This weekend I got myself into (for me) a hairy situation while rappelling. I got to a set of rap anchors that I needed to transfer to in order to get down to the next ledge, so I transferred myself to a quad anchor, took myself completely off rappel, and ran the rope through the rap rings at this anchor (retying my disaster knots of course).

These anchors sit at the top of a massive roof, so at this point the quad was completely loaded, and I was unable to effectively support my weight on the rock and clean at the same time. I set my rappel back up (guide ATC on a chain reactor, maybe 1 or 2 loops away from my harness), tied my prusik (below the ATC), and tested that the prusik would hold.

I'd never been in a situation where it's not possible to completely unweigh your anchor, so I made a rash decision to girth hitch a double length sling on my prusik, and tie a simple ladder into it.

My thought process was this:

  • I need to move my weight from the quad onto the rope in order to clean the quad
  • The prusik is a friction hitch that by design is meant to hold your entire body weight
  • By girth hitching directly to the prusik, I'm loading it and forcing it to grab the rope
  • I can then support my weight entirely on the rope and clean the anchor

It was only after I got back on the ground and my adrenaline was settling down that I realized I could have committed a grave error and unnecessarily loaded my one backup, which, if it had failed while I cleaned, would have led to my death.

I then came to the conclusion it would have been much smarter to simply tie the sling directly to the rope in a friction hitch, which may have failed but would not have affected my actual lifeline, the prusik.

Was I in any danger here? Or was I using the prusik as designed?

  • Trying to understand the scenario: So you were hanging from the quad anchor (no ledge to stand on) and then also supported by the ATC rappel, with prussik backup? Trying to shift weight from quad anchor onto rappel? And connected a sling with knots in it to the prussik and stood up on that? You couldn't connect the sling ladder directly to the rap bolts/chain/rings? Or do a pull-up to them just long enough to unclip the anchor carabiners?
    – endolith
    Jan 27, 2021 at 18:05
  • 1
    My memory is a little fuzzy now, but essentially it was a situation where the bolts were slightly above a small ledge, so that the carabiners clipped to them were almost horizontal when loaded (quad came over the ledge and the two biners on the end hung vertically). I tried several times to do a pullup, but it was just a little too awkward to get them unclipped. Surely I'm not the only one who has difficulty unclipping partially loaded screwgates above a roof! And yes, I know there were MANY other ways I could have done this safely, I just wanted to confirm that what I did was very unsafe. Jan 27, 2021 at 22:06

2 Answers 2


I'm not 100% sure I'm understanding all the details, but here's my take on this.

This seems like a really dangerous thing to do. A prusik is never guaranteed to hold. Sometimes a prusik needs to be tinkered with, dressed, etc., to make it hold. It should never be the only thing standing between you and the long ride.

Your method of unweighting the anchor seems sensible, but it shouldn't have been your only support. A simple way to back this up would have been simply to tie an overhand on a bight in both strands of the rope, down near your feet, and clip this to your harness with a locking biner. Then if all else fails, you just fall a few feet.

  • 1
    +1 for tying a "chicken knot." Doing so is good praxis for most things in the vertical world.
    – erfink
    Jan 27, 2021 at 4:47
  • 1
    After researching this a lot, this was my conclusion too. I didn't back up properly and that was the biggest mistake. Thanks, I'm gonna hug the ground now Jan 27, 2021 at 14:21
  • I agree with the answer, any knot can fail and should not hold your life as a single knot.
    – Willeke
    Feb 1, 2021 at 14:33
  • @Willeke There are many times we rely on a single knot to hold -- e.g. clove hitches, the various tie-in knots, munter hitches, flat overhand bends. Perhaps a better generalization is not to rely on a single friction hitch?
    – Ryan Amos
    Sep 10, 2021 at 19:27

Much depends on how frequently you use prusiks. Back in college when i was caving a lot, we used prusiks tied in 1/4" Plymouth Goldline using a 9/16" chunk of goldline as our ascent/descent line. I never had an issue, nor did anyone in our group.

For those of you unfamiliar, Goldline was a very hard lay stranded rope. As stranded rope, it was more obvious when it was damaged, easier to clean. Since caving ropes take a beating with dirt, they were cheaper to replace.

That said: We always used 3 loops: one for each foot and a chest loop, and a new loop took some training of it to cinch reliably.

Also, in my daily life, I've used either a pipe hitch or a prusic to pull well pumps. I would expect that having a knot fail on 1" poly pipe covered in iron bacteria slime would be more prone to slip than on a rope. But I've not had one of those fail yet.

And that's the issue, "Yet" You get few do-overs for a safety mistake in mountaineering. In general, you should never be in a situation of dying from a single failure mode. Two prusiks, and a knot below them? Sure. Single prusik? Ungood.

You probably could get away with it many times.

I don't like to use the word 'accidents' for times when things go pear shaped. I call them 'incidents'. Accidents implies "It just happened. Nothin' anyone could do" when in fact incidents fall into two categories:

Simple Incidents The group did not have a real plan or safety doctrine, and a single mistake led to injury or death.

Compound Incidents The group does have a safety doctrine. Several different events and circumstances combined to make a situation where someone was injured or killed.

You have done a clever thing: You analyzed a situation where no harm was done, but harm could have happened, and concluded that you could have done better.

All safety protocols are inadequate in that they cannot prevent all harm. "A ship is safe in harbour, but that's not what ships are for." We can mitigate risk. But the only way to mitigate it completely is to stay home. So we compromise so that the chance of harm is reduced to very small levels, and accept the remaining risk.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.