top of page

"Every time I answer a question I glance out the window to check the snow, and I see that there’s exactly the same amount, always. I believe that special employees of the town are hired to work through the night, spreading snow so that it remains at the same height  for six months.”


Steven Millhauser is an American novelist and short story writer. He won the 1997 Pulitzer Prize for Fiction for his novel Martin Dressler. The prize brought many of his older books back into print.

(The following interview was conducted over the Internet from 9th October 2009 to 6th June 2010 by Andrzej Gabinski and was published in Short Fiction 6.)

I would like to begin this interview by commenting briefly on the predominance of shorter literary forms in your recent writing: in an interview with Jim Shepard published in 2003 you praised some of the characteristics of the novella; in your article from 2008 you unveiled, not  without  a certain dose of  enthusiasm, the ambitions behind the short story; also, in that same year you published a collection of  short  stories  completed  in  the  previous  ten  years. Today, are you still equally attracted to shorter literary forms and the unexplored territories they might lead to or, on the contrary, have you considered re-embracing more extensive formats?

I continue to be drawn to short forms, for many reasons. I like concentrated effects, of the kind invited by short forms; I like intensity, sharp focus, heightened attention; I like the way something small can expand into something large. None of this should suggest hostility to the novel. What I dislike is the assumption that stories or novellas are by nature slight or unimportant, because of their shortness. I resist the idea that a writer is supposed  to begin with stories  and then work his way up to the real thing — a novel. It’s like saying that paintings become more and more important as they grow in size. The grand, the monumental, the all-encompassing, have an attraction of their own, and I’m not immune to the lure of largeness. But I’m also deeply suspicious of it, especially in American culture, where largeness is always considered a virtue.

Indeed, a number of your literary works disclose the apparently paradoxical pull towards, on the one hand, the minute and the intense, and, on the other hand, the vast and all-inclusive. And while it is true that American culture in general does hold novels in greater esteem, shorter  forms  do attract the attention of progressively  wider circles of readers and critics, simultaneously evolving towards ever-greater brevity and condensation, resulting in flash  fiction, vignettes and, ultimately, twitter entries. Do you believe these micro-genres based on word-count premises might represent a territory worthy of literary exploration?

I’m in fact very much attracted by the paradox you mention. “The Dome” is one example that springs to mind – a short piece that imagines increasing forms of vastness. The extremes of immensity and smallness are present throughout the history of American culture and continually offer themselves as models. Think of the sheer contrast in size between “Song of Myself ” by Whitman and any poem by Emily Dickinson, or Moby Dick on the one hand and Poe’s celebration of shortness on the other. Or, in the realm of nineteenth-century objects: the skyscraper and, say, the light bulb. In our day: the superhighway and the microchip. Why this fascination with extremes? I think it’s because America experiences itself as a vast  place that can only be expressed in gigantic forms, while at the same time there’s a distrust of vastness that expresses itself in a craving for the opposite.

As for radical brevity in works of fiction, I probably ought to be drawn to it, but for some reason I’m not. I need a certain amount of room. Poetic short forms are a different matter – I love the form of the sonnet, the extraordinary number of variations possible within a single tight design. Sometimes I’m jealous of poetry’s strictly ruled forms and like to imagine a prose equivalent: the ten-sentence story, the hundred-word story. But finally my interest in such things doesn’t run deep.

Another singular feature of some of your shorter pieces is their format: from cartoon-like vignettes “Cat’n’Mouse”, through lecture- structured works, “Kaspar Hauser Speaks” and “Here at the Historical Society”, to a story told through imagined paintings, “Catalogue of the Exhibition”, to, eventually, an embedded fictional object containing a separate narration, “Klassik Komix #1”, not to mention the formal complexities and delights of “A Game of Clue”, and others. Does the idea of such uncommon formats precede the actual completion of a story or, rather, does  their appearance run parallel to the process  of literary composition?

Before I give myself permission to write, I spend a great deal of time turning things over in my mind. When, after many weeks, I finally do sit down to write a story, I’m fairly clear about many things, though not of course everything. One thing I’m certainly clear about is the kind of format you’ve mentioned. It’s exactly why I spend time turning things over in advance, since otherwise the form would be uncertain and haphazard and therefore boring. By the time I sit down to write, I know that “Kaspar Hauser Speaks” is a first-person story narrated to a particular audience on a particular occasion, I know that “Catalogue of an Exhibition” will proceed as a series of descriptions of invented paintings.

