Mahler’s Fifth Symphony Considered as an Uphill Bicycle Race

“I want to see who can make a program for my Fifth!”

Gustav Mahler to Richard Batka, 1905


Procrastinating real work, I decided to write a short creative project, which does in fact come out of my most recent research. I discovered a year ago at AMS that Gustav Mahler was not only a cyclist, but also had a penchant for ill-advised excursions up mountains. On one occassion, he apparently tried to ascend the Loiblpass just south of his summer home in Maiernigg. According to the story recounted by Natalie Bauer-Lechner, he thought he was about to reach the summit when he realized he still had “1000 meters” left to climb. At this point he paid a boy to push his bike up the road while he took a short cut, and almost killed himself trying to claw his way up a drainage gully. I’m currently researching the type of bike he would have used, following up on the story, and putting together his probable route. The next step will be to figure out how and whether this has any impact on his music, but the 5th Symphony, written immediately after the incident just mentioned, seems like fertile ground. I chose a title for such a study: “Mahler’s Fifth Symphony Considered as an Uphill Bicycle Race,” a reference to The Passion Considered as an Uphill Bicycle Race” by Alfred Jarry (1905). J.G. Ballard also wrote a parody of that work, “The Assassination of John Fitzgerald Kennedy Considered as a Downhill Motor Race.”

And so what follows is my own parody, which will nonetheless reveal certain ways that I think a cycling-related program can be imposed on that symphony. The recordings referenced are: Hermann Scherchen, Philadelphia Orchestra (1964); Bruno Walter, New York Philharmonic (1947), Bernstein, New York Philharmonic (1963); Hans Rosbaud, Cologne Radio Ochestra (1951); and Bruno Maderna, RAI Milano (1973).

Klemperer, slated to race, was scratched.

The announcer, having listed the participants, gave the cue to begin. Scherchen, already on a bad footing, missed his first pedal stroke and almost ended his race at the start. He recovered with a strong second stroke and began his pursuit.

The race began with a “strictly measured” neutral start, but this did not stop the opportunistic competitors from trying to gain an advantage.

Walter took an early lead.

Italian mechanics being notoriously deficient, failed to tune Maderna’s machine prior to the race.

In those days, according to the excellent sports commentator Kalbeck, it was customary for the racers to begin by paying homage to colleagues who had fallen off on the previous day’s racing. For this race, and because of particular carnage in the previous week, organizers took the unorthodox step of beginning the race with a “funeral march.”

The course for the day had a somewhat unusual profile; the climb consisted of a lower peak followed by a quick descent and a false flat, followed by a long climb up to the higher elevation that constituted the day’s finish. In this, the climb was not dissimilar to the popular combination of the Col du Telegraph and the Col du Galibier used frequently in the Tour de France. The course had been designed in 1904 by the famous race organizer, Gustav Mahler.

Indeed, though this exact course has been used many times subsequently, there is always some confusion as to whether the course should be divided into three or five sections, as sectors one and two, and four and five seem to flow naturally into each other. There were however, five clearly marked “time-checks” including the summit finish.

The competitors were more or less riding in a group formation as they reached the first turn. Here things got suddenly faster. One might even say the races began riding wildly, or passionately.

And here, already, came the first attack of the day. Scherchen leapt out of his saddle and began riding so quickly that his wheels had trouble keeping in contact with the ground, and many spectators had the feeling he would simply veer off the road altogether. Rosbaud made a quick chase followed by Maderna and Bernstein. Walter kept his cool and, knowing that others would crack, and gradually brought the breakaway back, and subsequently rode them off his wheel.

Maderna, meanwhile, was somewhat distracted by the group of Mariachi trumpeters that greeted his arrival at the same turn.

At the first time-check, Walter crossed with a substantial lead, followed closely by Rosbaud who recovered well from a slow start, then Bernstein. Maderna just edged out Scherchen to round out the top five.

The section following, which led up to the first major peak of the day, also saw a storm descend on the race. The racers responded with the greatest of vehemence.

