Director Patrick Kelly Looks Back At The Ghost Sonata


Last Saturday, Undermain’s production of The Ghost Sonata concluded its run.  Stephen Foglia sat down with the director, Patrick Kelly, to do a post-mortem on his most recent success.

Stephen Foglia: We just closed on Saturday, and you watched [the show].  How was it?

Patrick Kelly: It was terrific.  I hadn’t seen it for a bit more than a week.  The performance strengthened as the repetitions went on.  The cast was – I would have said remarkably consistent in their performances, but at Undermain that has been the rule rather than the exception.  The Birthday Party was the same way.  The actors, once they achieved the performances they were after, maintained them very well.  It only got better.

Jumping way way back to the beginning of the process, you’ve directed Ghost Sonata once before.

You may need to jump back all the way to 1981, when I directed the play at the University of Dallas.  And that was also my first collaboration as a director with John Arnone [Scenic Designer of The Ghost Sonata].


He was establishing himself in New York and in the regional theatre at that time as a scenic designer.  I called John just to see if he’d be interested in doing a job here.  He said, “sure,” and we then did the play.

So I had a very happy memory of the play, then when John and I were talking about a project to pitch to Undermain after The Birthday Party, Ghost Sonata was one of about five that we looked seriously at, and then it was the one that we pitched to Katherine [Owens, Artistic Director of Undermain], and that she picked.

Does this play require something of you as a director that other plays don’t typically?

Yes.  I think that Strindberg in this play is harder on actors than any writer I have ever worked with.  Any writer.  I think that Strindberg makes demands on actors, not necessarily on directors, but upon actors that they are unfamiliar with.

An image does it better: when you’re doing a part in Shakespeare, or Ibsen, or Arthur Miller, or Molière, or even Aeschylus, the writer gives you a necklace or a bracelet of beautiful stones, and you put this on you and animate it that way.  What Strindberg gives you is a handful of gems and says, ”you string them together.”  And what you string them together with is your own sinew.  You make the part.  And that’s a very heavy lifting.  In a sense it’s creative rather than interpretive work.  It calls upon [the actors’] imaginations and their sensibilities in a way that is more profound than almost any other writer that I’ve come across.  He leaves it up to them.  And the only one I could compare him to would be Beckett.

Do you talk to the actors about that?

No.  No I don’t talk to the actors about any critical perspective at all.  Because their work is so personal.  It makes sense to talk to a dramaturge, or a member of the audience, or to a part of the design team.  But the actors are dealing with the work so directly, so profoundly – it’s them.  I mean, a director looks at something and is always outside of it.  Designers, same thing.  But a performance is not out there.  A performance is in the performer himself.  It’s a matter of his own life or her own life that is being used.

Do you even mention this idea of difficulty at all to the actors?  “Ok, we’re in something that’s going to require more—“

— No, because I didn’t know it until after the play was on.


No!  It’s entirely something new that I didn’t know until we were open.  I just knew it was very very very hard.

We have had really wonderful responses from critics and audiences, and I was wondering if there were any that stuck with you.

The greatest compliment that Strindberg received was that of almost everybody who saw the play, the most common comment, somewhere in there was, “I can’t believe it was written in 1907.”  What better compliment can you make to an old play like that?  You wouldn’t say that about The Cherry Orchard, wonderful play that it is, it doesn’t have that strange sense of challenging your imagination and surprising you.

None of the critics expressed bafflement as to the meaning of the play.  If anything that’s the most profound praise that a production of the play could have.  I think it’s very strong praise for the performers and the designers.  And indirectly for the director.

May I infer from that that clarity was something you concerned about going in?

I’m always concerned about clarity.

How do you harness the narrative in Ghost Sonata?

Just making the events that the audience is witness to clear was the main thing.  The background in the play – the guilty past and all that – is very hard to keep track of as an audience member.   The narrative that we had to be very careful to show was contained in the relation and the changing of the relations between the characters that the audience sees.

This had to do also with the floor plan that Arnone figured out.  Putting the house behind the actors who are looking at it.  Having the actors look out at the audience and light up the people they were talking about, up-stage of them.  We’re not the first people to do that.  I think Bergman did it a couple of times in that way.  But I think the space down here accommodated that idea very well.

It was a struggle.  That last scene in particular was very much a struggle.  And remained so to the last nail-biting moment.

I’d like to hear more about how the design came about with John and what questions it was meant to address in the play.

