Tie in knots

climbing friend,

if you would like to experience true terror, tie in with a "instrument of death" bowline and get someone with a PAS thong to belay you with deadly ATC on runout trad route. if you are making out alive, you would experience the true personal growth and power most profound.

Nick Goldsmith wrote: I have never heard of a  figure eight coming untied and killing someone. The same can certainly not be said for the bowline. 

Nick, here are a few accidents, unearthed at random, involving figure 8-s, one of which involves the knot untying (with a short end).  Not all were fatal, but some were.

--------------------------------------------------------------------------------------------------

 ...“I completed my figure 8 knot, but at the exit of the knot I didn’t leave enough rope. I left only 1.5 to 2 inches of rope. I really didn’t feel like redoing my knot and had done this [in the] past with no issue. And I just started climbing without performing a partner check.…I fell away from the wall as the rope took my weight and slowed me down for a fraction of a second. The knot had come undone [and] I was falling toward the ground.”

http://publications.americanalpineclub.org/articles/13201213850/Fall-on-Rock-Incomplete-Tie-In-Knot

...Savageau leaned back to scope out the pitch, and immediately fell, as the figure eight knot had not been finished.

http://publications.americanalpineclub.org/articles/13200105202/Incomplete-Tie-In-Fall-on-Rock-California-The-Needles-Sorcerer-Needle

…an online analysis from a close friend who was belaying at the time stated that the climber was likely distracted after passing the rope through the anchor chains. It is believed she threaded the rope through the tie-in points in her harness without beginning or completing her usual retraced figure-8 knot. She is believed to have tied an overhand knot, intended as a backup, that briefly held her weight after she unclipped her tethers and before she fell.

http://publications.americanalpineclub.org/articles/13201214439/Fatal-Tie-in-Error-in-Maple-Canyon

...According to witnesses, the cause of the accident was an unfinished tie-in knot. The climber tied her own knot, but unfortunately did not finish it….The rope end was found at the anchors with a figure-of-eight knot and working end of the rope trailing from the knot.

http://publications.americanalpineclub.org/articles/13201202400/Fall-on-Rock-Unfinished-Tie-in-Knot

----------------------------------------------------------------------------------------------------

I'm totally done arguing relative merits, but think it worthwhile to set the record straight about something else.  This teapot tempest started when I said I didn't care if others can't check my bowline with tucked Yosemite finish.   I didn't issue some blanket rejection of all partner checking.  Whether the knot is a bowline or figure 8, accidents happen because someone doesn't  finish the knot (or in some cases doesn't even start it).  So checking that your partner has a knot makes a lot of sense; its a part of attending to critical climbing issues as opposed letting other things preempt your attention.  I always watch my partner tie in (when we are together), and ask them if everything is ok if something looks amiss to me.  I also watch what they are doing at the anchors, where there are also various possibilities for effing up.  Of course, we aren't always visible to each other, first because of my usual practice of not having the second tie in until the ropes are taken up, and second because some lowering protocols require the tie-in to be untied and retied out of sight of the belayer.  

As for the knot that started all this, I tie it exactly the same way every time and could do it blinfolded.  I then tighten it by pulling, one at a time, on the four strands exiting the knot body, at the same time verifying that the rope is through both tie-in loops. This constitutes a deliberate re-checking of what every strand is doing.  Then I check my harness to make sure it is properly buckled, and away we go. I don't shortcut this little ritual, and once completed there is no question that knot and harness are as they should be.

I wasn't always this careful, and many years ago made a mistake like Lynne's and John's that could have been fatal, but wasn't because it just converted the ascent from a roped endeavor to a solo, and I was used to that and could handle it.  In those days there were no extended discussions like this about the potential for being distracted and the fatal consequences that might ensue, so this extra level of care turned out to be another in the extensive list of self-taught attributes.  We live in different times, when it is more than abundantly clear---and extensively publicized---that inattention kills.  Whether you fight inattention with formal buddy checks or find other ways to keep what is critical in focus, the message hasn't changed since Whymper's Matterhorn accident a century and a half ago.   "...remember that...a momentary negligence may destroy the happiness of a lifetime. Do nothing in haste; look well to each step; and from the beginning think what may be the end."