The less-experienced riders were surprised upon reaching the second time-check that this was not, in fact, the finish to the day’s racing; many who rode too hard during this section found themselves dispirited when when the road turned suddenly downward and they realized they still had over 1000 meters of climbing in front of them.

The standings changed little between the first and second time checks. Walter, continuing his blistering pace added over a minute to his lead. meanwhile a flat tire all but ended Bernstein’s day, and he crossed the line forty seconds behind Scherchen.

The course guide for the day warned riders not to dismiss the section from time check two to three, but many nevertheless considered it a “joke,” or “scherzo” as the Italian riders were fond of saying.

Walter’s seemingly unassailable dominance persuaded a dispirited Scherchen to dismount his bicycle. He handed the machine to a spectator and took an ill-advised shortcut by foot, missing over half of the race course in this section. The other competitors immediately lodged a formal complaint.

The spectator, meanwhile caught a lift from a race official and met Scherchen near the third time check with bike in hand. Scherchen emerged from the side of the road, having almost fallen to his death scrambling up a smooth water drainage gully. Some have speculated that Scherchen’s motivations were not as nefarious as they may have seemed, and that he fancied himself more of a race organizer than a bicycle racer.

At the third time check, Scherchen now had a commanding, though contested, lead. Having lost almost ten minutes, Walter now trailed Scherchen by six and a half minutes. Walter, who had been well in front when Scherchen made his short-cut, only found out about the deception as he crossed the line. His pride, however, prevented him from relying on official sanction, and resolved to win the race in spite of the loss of time. He set off in pursuit of Scherchen, the others now well behind.

The next section, by far the most beautiful aesthetically, nevertheless contained potential pitfalls for the competitors. Its terrain is a long “false-flat” where a fit rider can push a large gear, but a tired rider can easily come to grief. A famous sports commentator, and partisan of the so-called Frankfurt school of bicycle racing, found this entire section too “pretty,” and was heard muttering something about its “ingratiating sound.”

And here, Walter’s persistence paid off: Scherchen, clearly winded from his cross-country excursion took nearly twice as long to arrive at the fourth time check. Indeed, it is said that no other professional cyclist has ever taken longer to travel that section of road, even Scherchen on his previous attempt at the same course with a rickety state-issued Viennese bicycle.

At the fourth and final time check before the finishing climb, Walter was now again in the lead. A very tired-looking Scherchen crossed the line a minute later, followed by Rosbaud four minutes after that. Bernstein though not quick, managed to pass a faltering Maderna just before the line.

Walter set off in search of glory, his victory now all but assured. While he was not the fastest on the day through this section, Walter climbed well. Bernstein, surprisingly, was about 10 seconds faster than Walter, and over a minute faster than Rosbaud. Rosbaud, however, still had enough of a buffer that his podium place was never in danger. Maderna, no longer caring about overall victory, stopped to admire the stunning vistas and finished last of the favorites.

Walter, near the summit, slowed to wave at his ecstatic fans. Meanwhile, Scherchen made an amazing recovery–amazing enough that he has always been suspected of pharmaceutical enhancement. As Walter went under the one-kilometer-to-go banner, a sprinting Scherchen flew past on his way to victory. Walter slammed his fists on his handlebars in disgust as he crossed the line.

The results are contested.


Just when you think you know all the hills in Rhode Island. I put together a nice new route today, heading out over Capron, Mountaindale, Pine Hill, before heading towards the Scituate Reservoir via Gleaner Chapel, another hill I discovered relatively recently. Feeling good today, I set personal records on a bunch of the hills, picking up a Strava KOM on Gleaner Chapel. (I missed the KOM on the hotly contested Capron by 1 second…) From Gleaner Chapel, I wound my way back via Peck Hill Road, which was new to me. Not the most scenic road, at least not by traditional standards. You do get many exciting views of the Johnston Landfill – the whole stretch seems somewhat post-apocalyptic.