One of the things about the play that I think both of us felt or appreciated was its somewhat fragmentary approach to painting the picture of the lives of these people.  There are shards of story that are sprinkled throughout the play.  And you have to put them together yourself.

What I compared it to, and I’m not sure John and I discussed it this way beforehand, but once we had the set I could see it, was that it’s like cubism.  Analytical cubism was arising at just this moment.  The Desmoiselles D’Avignon by Picasso dates from nearly the same time [as Ghost Sonata].  What the painting does, and what the whole project of analytical cubism tended to do was to look at a subject from different angles and to somehow arrange those different perspectives in the same picture.  So you get that kind of fragmentary and broken up, not quite put back together, look.  That I think would explain the underpinning for [Arnone’s] set.

You also see in the Picasso, I guess, that perspectival fragmentation can be not just literal geometric perspective, but you could see at the same time the idea of the woman as physically rendered, physically present at the same time and right next to the woman as primitive mask.  And I see that in Ghost Sonata as well.  You can have the idea that somebody could be a vampire and an old man.

Or a character can be in a winding sheet and walk out the door, and it doesn’t turn the play into this sort of ghost story.  The play seems to exist sort of outside of genres.

This is a part of modernism, at least as I understand it in music, in literature, in poetry, that there was interest in how you render reality but also this idea of how do you categorize a rendition of reality.  So that the old traditional genres tended to disappear.

You did a significant amount of pre-blocking on this.  How early were you attacking that?

I plan the plan the blocking of every play that I do.  Only because I don’t like the experience of having actors turn and say, “well, what now?”  Once I have a basic movement and the actors on their feet, I can work with their impulses and we change things.

But it seems to me that the staging of this play is rather set out for you.  The first scene is about a character in a wheelchair and a guy standing there talking to him.  So that immediately puts limits on where you can go.  Second scene is about basically five people sitting down to have tea together.  So when you put a table on the stage it fills the stage and everything happens around that.

So the pre-blocking I did was not intricate or difficult.  In fact, in the case of the third scene it was thrown completely out, and the scene was re-staged many times.  By the end we found the staging that was required but I had not anticipated.

You cast a young man as the Old Man.  I think audiences would be interested to know how that decision was made.

Very easily.  Blake could play the part.  His age was not a main an issue.  He had very good make-up advice, but he had the imaginative and artistic scope required.

Casting was essential to the play, and we spent eight months on it.  Which is a long time, but there are twelve actors in the play.  We had to get the right dozen people.  Thanks to the patience and the resources and, if I may say, the reputation of Undermain Theatre – people are anxious to come here to work – we were able to get the cast that I thought was perfectly suited for the play.

Dylan Key [Undermain’s Associate Producer] worked as a casting director, really, through the very long period we spent interviewing actors…he certainly has a great amount of the success of the play to his credit.

What was the most challenging part of the process?

Getting the three scenes to feel right together.  To establish a tone or to establish the progression of tone that the play does in a believable way.

The actions of the play are very different.  Each very odd in themselves, but they’re not very much like each other.  The first scene is almost completely narrative.  People talk about things that happen in the past.  The second is a kind of operatic interlude in which a number of characters express themselves at some length.  Then in the third scene you have a duet, interspersed with a few small comments from the Cook.  Which seems to have really nothing to do with what went before.  To get all those things working together and making sense to an audience was very challenging.

In practical terms, how did you effect that.

Trial and error.

How early did you start running the scenes consecutively rather than as separate units?

Early.  Early.  We had some conflicts – certain of the lead actors had to leave town or not be at rehearsal until the middle, but I thought it was important to face the challenge of doing the play all the way through from early on.

If you had time to re-examine any part of the play, is there a part of the play that you’ve had new thoughts about?

I did think it might have been a good idea to have the ghost of the Milkmaid appear more frequently while Hummel was on-stage.  As I was watching the show – she was so good, the actress who played it was so striking and so remarkable – I began imagining, “oh, if the ghost was here!  If she was just standing silently, that might be very interesting…”

Were there particular passages that exceeded your imagination?

When you have an audience and performers.  It’s always greater than you imagine.  I thought it was wonderful the way they almost always laughed when the Young Lady says, “there’s a maid I have to clean up after.”  And I thought that was great because the Young Lady would be dying in a matter of minutes.  And yet the laughter didn’t do anything to take away from that.  Strindberg was realizing that you have to relieve a certain kind of tension that you can only do with a laugh in the theatre.  The more serious she was about it, the funnier it was.

Well thank you very much, Patrick, for talking to me.

Thank you.