I think it goes like this:

You only hear about bowlines being mistied
John Long and Lynn Hill used bowlines
John Long and Lynn Hill are famous
You don't hear about non-famous people

Conclusion.... Famous people use bowlines

Mostly not, but Adam Ondra does.
 

finished Fig 8 will not come untied. finished standard bowline will loosen up if not constantly weighted. apparently you use a version that addresses that issue???

Nick, over the years I used two types of bowline finishes.  One was the Yosemite finish with a barrel knot (half of double fisherman's), and more recently the one I mentioned in the post, the tucked Yosemite finish, which is more compact.  Both of these have lasted through many long days without loosening with the ropes I've used.  I do make a point, as I said just above, of very carefully tightening everything up when I first tie the knot.  (Occasionally, I recall the barrel knots needed tightening.)

rgold wrote: To help Brian out a bit, here are some articles listing several instances of bowlines coming untied.  Lynne Hill's situation is included, so we don't know how many of the incidents were due to not tying a knot at all.  In the first one mentioned, a bowline was tied, but we don't know about the presence of a finishing knot.  http://rockandice.com/climbing-news/tnb-when-your-rope-falls-off-and-5-ways-to-prevent-the-nightmare .

http://rockandice.com/climbing-accidents/bowline-comes-untied-climber-falls-to-ground/

From Abram Herman: "I have a coworker who fell at a Denver gym after her half-tied figure eight failed when she went to lower from the top of a climb."

I've seen two more accounts of figure 8's failing on reddit but can't find them now.  As far as cursory sampling the media goes, I'm finding both types of failures but more bowlines.  The confusion between inattention causing knots to be incomplete or not even begun and knots "untying" is pervasive.

As I said earlier, a bowline that isn't equipped with a finishing knot isn't a climbing knot, and if you are going to check something it should be whether there is some kind of stopper knot installed.  Leaving a bowline (or a double bowline) tied loosely with no stopper knot is indeed a recipe for death.

I do think I'm obliged to recalibrate some of my comments, because of the situation of a loose unsecured bowline coming untied.  Tying such a knot is a kind of inattention, but doesn't seem to be quite the same as inattention causing no knot to be tied at all, and this makes Greg's challenge about bowlines being easier to eff up more substantial, even though I'd prefer to say an unsecured knot hasn't been finished.  There is no equivalent error with a figure 8; you can tie it pretty loosely and stopper knots aren't required anyway and the knot still won't untie.

Once again...very well stated.  

Essentially all of the examples of accidents involving any type of knot are cases where people did not tie their knot "wrong", rather they did the first step, put the rope through their tie-in points, and then got distracted and failed to complete the knot.

So this business of one knot being easier to inspect than another is really silly.  Yes, anyone looking closely can tell whether you have rethreaded your figure 8 or not.  Now think how much easier it is to tell whether you've completed your bowline.  If you start your bowline, put the rope through your tie-in points, and then get distracted, what you have is a rope through your tie in points - no knot whatsoever.  Not difficult to recognize if you look!

I'm not just being pedantic here.  I once was climbing with an acquaintance.  While I was doing other things, he tied in, and then said he was ready to climb.  I put him on belay, but failed to ask to check his knot.  I could see something behind his hip that looked like an eight, but had a really long tail.  I had never climbed with him before, and assumed that rather than tie a backup knot on his fig-8, he preferred to just leave a long tail.  Fine, I thought.  Until he got 20 feet up, placed his first piece of gear, reached down to clip, and pulled his rope out of his harness.  He had only started the fig-8 and threaded it through his tie-in points, but never followed through the knot.  That explained why I saw something shaped vaguely like a fig-8 with a long tail behind his hip/under his leg.  Fortunately he was at a good enough stance that he was able to rethread the knot and continue.

Had the same climber tied in with a bowline, I would have seen nothing resembling a knot, and might have asked him to turn around so I could see his knot (normally I check my partners knot).  I'm not trying to excuse the fact that I failed to check him, but a rope with no knot whatsoever in it is distinctly different in appearance.  It was warm, and we weren't wearing layers like in Lynne Hill's situation.  

Cheers,

GO

To simplify the massive responses here;

A properly tied figure eight knot with a sufficient tail will never fail.

A properly tied bowline knot with a sufficient backup will never fail.