From there, the last event of the day was Ipswitch St., an intriguingly steep little climb I discovered on Strava. I knew it was steep going in, but was still rather unprepared for what I found. Here is what I expected:

ipswitchWhich is to say a tough, but sustained 12% climb, with a steeper part in the middle. The climb starts on Monson St, which takes a tight left turn. You can tell that there’s a steep hill in front, but you can’t see Ipswitch until you make right turn and see a very steep, very narrow road in front of you. But “sustained” isn’t really the right word. The next section is comprised of two distinct sections of (probably) 20%. You think when you get to the top of that and the road swings to the left again, that you’re done, but instead it levels off briefly before kicking up to the 20s again. That last section really hurts. Needless to say, I’ll be headed back to try it again when it’s not the last hill of the day. Almost certainly one of the hardest in the state, in the same league as Jenckes and South Court. Probably slightly easier than Jenckes, slightly harder than South Court. Brentacol has now been updated with a more accurate (but still not quite representative of what it actually felt like):



Case Mountain and Hop River Trail

Easier day on the bike today for my second day in CT. This time I put cross tires on the bike before hitting some nice dirt in the Manchester area. I’ve had the Strava KOM on the Case Mountain segment since back in 2010 when I first got the cross bike (or to be precise, I didn’t get the segment until I joined strava and uploaded all of my old rides last year sometime). I always figured I would lose it at some point and indeed, my time is now only holding the lead by a mere second. So I figured that now I’m back in shape I should go out and see what I could do about lengthening my lead. But as I discovered yesterday when I was only able to beat my Mount Tom time from a similar period, I was actually in decent shape back in the fall of 2010.

Add to that, Case Mountain is a mess (which is also probably why people haven’t been lining up to steal my KOM). Since the last time, there’s a massive construction project on the bridge at the bottom, so just getting to the climb is a chore. And the trail surface is greatly deteriorated, now featuring many ruts and larger rocks in the trail. I rode it as hard as I could, but about 1/2 way up, I was a bit confused about which way the trail was turning, hit a patch of loose rocks, and was forced to unlcip. I was in a pretty easy gear and it took me a while to get the momentum/balance I needed to get going again. Kept going to the top. Later in the ride I tried to calculate whether I might have gotten it since I forgot to set a lap. I figured the total time for the climb was about 1/2 Mount Tom, which I had managed to beat by a minute, so if I rode it perfectly I might expect to shave 30 seconds off the time. I guessed my mishap in the middle probably cost me about 30 seconds, so I figured it would be a wash and could go either way. When I checked at home, I missed the KOM by 4 seconds. Oh well.

Then I rode the Hop River Trail, which is luxuriously smooth and fast on a cross bike. It gets a little bumpier the further east you go, but there are also fewer other riders/walkers, so it’s easier to cruise.

Mount Tom/Mount Holyoke/Sugarloaf


Scratch that. No Sugarloaf. In my never-ending quest to make sure that no one will ever agree to accompany me on a ride, I came up with the following route and conned Dave B. into going with me:

The plan was to hit Mount Tom first, but take an alternate route looping down via Easthampton Road, and around Whiting Reservoir. Dave probably had his first clue that I didn’t have my shit together when we turned onto Fort Road and it quickly became dirt. Not problem, however, since I was riding my cross bike with road tires (25s) and Dave had a cross bike (for gearing, primarily) and cross tires.

The first hill of the day was the west side of Reservation Road that climbs up the to the western entrance to Mount Tom Reservation. I’ve done the eastern approach before and assumed this would be similar. A few meters after turning onto the road, however, I wondered what the sign about “No through traffic” meant. Hmm…ignore it. After climbing on pavement for a little while, we came to a gate followed by what looked like a messy gravel road. Checked with Dave to see his opinion, he said he was game to give it a try. For about 1/4 mile it was fine. It became clear that there once was a paved road here, since there were many small-to-medium patches of pavement mixed in with all the gravel. Then we got to the part pictured at the top of the post with huge ruts that were not navigable on a road bike, at least not for me. Options were to turn around (and risk flats descending on the gravel) or hike the remaining portion, which I estimated to be about 1/4 mile. We chose the latter. At the top, we rode through the park to 141 and then took a left into the Whting Reservoir. More gravel, but this was nice and smooth. Lots of hikers and recreational walkers, so we took this nice and leisurely.

