Tweet Score Preparation for Silent Film Orchestra

For those of you who have accompanied silent films with a small ensemble, band or orchestra, how do you approach score preparation for the conductor/leader and/or individual players?

Do you use the original scores or prepare a modern score/parts?

If you prepare a new score, do you include any special marks, notations, rehearsal marks, etc. - such as "optional repeat", "look at conductor", notes referring to the action or title cards from the film itself? All before you begin the rehearsal process.

What's the best way to prepare a score to keep musicians and film together?

Any insights, advice, or things to be avoided would be much appreciated.

Thanks!

Scores that I've done in the past are usually written (or transcribed on computer) as parts, with the cues marked out individually. You have to work out the cuing in rehearsal, of course, but special notes can be hand-written in the margins.

I'm of the opinion that re-orchestration can be beneficial for a number of reasons, but also realize that not everyone is Ravel, so the original arrangements can be useful for short notice.

Rehearsal is the key to synchronization, and professional players will pick up faster than an amateur player, of course.

Do you in essence prepare a full score of the entire film, or use discrete selections and handle transitions during rehearsal as you suggested, beyond just the cuing of instruments based on the score and what may or may not be in the ensemble?

I was trying to decide if I should simply reproduce the period parts and work from them as the would have been in the 1910's 20's, or craft new parts with some cuts and repeats left in or taken out based on my work at the piano from the piano/conductor's score with the film playing?

So that, for example, I include an Agitato just as originally published, or should I include additional notation, optional endings, optional repeats, etc. that would not have been in the original parts.

Would the players scores end up looking like a musical score and parts, which often include optional endings, vamps, and optional repeats to handle on stage delays and variations, or should the score instead be a selection of individual numbers in sequence with little or no cutting or emendation; and all of those bits - cuts, repeats, cues, etc worked out as you said in rehearsal? Which approach provides the most clarity for the players? Anticipate in advance or work out in rehearsal.

I think I'm doing a poor job of articulating some of my concerns so please bare with me.

It makes sense from what you wrote than many thinks will require the ensemble to be together with the film playing to iron out some issues.

Do you in essence prepare a full score of the entire film, or use discrete selections and handle transitions during rehearsal?

That is up to you. I usually write transitions and modulations into the score, usually played by piano or strings. Usually, you'll find that individual parts will fit about a page or two per piece, so it's not difficult to lay out, and if you number the individual cues, it makes it easy for the player to spot where the next cue will take place.

I was trying to decide if I should simply reproduce the period parts or craft new scores with some cuts and repeats left in or taken out based on my work at the piano with the film playing.

So that when a piece returns you have the part again, or do you shorten pieces as needed.

Again, up to you. I red-line a lot of stuff to fit cues, and many of the pieces written during that era are arranged exactly for that.

Would the players scores end up looking like a musical or opera score and parts, which often include optional endings, vamps, and optional repeats to handle on stage delays and variations, or should the score instead be a selection of individual numbers in sequence with little or no cutting or emendation? Which approach provides the most clarity for the players?

The former is easier if you haven't timed the cues out (usually when you have little time to put a score together), but the latter is clearer. Remember, if you are the conductor, it's your job to lead the orchestra in time with the film, so it largely depends on you to keep sync with the film. Writing detailed cues in the piano/conductor part will help you greatly in this case.

I re-edited my post above for clarity, but you had already replied, and you really helped clarify my thinking in the reply, and answered what my clumsy original post was after, although, I'd love to hear other opinons and thoughts on the matter.

Currently, I think, since I have time, that I'll select the pieces after studying some of the different approaches that have been taken by others in scoring the film, then play those pieces against the film on piano.

After I've made my final selection, see which pieces would work best if cut, developed, repeated, transposed for variety and over arching structure, which themes need repeated, decide on cues, based on available forces, etc.

Also allow space for piano to handle the improv in a few places, or to allow for better timing, then prepare in effect a full score and parts with annotations that can be expanded upon or modified in rehearsal.

Do you recommend leaving in a little extra music as it were or relying on piano, pauses, and the leader to carry things through?