As for what happens in my mind before I begin writing, that’s much more difficult to say. In the case of “A Game of Clue,” I began by wanting to write a story about a family that played a game of Clue. But that’s a very vague, empty idea. Only as I began to think about it, to ask myself what excited me about it, did I come to understand that it was the Clue board itself that was going to be at the center of the story. The characters would be given the same kind of attention as the human characters. And only after that did the idea of the structure begin to emerge. Each story has some center of excitement that I try to locate and understand, and from there the formal elements begin to present themselves. Only then does writing begin to be possible. The process of getting there is murky, and I’m happy to leave it that way.

In the story you mention the board binds as  many as  three distinct narrative layers, a strategy that also surfaces in “Alice, Falling” or “The Eighth Voyage of Sinbad”, which makes me wonder whether you sometimes attribute the resulting multiple sub-narrations a certain level of autonomy in your considerations or, rather, do you always feel they belong to a greater whole?

A tricky question. When a story presents itself to me as a series of layered narratives, I naturally want them to be part of a whole – if they’re not part of a whole, what are they part of ? Incoherence lies that way. But it’s equally true that narratives of this kind invite certain kinds of freedom that distinguish them from more conventional arrangements. It may be that I’m attracted to narrative layers precisely because they  fight against  my tendency  to  drive a narrative to  a predestined end. After all, what is a story? What is any work of art? It’s the result of a bloody battle between two forces – the freewheeling imagination and the constraints of form. Multiple narratives are a way of disrupting a strict sense of form – which fights back in its own way. As I said, a tricky question.

It appears, nonetheless, that form always gets the upper hand, in the end – especially  in the case of  closures,  which irrespective of  their interpretative straightforwardness or haziness  are logically unavoidable. Have  you  ever considered  ambiguous  closures,  such as in the case of “The Dome” or “The Dream of the Consortium”, for instance,  as the ultimate attempt to somehow  counter the rigid confines of literary format?

I think of myself as attempting two opposite tasks simultaneously: driving a narrative or a vision toward its necessary conclusion (the steady enlarging of the dome until it encloses the universe, the increasingly dangerous throws of the knife thrower until the unthinkable takes place before our eyes) and at the same time entering territory, at the conclusion, that opens up the story even as the story completes or exhausts itself in its final movement. I’d argue that there are two kinds of closed stories: the kind that shut down at the end, leaving you in a small closed room, and the kind that open up even as the reader hears the sound of a closing door – but the reader is now on the other side of that door, stepping into a new, open place.

That  last  comment  made me think of “Paradise  Park”  – although in the end fire mercilessly devours Sarabee’s fantasy, I cannot stop thinking it was there, at least momentarily. Actually, I often wish that particular story had left me in the open place you mention, yet I cannot imagine what exactly it would look like. By the way, does the last image of the male swimmers bear any relationship to Section 11 of Whitman’s “Song of Myself ”?

But it IS there, from start to finish, a complete world that can be entered or ignored as you wish. The destruction of the park, within the fiction, is no more than the act that completes the fiction. The open place that I imagine is independent of particular acts within the fiction – it’s the place you’re carried to when the story is over. But that’s not for me to describe. The final image of the male swimmers comes from a photograph of Coney Island at the turn of the century and has no connection with Whitman.

Bearing in  mind  your  last  collection  of  stories  I  cannot stop asking myself whether unambiguous endings, as far as their interpretative dimension goes, might be useful to a writer like yourself in any way?

I would never argue that an ambiguous ending is necessarily superior to an unambiguous ending. The  ending is  in the  service of something  else,  which I might call the vision  of the story,  and is used  to complete or strengthen  that vision.  In this sense, yes, an unambiguous ending might in certain instances be more useful to me than an ambiguous one. Consider “The Disappearance of Elaine Coleman.”  The story takes the form of a mystery:  How did she disappear from a locked room? The narrator is a kind of detective. He gradually arrives at a solution to the mystery, which he presents to the reader. The ending, or revelation, is explicit, though it’s complicated by being (strictly speaking) impossible. But it’s important for the final explanation to be clear, because the solution to the mystery is also the narrator’s discovery of his complicity in the disappearance. An ambiguous ending would not permit that effect.

A story like “History of a Disturbance” works differently. Here, the ending is what I would call ambiguous. By this I mean that the narrator explains himself very clearly, but his words at the end, as he invites his wife to join him in a new world of revelatory silence, are so extreme that the reader questions them and is invited to dissent from them – to ask, in short, whether the narrator is insane. That kind of ambiguity is crucial to this particular story, in which the narrator’s mental state is part of what the story is about. In “Elaine Coleman,” the explanation is extreme, but the story doesn’t ask you to question the mental state of the narrator. It asks you to share the narrator’s sudden flood of self-understanding.