Then for the main attraction. Mount Tom. For the first time I went into the climb not being particularly afraid of it, but with a nagging suspicion that I should be. My previous best on Strava (though probably slower than the time I did it with a 34-29) was 12:01, and I was hoping to crack 10 minutes today. In the end I only managed to shave a minute off and got to the top at 11:01. I was hoping the Strava comparison would be more useful in seeing if I went out too hard today or anything, but my time today gained pretty gradually and consistently over the whole climb. In any case, I felt good for the first half, but had to fight the urge not to stop for a breather through the entire last section. The road surface, by the way, already bad, has deteriorated significantly since last time. Dave made it up without walking but had to stop a couple times.

No more dirt-road drama after this point, though we did take the dirt southern approach to Mount Holyoke, but that actually makes the climb slightly easier than taking the more northerly road up from Rt. 47. After that, the plan was to ride up the eastern side of the CT river to Sugarloaf, but given all the dirt (and Dave had a flat early on) we were already at over 3 hours for ride time, and I was already thinking about Korean food in Hadley, so we ditched the last part of the ride when we got back to Hadley.

quick update on Jenckes

So I added to my arsenal of gear for measuring max gradients and bought a car mount for my phone. Calibration is a bit tricky, because you obviously need to have your car on a completely level surface when you set it (you also need it turned off because the vibrations are enough to keep the inclinometer from settling on a number.) But I think I had a decently flat spot when I set it. That said, I can’t guarantee the accuracy of the following measurements to greater than a couple percentage points. I also don’t know what sort of compression happens on the shocks, etc, that might keep the car from matching the true angle of the road. I drove up Jenckes from Star. On Star I got readings in the low 20s, which is about what I expected. And I picked up somewhere around 19-20 (I didn’t write it down) on the section of Jenckes just before Pratt (I didn’t walk that far down the other day). And the section near Congdon came up with a pretty similar reading, around 19, to what I got the other day. Definitely not the most accurate way of measuring this, but a nice option to have if I want to estimate a hill while I’m out driving.

Blog Beautification

As you can see, if you’re looking at this site, I finally decided it was time to upgrade my wordpress installation (which was probably 5 years old and horribly insecure…) and decided to move it into the brentacol directory. Any old links will probably stop working, but if you go to the old address you will be redirected here.

Max Gradients

Maximum gradients are kind of silly, but they are also fun to know and talk about. The problem is that you can’t get a reliable max gradient from any kind of GPS/Map data (I mean, I looked down and saw 75% on my Garmin going up Hurricane Mountain back in June…It’s steep, but not that steep). In order to make the mapping data reliable, you need to make your sample (and minimum segment length) long enough to average out any really wacky spikes. That means that even if a hill has a small very steep section, I generally won’t trust it without having seen it for myself. And your subjective opinion from looking at a hill is not a whole lot better. I’m pretty good at judging grade by look and feel, but it’s also very context-based; 12% in the middle of Providence looks much steeper than the same gradient on Mount Washington. So my BRENTACOL maps show max gradient, but it’s only the steepest recorded segment, not the true max gradient spike. For that, the only real way to get it is to measure it by hand.

With that in mind, I was starting to research getting myself a cheap inclinometer, but when I started googling “inclinometer” the first suggested search was something like “inclinometer android app,” so I realized I probably had everything I needed to do this already. The first one I tried was Inclinometer Free, which seemed to work well, but had a few drawbacks, one being that it only gave angle of incline (not the biggest deal since you can just calculate gradient from angle by taking the tangent of the angle). The bigger issue is that it only gives whole-number angles, which translates to showing only about every 2%. So I settled on purchasing the 99-cent Clinometer app, which took care of both problems. I’d like to mount the tablet to a board or something, to take a slightly longer sample size than just the tablet, but for now I can just put the tablet right on the ground to get a measurement.

