to the second–person pronoun: 1. generalized you (like German “man” or French “on”); 2. fictional reference (to a character); 3. fictionalized (=horizontal) address (to a narratee); 4. apostrophic (=vertical) address (to the reader); 5. doubly deictic you (which combines at least two of the previously mentioned possibilities and thus creates ambiguity). What function the “you” assumes in a given text will ultimately also depend on who is (potentially) addressed by this “you.” It seems to me rather reductive to say that second–person narration often supports a distancing function merely because authors rationalized their own use of the technique in these terms. I will come back to this point below.

Even though Parker's main objective is precisely not to present types of second–person narration but instead what authors of such narrations say about their writings, at least some narrative–theoretical considerations might have helped Parker to avoid some of the terminological imprecision I mentioned above. Other typologies and explanatory accounts might have been useful in this context, e.g., James Phelan's (“Self–Help”) rhetorical approach with its differentiation between a textual “narratee” and a wider “narrative audience” and Monika Fludernik's (“Second–Person Fiction”) classification of second–person narratives into “homocommunicative” (i.e., members of the communicative level such as narrator and narratee are also protagonists) and “heterocommunicative” (i.e., the communicative and storyworld levels are kept separate). More recently, Fludernik (“The Category of 'Person'”) has further specified her typology and has provided useful graphic presentations for the various kinds of relationships between “you” and “I” on the discourse and story levels. [→page 71] Although she in the end admits that “texts deploy a variety of constellations that suggest a sliding scale between you and we narratives” (“The Category of 'Person'” 122), she identifies six basic types of you–narration: 1. reflectoral you narrative; 2. non–communicative I–and–you narrative; 3. first–person narrative with you protagonist; 4. homodiegetic you narrative; 5. self–address narrative; 6. communicational I–and–you narrative (107–13).

One may debate whether it is possible, as Fludernik suggests, that there is narration without a communicative level.⁵⁾ This will depend on whether one is willing to embrace a poststructuralist paradigm that allows for “narratorless” narration. What this typology makes sufficiently clear, however, is that one cannot operate on the level of narration and on the level of communication between author and reader (whatever that may be) at the same time. For theoretical but also practical, analytical purposes these levels need to be considered separately. Having said that, narratologists do of course think about what potential effects techniques such as second–person narration may have on readers, and some even consider the role real–life authors play in the conception and anticipation of such effects.⁶⁾ I will explore these points in the next section.

4. Functions and Effects

In her 2011 article, Fludernik also comments on the ways in which readers may respond to you-narration:

[I]n many you texts the foregrounded address function implies the existence of a person who utters these exhortations, comments and commands. To the extent that the (real) reader initially feels directly implicated, he or she will also take that voice as emanating from a real person, i.e. the author. Only when the fictionality of the text has been established does the reader move on to a reinterpretation of a text–internal, though extradiegetic, communicational setup, recognizing the speaker as a narratorial speaker without an existential link to the real world. (119)

[→page 72] The assumption that real readers may at least to some extent feel addressed through the pronoun “you” features in most, if not all, accounts of this narrative technique. One may object to Fludernik's comment that not every reader will initially feel implicated and will take the narrative voice to belong to the author. Second–person narration is, after all, still a marked narrative technique and therefore potentially alienating for readers. At the same time, not every reader who did initially feel “spoken to” by the author will necessarily reinterpret the communicative situation as located on the storyworld level and identify a story–internal narrator as soon as markers of fictionality become obvious. How different readers respond to you–narration is likely to depend on their expectations, their previous reading experiences and perhaps training. Kacandes's book Talk Fiction suggests that quite a number of readers (and not only “naïve” ones) will feel “talked to” by certain fictional texts and that, moreover, there are types of text which invite this kind of response. Kacandes's initial anecdote of a student who read Calvino's If on a Winter's Night a Traveller as being “about him” also points out the possibility of a more sustained sense of “identification” in readers. Parker argues along similar lines when he writes:

It is always a memorable moment when we, as readers, identify with something in a literary text. Perhaps even more memorable is the moment in which we can say not, “That's me!” but instead “it could be …”—something which second person texts, much like slips into second person in oral narration, would seem to promote. (173)