Individual self-understanding in the midst of a larger community to which the narrator belongs and on whose behalf they often,  but not always, speak…The theme was brought up in the 2003 interview with Marc Chénetier,  in which you mentioned  your deep interest and attraction for the “we” pronoun,  especially when it operates in conjunction with the singular “I” – a strategy employed precisely in “Elaine Coleman” or, for instance,  “Dangerous  Laughter”. Do you still feel, like you explained you did in 2003, that you are “certainly not done with” the plural narrative voice?

I’m far from done with “we.” It has many advantages. One is its ability to swerve into “I,” as you’ve mentioned. Another is a splendid refusal to participate in the kind of conventional story in which one character is favored over others, a character with whom the reader is invited to sympathize. In a “we” story, a community is presented as a character in its own right. It’s a strange sort of character, revealed and hidden at the same time. It’s a character without a face. And a communal narrator invites a type of narrative that interests me: the story of a battle between  a community  and a dangerous  outsider, or between a community and something  disturbing  that takes  place within it.

In addition, “we” always hovers on the edge of paradox. It’s more intimate to say “we” than “he”, since the pronoun includes the speaker directly; at the same time, “we” is deeply impersonal, since it forbids the reader the  familiar  pleasure  of  connecting  with  an  imagined character. Such paradoxes strike me as rich and worthwhile. I intend to explore them in future stories.

Particular communities in your stories enjoy singular habits/ traditions/circumstances disclosed by a plural narrator, which, bearing in mind its paradoxical impersonal  intimacy, might actually suggest tales of cautionary character. Are some of your narrators confessing, as opposed to simply relating, their communal experience?

Yes, definitely. This is especially the case in “The Knife Thrower,” in which the plural narrator doesn’t simply report a disturbing event, but continually reveals moral discomfort at witnessing that event. The watchers know perfectly well that they’ve come to the auditorium in the hope of seeing blood, just as they know that they’ve carefully evaded confronting their secret desire. They know they’re complicit in whatever happens, even as they deplore what happens. The story is presented as a report, but the tone repeatedly  suggests a troubled confession.

The confessional tone is equally strong in “The Disappearance of Elaine Coleman,” though it doesn’t reveal itself fully until the end. “We are no longer innocent,”  the plural narrator says. The explicit confession comes from the singular narrator — “I too murdered Elaine Coleman” — but his revelation  of guilt attaches inevitably to the “we.” The story is a kind of mystery tale, in which the narrator becomes a detective who solves a crime. The solution leads to a confession: we are all guilty.

Such tales of collective responsibility are clearly opposed to less disturbing accounts of shared enthusiasm, as for instance in “Here at the Historical Society”. Nonetheless, the covert sensations of unease or guilt appear to surface in your stories more often – do you believe a narrator’s experience of and participation in the uncommon necessarily marks their accounts with self-reproach, rather than, for instance, sheer amazement or wonder?

Even  if  the  narrator  is  plural, a  disturbing  or  uncommon experience will leave its mark on the narrative voice. The author – as  opposed to the narrator —  can handle this in various ways. The author may give the plural narrator a cold, objective voice, which will allow disturbing events to emerge as  if  they were ordinary; a tension  will inevitably be created between the voice and the things the voice is  reporting, and the voice may show occasional signs of agitation that suggest a kind of failed repression. Or the author may allow the  disturbed  material to  erupt  powerfully  into  the  voice, which  may reveal a host  of  feelings:  self-reproach,  anxiety, harsh disapproval,  defensiveness,  nervousness.  Because  “we”  suggests  the idea of thoughtful summary, of a merging of individual identities, it is associated with the impersonal, and therefore the slightest ruffling of that surface will be immediately noticeable.

Nonetheless, these rufflings, however slight, tend to disclose negative sensations and states of mind – embarrassment, disturbance or tension. Contrarily, do you think a hypothetical coming to terms with and openly embracing the extraordinary would render a narration as uninteresting?

Trouble is always inherently more interesting than its opposite, but lack of interest comes only from bad art. I can easily imagine an embrace of the extraordinary that would be seductive, enchanting, compelling. But the question makes it sound as if a story presents a single state of mind. I tend to think of stories as battles.

Earlier you observed that the plural pronoun “suggests the idea of thoughtful summary” – do you ever imagine your communities considerately selecting an individual as their “voice” or, rather, conceive a persona that spontaneously assumes the task of relating a worrying shared experience?

I’ve actually used the second strategy you mention. In a story called “A Report on Our Recent Troubles,” which was published after the stories in Dangerous Laughter, I chose a plural narrator composed of a number of concerned citizens who present a report to a Town Committee  about  a  series  of  troubling  incidents.  The  reader  is invited to imagine that, as the town experiences  a wave of suicides, a Committee is formed to investigate  the cause;  several members  of the Committee  are chosen  to conduct the investigation  and report back. The story  takes  the form of that report. It’s exactly what you suggest: a persona (a multiple persona) that assumes the task of relating a disturbing experience.