I'm concerned about writing a score that is too tight to the film and runs the risk of getting out of sink or being a constant battle to keep in sink, or that because a new score as opposed to emulating, for lack of a better word, historic practice; not too mention sapping the fun out of it.

At this stage I'll either be at the piano with another conductor who I'll work with in the score creation phase, and he will take charge of rehearsals or lead from the piano and conduct the rehearsal from there myself.

If I understand your concern, it's that you're going to run a little short on cues. Obviously you can synchronize a piece only so closely, so what I would recommend is to either a) hold the end chord, which judging from recorded scores was the method of choice, b) have a two-to-four bar section that you can repeat until next cue, or c) play to end of cue and vamp on piano until next cue.

I've done fully-composed scores as well as a composed & compiled score. Since these are scores that are meant to be conducted by not me, I have found timing things out meticulously and putting metronome markings in the score, both at the start of the cue and at regular intervals has made synch possible. All the conductors I've worked with, from high school band directors to pro-level symphony orchs are people who can stay at 116 bpm or whatever, and I have never had a (major) problem with synch in the last years. Giving the conductor permission to slow down or speed up if they get ahead is also useful, otherwise

When coming up with the cue sheet for my composed/compiled score last year, which was 75% "photoplay" cues and the rest by me, I devised a mathematical formula - housed in an excel spreadsheet - to work with so that if I knew the duration of a cue, I could determine how many bars I needed and at what tempo in a particular meter. Then I'd find the cue that fit and did the cut & paste. In some cases I would remove 1st endings or add repeats, or cut bars (usually a mutliple of 4) to fit.

Ben

Thanks Gentlemen!

That is exactly the sort of insight and practical advice I was hoping to find.

I also talked to Dr. Kenneth Sarch at Mansfield University who has written two and is working on a third or fourth silent film score, and he admitted his first problem was issues of synch, and he took a route similar to what you suggested Ben, (these are original scores for string orchestra written with College, High School, and Beginners in mind - a fantastic idea by the way to bridge film, silent film, and music education), that his first scores had the problem of being a little too perfect, and the result was having to hold chords etc., remember these were played by students, and that he was going to allow a little space in future works to cover any variations in tempo.

He used an onscreen timer that was mated to metronome marks to create his score's cues.

I plan on using as much original photo-play and period music as possible, with perhaps some sections being developments of earlier themes when appropriate - in a period style, with improv and piano solos to bridge some sync issues and moments when you really want something to fit precisely, or to have a very wide rubato, also provide some tonal and physical breaks for the other players; but will probably keep all of those to a minimum, or at least a minimum that's aesthetically pleasing.

Glad to know about the usefulness of the metronome markings, at least the conductor if we go that route and I are good friends of old, and we will have time to work on those issues of tempo together.

The cuing via excel sounds excellent, isn't modern film scoring done in a similar fashion so many frames to the bar, or something very like?

Thanks Gentlemen!

That is exactly the sort of insight and practical advice I was hoping to find.

I also talked to Dr. Kenneth Sarch at Mansfield University who has written two and is working on a third or fourth silent film score, and he admitted his first problem was issues of synch, and he took a route similar to what you suggested Ben, (these are original scores for string orchestra written with College, High School, and Beginners in mind - a fantastic idea by the way to bridge film, silent film, and music education), that his first scores had the problem of being a little too perfect, and the result was having to hold chords etc., remember these were played by students, and that he was going to allow a little space in future works to cover any variations in tempo.

He used an onscreen timer that was mated to metronome marks to create his score's cues.

I plan on using as much original photo-play and period music as possible, with perhaps some sections being developments of earlier themes when appropriate - in a period style, with improv and piano solos to bridge some sync issues and moments when you really want something to fit precisely, or to have a very wide rubato, also provide some tonal and physical breaks for the other players; but will probably keep all of those to a minimum, or at least a minimum that's aesthetically pleasing.

Glad to know about the usefulness of the metronome markings, at least the conductor if we go that route and I are good friends of old, and we will have time to work on those issues of tempo together.

The cuing via excel sounds excellent, isn't modern film scoring done in a similar fashion so many frames to the bar, or something very like?

Check your email in a few minutes. And feel free to post my response here.