I would contest the rather imprecise, everyday use of the term “identification,” though. Is it really true that (adult) readers identify with characters? In other words, do we suspend our awareness of being in the process of reading and then merge with the presented characters in our minds? What rather seems to be the case is that we enjoy being allowed to imaginatively peep into other “people's” lives, to empathize with or feel sympathetic towards characters' predicaments (see Sklar) and, yes, perhaps to see some of our own feelings or circumstances reflected in the novelistic presentation, which in turn may [→page 73] make us think about our own lives and so on. This, however, is not the same as “identification.” In that sense, I also disagree with Parker's contention that readers “imagine the other as self” (173; my emphasis). Admittedly, the term “identification” has been widely used in literary studies when discussing readers' responses to characters. However, it has also been criticized for being misleading and imprecise (see Schneider 613; and also Eder, Jannidis and Schneider 47). Perhaps a better term to use is “immersion,” which captures the way in which readers can become engrossed in storyworlds spatially, temporally and also emotionally (see Ryan 89–119). I also disagree with Parker's assumption that second–person narration promotes identification “much like slips into second person in oral narration.” The real–world deictic and referential frameworks in conversational settings lend those conversations a fundamentally different ontological status. Especially when interlocutors share the same time and space and talk face to face (and even when they do not as, for example, in email communication), the ascription of the personal pronouns “I” and “you” is quite different from a situation in which I take a book into hand and may not be able to decide with certainty whom the “you” addresses. Kacandes's “Talk” as metaphor makes sense, taken literally it does not.

Even though it is not unproblematic to guess what the reader (this construct of an idealized reader whose reading experience in most cases is based on the scholar's own reading) feels or thinks in relation to the narrator and/or author, Fludernik's comment above is instructive insofar as it demonstrates that the onus of making sense of a text and of attributing or not attributing the text's message(s) to (an image of) a real author is on the reader. This is presumably the kind of theorizing—which concentrates on reading responses—that Parker implicitly criticizes. However, there are literary–theoretical studies which reinstitute the author in an analytical framework. A somewhat radical attempt is made by Andreas Kablitz, for example, who argues that one should give up the strict usage of the term “narrator” in lieu of “author” in narrative analysis except where it is obvious that a narrator [→page 74] persona has been created. However, Kablitz cautions us against sliding back into what he calls “biographism” / “Biographismus” (42), a danger I see in Parker's approach. James Phelan's work is another case in point. In his book Living to Tell about It, for instance, Phelan explores what he terms “character narration” by paying attention to textual design and its possible effects on readers. He uses the theoretical concept of “implied author” to unravel the ways in which literary texts can give out discrepant messages to readers or contain seemingly redundant information addressed to the narratee, thus displaying the “author's need to communicate information to the audience” (Phelan, Living to Tell 12; italics original). Still, one cannot “know” the real–life author and his or her intentions, Phelan argues, and his discussions of fictional and non–fictional texts show that textual effects are subject to interpretation. This is why his approach is ultimately also an exercise in reader response (as all literary analyses presumably are).

So, is Parker justified then in arguing for a “writer response” approach? Writers' own comments have been used in author–centered, historical literary studies, where interpretations of texts are triangulated with what authors and others wrote in letters, essays and diaries about those texts. Such commentary may also be interesting from the perspective of the sociology of literary texts. However, even in these lines of research, to merely collect author statements and to take them at face value, disregarding the contexts in which such comments were made, would be naïve. And unlike Phelan, for example, who observes narratological distinctions in his analyses, Parker does not treat “narrator” and “author” as discrete theoretical categories when he talks about authors' intentions. This can be observed in Parker's discussion of Pam Houston's self–commentary: “Admitting her place as narrator to her public, while psychically avoiding it herself, she transforms her own experience into something 'fictional,' an ironic disguise, frighteningly close, but othered” (170; italics my emphasis). I think it is important to distinguish between author and narrator since the two need not coincide. Even if authors use auto–biographical material or inscribe [→page 75] themselves into their works, as it were, there is still the creative process which transforms such material into an artifact. The artifact offers an aesthetic experience, and this experience can be very different for different readers. To interpret a literary text does not mean to boil it down to one conclusive meaning, quite on the contrary. Within the overall economy of a novel,⁷⁾ the use of second–person narration can assume a whole range of functions and create numerous effects not anticipated by the author. Is it helpful to say that these functions and effects do not exist merely because the author does not mention them? Can authors' self–commentaries really help us interpret texts, or do they rather limit us in our interpretative vision? It is perhaps not necessary to recapitulate Wimsatt and Beardsley's warning against the so–called “intentional fallacy” here, nor the debate their article engendered.⁸⁾ I do not think that a “writer response theory” could take us very far when thinking about fictional texts because artistic expressivity is likely to encompass more than what may be in the conscious grasp of a writer and is certainly as much related to what readers make of a text as what authors may or may not have intended.⁹⁾