Narratology theoreticians  generally consider  a third-person impersonal narration as more credible, principally due to its supposed objectivity and neutrality, when compared to an I-narration; nevertheless, the “we” pronoun is barely mentioned in similar analyses. Do you ever consider plural narrators in terms of their credibility/ storytelling authority – especially  if juxtaposed with a singular first- person speaking voice?

I consider  all narrators  in terms  of their credibility. There’s  a particular kind of authority inherent in each pronoun.  An “I,”  for example, has the authority of witness,  whereas  a “he” or “she”  has the authority of the impersonal, of objectivity. “We,” like “I,” has the authority or credibility that comes with active engagement in events; at the same time, it shares the burden of the subjective that always haunts personal narration.

But  the  subjective  isn’t  the  same  for  “we”  and  “I.”  A reader understands  immediately that a single  narrator might have feelings and opinions that derive from temperament and might therefore be untrustworthy. But what is  the  temperament of  a plural narrator? What, in fact, is a plural narrator? In real life, there are situations that regularly call for a “we” – for instance, many people, with extremely different temperaments, might agree to a number of statements in a political document that they’re  all willing to sign.  In the same  way, one can imagine a very limited sort of plural narrator, in which a small number of separate personalities might agree with all the statements that constitute a story. But as the number of narrators included by “we” increases, it becomes difficult to believe that every statement is shared in the same way by every member of the group. Many of my plural narrators represent an entire town. What does it mean to say that “we” were made anxious by a certain event? All twelve thousand of us? As soon as you begin to think of it in that way, the authority of the plural narrator becomes  undermined, and it’s  precisely  this  undermining, this continual questioning of statements made by a plural narrator, that I find intensely interesting.

When “we” is juxtaposed with “I” in a single story, the first-person narrator assumes the authority of personal witness, and reminds us that statements made by the plural narrator are necessarily more general.

The first-person narrator has the authority of the case history, whereas the plural narrator has the wider authority that comes from enlarging particular instances  into a broad truth.  This  distinction  – between two kinds of authority –takes place even as the larger question about trustworthiness is being asked.

Your stories often feature grammatical structures that emphasize their hypothetical value. Do such modal devices constitute a strategy aimed at better accommodating plural narrators – who thus provide multiple interpretative possibilities of events, rather than an “artificial” and only truth?

Devices of the kind you mention might be used for any kind of narrator, but when they’re  used  for plural narrators  they do have the effect you suggest. A “we” narrative is already charged with the kinds  of uncertainty I’ve spoken  about; grammatical structures  that emphasize the hypothetical are very useful for undermining the notion of a single, easily grasped truth.

The  narrative voices  present  in  your  stories  often  provide readers with a variety of sometimes opposing interpretations of the events described. It thus appears that readers are eventually faced with a rather difficult choice regarding the ultimate construal they so long for. Should they abandon that longing?

In certain stories I want readers to be faced with that choice. Or it might be more accurate to say that I don‘t want readers to be faced with that choice, as  though  they can feel peaceful  only after  they choose among many contradictory interpretations – rather, I want readers to experience, to be aware of, those many interpretations. This has nothing to do with any sort of pleasure in confusion. As for the longing to find an ultimate meaning – you might as well demand the same thing of life.

Honestly, I much more prefer the plenitude of uncertainty. I would like to conclude the interview by asking you about your recent visit to France and how an author like yourself may defend the short story in the homeland of great novel-writers?

A defense of the short story in the country of Maupassant? Hardly necessary! The University of Angers hosts an annual conference on the short story in English, and they also publish a journal called Journal of the Short Story in English. My job was to give a short reading (I read selections from “Klassik Komix #1,” which is composed entirely of images) and answer questions put to me by a panel. The questions were of two kinds: general questions about the genre of the short story (example: “Do  you find it more difficult to write the beginning or the end of a short story?”) and specific questions about stories from Dangerous  Laughter (example: “’Cat ‘n’  Mouse’ evokes the animated cartoon.  It  is  written  in  short  narrative statements  in  the  present tense  in a very stylized  way. What narrative or thematic powers  do you  perceive in  the  images  of  an animated cartoon?”).  Although all questions  about stories  strike  me as  impossible,  the questions  in this case were sent to me in advance, so that I could brood over the precise nature of the impossibility of answering them. They tell me the interview will appear next year in their journal. Everyone, incidentally, spoke English fluently, and everyone was extremely gracious.


I sincerely hope to be able to, one day, attend a reading by yourself or a panel with your participations. I would like to thank you for your time and answers, which will now allow me to explore in depth your impossible answers.

bottom of page