My scores are photocopies of original parts, marked in pencil. Mont Alto tends to love its rubato, so I can't just count measures -- I actually play along with the film to figure out how much music is needed. Now that I can get the video on my computer and sync it to Digital Performer, I can record MIDI from my piano in real time. That makes it easy to decide how many more measures I need to fill a scene without having to play the entire cue again, or watch to see if it would be more effective (for instance) to get to that minor key section earlier.

Once the quintet is with me, I often find that they don't agree about the speed a particular piece goes at. Since we don't have a conductor, it's always easier to play pieces at the "natural" speed that the orchestra likes (since that's what we're likely to do in performance), so I'll cut or add measures during the first and second run through as needed. Then we change the pencil marks to dark ink for easy reading in live situations.

Occasionally I leave in space for piano improvisation. Sometimes it's to give the other instruments a break -- if we're playing a 90 minute movie and can't have an intermission, I put in a break of up to five minutes somewhere in the middle, long enough to get a drink of water and let the mind relax a bit. Other times it's because there's a short cutaway scene with a different feel, but short enough that it doesn't make sense to bring everyone in. Sometimes it's because a sequence calls for close Mickey Mousing, which can be easier to improvise. And other times it's because I know the next piece is the big storm or battle or whatever, and want to give the musicians time to look ahead and prepare and stretch out their fingers for a six-minute agitato.

My most recent project -- scaling up Amarilly of Clothesline Alley for full orchestra -- will be done on computer from scratch, since the conductor wants a full conductor score (and because it'll be a lot cleaner). We'll see how that goes.

Thanks Rodney, and Andrew for the very helpful email.

It's very interesting to see the variations in approach. I can see the sense of taking what is in essence a more chamber music approach, especially when you play together a lot, I think the key, on how to approach the score, will be the number of musicians and film length. We won't have the luxury of playing together frequently, unless we end up having more opportunities for rehearsals than I anticipate, on the other hand these are professionals, so that will help.

With a small ensemble, a trio to quintet, I think having things a little more open ended, might work better than I'd originally thought, but if we go over five players and especially if we have a conductor, I think preparing the score to be more closely synched with much more specifically scored parts, where things are "just so", might make more sense. The clarity of a conductor and the richness of a larger ensemble will require some more specificity. I wish we could do two different scores for the same film and try both approaches and see what works best - maybe we will for a 10 minute segment - that might have some merit and time should allow for that.

I can't thank you all enough. Everyone's input has been most helpful. I want to say, yes, that's it eureka! I'll do it that way! Thank you Mr. X, you've solved my problems, but I'm beginning to realize, I/we won't know until we try.

I think as I work through the music and decide what to do, my approach to the score will end up taking from all of your suggestions and trying them against the reality; that is the conflict of my abilities and my ambitions.

I think the proof in the pudding will be in the first rehearsals. Having arranged for choir and orchestra, worked on editing an unpublished 18th century opera score, and writing for brass and organ, plus arranging organ solo film scores, this is the most intimidating musical project I've ever undertaken. Which, frankly surprised me, somehow I thought it would be easy - yes, I'm an idiot - I thought you'll pick the scores, copy out the parts, call up some friends and play the picture; now - not so much, although I find it endlessly exciting.

Ultimately, I want this to be good, very good, for any number of reasons: number one, I want the audience to go - wow! that was great - I want to see more silent movies. When are you guys going to do this again. Great models, and suggestions, I hope I make the most of them.

Thanks again,
Eric

I have found that the more information the conductor has, the better—timings, even screen shots in the score if that's possible. When working with pre-existent material, even if it was originally timed to fit certain scenes, there is no guarantee that the film speed will be the same as it originally was (somewhere) and what is 21 fps in Italy may not be 21 fps in Seattle, so timing the cues is extremely important. I tend to favor very precise cues for the images (what film are you talking about?) vs. generic music that matches the mood of a scene overall while passing over individual moments. My new score for Harold Lloyd's NOW OR NEVER is a case in point. It consists of 12 long cues, about 3-4 minutes each, with an extremely precise match of music to image, so that the conductor found it advisable to use a click track to keep the players right on cue. On the other hand, I am scoring UPSTREAM with a series of lead sheets that have some optional repeats and fermatas built in so that if we are slightly ahead or behind the action we can manage. Best of luck! Donald

Dear Donald,
Thank you for the additional information and advice, I love the idea of screen shots in the full/conductor's score, that makes so much sense. Our committee has narrowed down our film choices to two selections and we need to obtain the films and licensing for them next, when that's done I can plan the actual scoring work in earnest. We hope to send out the requests to the copyright holders on Monday.