With that, I took the tablet out to Jenckes and Bowen, and here’s what I found:

1. Jenckes – I only tried a couple spots, but the steepest I found was 18.2%, about what I expected. (This is from just below Congdon.)


2. Bowen. The highest I found on the sidewalk section at the top was 32.2%, and on the pavement past Pratt, it went up to 32.9%. You’ll also notice that on gradients that high it becomes necessary to stop the tablet from sliding back down the hill.


Fun with Strava Names

Speaking of Strava. For a long time I had the KOM on River Ave., a short little hill less than a mile from my house. I blogged about it way back when. I was pleasantly amused when it first showed up in my KOMs that someone had entered it as “Mont Fleuve.” And I was even more pleasantly surprised that the other two people listed in the standings were none other than Keith Kelly and Adam St. Germain, and both can still beat me on any hill they care to challenge me on. (Adam doesn’t seem to be on there any more, but I’d swear he used to be…who knows.) And my time (somewhere around 40 seconds) was good, but not as good as I knew I was capable of. So I went back sometime this year and tried to best my own time. I did, by like 1 second. But the funny thing about strava is that when you do that, it sends the “You lost your KOM” notification to whoever is second place. So I think that must have pissed off Keith enough to bring him out of hiding to get it back, which he did. But then last week, now that I’m finally hitting my stride, I managed to get it back again. No doubt Keith will punish me for my insolence. But still. It’s fun while it lasts.

When I got it back, I posted something to facebook and mentioned how much I appreciated whoever had put the segment in as “Mont Fleuve.” I actually kind of suspected Keith. Turns out it was my new friend and new teammate Dave. So when I decided to add Mount Pleasant (up Atwells) to Strava, I decided to have a little fun and put it in as “Aangenaamberg” – literally “Pleasant Mountain” or “Pleasant Hill.” Dave hoped this might become “a thing” so I changed the names on a few more that I had already created:

  • Aangenaamberg
  • Liefdegildemuur – literally “Love-guild-wall,” which was as close as I could get to “Lovecraft Wall” for the combination of Star and Jenckes in Providence. For the record, there are already WAY too many Jenckes segments, but none starting at South Main, and frankly I don’t care how fast you climb the second part if you don’t start with Star. I’ve started doing the top section from Benefit occasionally because the whole thing wasn’t on Strava (and somehow managed to KOM on a version someone made that goes well past the crest of the climb), but really if you don’t go all the way from Main, it’s not the real thing.
  • Eikelberg – In Phippsburg, ME, a nasty little dirt/gravel climb called Acorn Hill Road.
  • Kalksteenstraat – Chalkstone in PVD.
  • via Astuta – Smart St. in PVD. (recently reclaimed the KOM)
  • via Cristoforo – Christopher St. in PVD. (also recently reclaimed)
  • and my favorite: Helling van de Varkensduivel – Dark Entry Road, the road up to the haunted (by pig demons?) abandoned town of Dudleytown.

New Brentacol Feature

Hilljunkie just posted a couple weeks ago his annoyance at the proliferation of stupid STRAVA segments, which prompted me to add something to BRENTACOL that’s been knocking around in my head for a while. Namely, because there are multiple versions of pretty much every hill on STRAVA, and no real moderation system, there’s no real sense that any one of those segments is the real thing. In most cases, if you win one, you’ll probably win the other, unless they’re mapped very differently, but it’s kind of annoying nonetheless. So, granting that BRENTACOL isn’t really high enough profile to really make this work, I figure I can “certify” a given route by displaying it’s leaderboard from the hill’s description page. Just click on the link that appears for any hill that I’ve registered the strava leaderboard and you’ll see the hill. Currently I’ve only added it for a couple, but here, for example is Blue Hill.