We've always worked from projected DVD's, as this is an event sponsored by a local church for the entire community, and is provided completely free to all who attend. Therefore, we have very little budget and any thoughts of expanding were nixed when our last movie house in town was closed and demolished around two years ago; so the format isn't a "real" film, but it's the best we can offer. Of course it does provide rehearsal advantages, it's always exactly the same each time, and I can work from the DVD in advance.

Having considered everything and reviewed some of the secondary historical literature, and thought about the films we hope to show, I think I lean toward a hybrid approach: so that the score will use some long cues that may not track perfectly with every small on screen gesture, but will provide the best overall mood for the cues, and support the action without being over powering; and also hopefully it will contain a few cues written and scored to track the action very closely, perhaps at key moments, and developing some of the themes from the more general cues; and then for the sake of tone color and a break for the instrumentalist, also use some piano fill-in and improvisation for other cues. I think the majority of the score will be photo-play or theatrical accompaniment music from the period, I might write one or two themes and/or developmental sections, or adapt a photoplay piece and weave some themes in as counterpoint or develop them and/or re-write and/or re-score these few sections to tie closely to the onscreen action. It would give the score flexibility, and a personal stamp, I hope, without being too idiosyncratic. I've decided I'll do the evening's short with organ only as a contrast in performance practice and for variety.

It seems to be what might work best all things considered. It will allow for some particularly satisfying creative work, should mean that rehearsals can focus on the general cues and the more intense scoring for effect, and allow for the musicians to have a successful outing and not be too taxed.

I have already thought a great deal about some of the scoring in general for the two films we wish to do, but I do not want to embark any further until the actual pictures are settled upon, and then it will be back for more advice. The size of our ensemble is also dependent on a grant we will hopefully be able to receive. But I wanted to start the planning as early as possible - if funding falls through, I'll probably do the score as an exercise and an outlet and perhaps use it some day in the future, or at least use it for the organ solo score I'll create in it's place.

Thanks again for the great advice,
Eric

slightly OT...do any of you with collections of sheet music have "The Philanderer" by Bertram Srawley? I have a recording of it and have been trying to track this one down for some time...

Ben

No, I'm afraid I don't. There are two orchestrations of Srawley's The Philanderer at the University of North Texas (as well as a different composer's piece by the same title), but they're strict about copyrights, so you may have to give a stab at clearing it or getting permission before they'll make you a copy.

http://iii.library.unt.edu:81/search/t? ... mit=Submit

I see that the New York Public Library has several other Character Sketches by Srawley (The Flapper and The Hustler, both of which sound like fun). You may be able to look at those even if they won't copy them for you (I assume that you live closer to the NYPL than to Denton Texas).

I have some links to online catalogs of archives of photoplay music at the bottom of the Mont Alto "Silent Film Music" page, if you want to search those archives in the future. And please let me know if you know of other online catalogs that I can add to the list.

Rodney. thanks for the info. I've been meaning to take a more in-depth look at the NYPL collection. I certainly know enough people who work there now to at least get access...

Ben

The National Library of Australia also has "Philanderers." They're very easy to work with as long as you have your copyright stuff in order.

FYI, Srawley was published through George H. Sanders. Sanders renewed some of his pieces on a per-author basis, but not all of them. Unfortunately, I do not have the renewal book for 1925 to check whether the copyright for Srawley's works (which were mostly published between 1924 and 1926) were renewed.

The National Library of Australia also has "Philanderers." They're very easy to work with as long as you have your copyright stuff in order.

I've worked extensively with them before, they've got quite a collection of stuff, and even if they're unsure of copyright status, they'll point you in the right direction (if you do your research) and can give you email addresses for the current copyright owners in Australia. I had to do that for a number of pieces, but was able to work out everything with them within a few days.