In my opinion, the belayer should check to confirm the rope on the climber is going through both hardpoints and has a knot tied afterwards. The knot itself should be the responsibility of the climber; use whichever one of the two knots above you can tie correctly 100 out of 100 times.

Only five pages? Weak 

GabeO wrote: Essentially all of the examples of accidents involving any type of knot are cases where people did not tie their knot "wrong", rather they did the first step, put the rope through their tie-in points, and then got distracted and failed to complete the knot.

So this business of one knot being easier to inspect than another is really silly.  Yes, anyone looking closely can tell whether you have rethreaded your figure 8 or not.  Now think how much easier it is to tell whether you've completed your bowline.  If you start your bowline, put the rope through your tie-in points, and then get distracted, what you have is a rope through your tie in points - no knot whatsoever.  Not difficult to recognize if you look!

I'm not just being pedantic here.  I once was climbing with an acquaintance.  While I was doing other things, he tied in, and then said he was ready to climb.  I put him on belay, but failed to ask to check his knot.  I could see something behind his hip that looked like an eight, but had a really long tail.  I had never climbed with him before, and assumed that rather than tie a backup knot on his fig-8, he preferred to just leave a long tail.  Fine, I thought.  Until he got 20 feet up, placed his first piece of gear, reached down to clip, and pulled his rope out of his harness.  He had only started the fig-8 and threaded it through his tie-in points, but never followed through the knot.  That explained why I saw something shaped vaguely like a fig-8 with a long tail behind his hip/under his leg.  Fortunately he was at a good enough stance that he was able to rethread the knot and continue.

Had the same climber tied in with a bowline, I would have seen nothing resembling a knot, and might have asked him to turn around so I could see his knot (normally I check my partners knot).  I'm not trying to excuse the fact that I failed to check him, but a rope with no knot whatsoever in it is distinctly different in appearance.  It was warm, and we weren't wearing layers like in Lynne Hill's situation.  

Cheers,

GO

Actually, depending on the bowline, you can have a full strength knot if you don't finish it.  I use a bowline on a bight, where i thread the rope through my harness, tie a bowline, then run the tail through my harness again, retrace the bowline, and then tie a backup knot.  If I don't retrace, I still have a completed bowline tying me (with no safety knots) in to the rope.  With an 8, you end up with nothing, as you noted in your real life scenario.

Andrew Hess wrote:
+1
Both are pretty damn easy to tie. I can think of....well...almost anything else in climbing that is easier to screw up.

Why don't we just tie a bowline and then back it up w/ a rethreaded f8. There. Everyone's happy.

Funny enough, I tried this once. The figure 8 cinches down before the bowline, negating the benefits of the bowline.

I have been tying in with a double bowline w/a double fisherman's finish for 20+ years with zero issues.

I have listed this before, but there are 5 reasons I prefer the bowline over the figure 8:

1. No secondary knot to untie before pulling the rope.
2. Easy to untie after a hard fall
3. It is virtually impossible to tie a bowline wrong - it just falls apart.
4. The tail is towards the harness helping avoid clipping issues (I know, the figure 8 can be finished this way as well)
5. It is an easily adjustable knot.

Matt Himmelstein wrote:

Actually, depending on the bowline, you can have a full strength knot if you don't finish it.  I use a bowline on a bight, where i thread the rope through my harness, tie a bowline, then run the tail through my harness again, retrace the bowline, and then tie a backup knot.  If I don't retrace, I still have a completed bowline tying me (with no safety knots) in to the rope.  With an 8, you end up with nothing, as you noted in your real life scenario.

I think Edelrid depicts this knot in the rope safety guide, sold with its rope.  I could not understand the knot as depicted by Edelrid. Can anyone post a link showing how to tie MH’s not.  

Floyd Eggers wrote: Only five pages? Weak 

I know, I was hoping for bloodshed when I started this thread.

Mark Cowan wrote:

Funny enough, I tried this once. The figure 8 cinches down before the bowline, negating the benefits of the bowline.

What? How? That's not supposed to happen.

Tyler Newcomb wrote:

What? How? That's not supposed to happen.

It's because the standing end of the figure 8 ( the pre follow through fig 8) is still on the working end of the rope, therefore you cinch it down if you weight the rope.

Mark Cowan wrote:

It's because the standing end of the figure 8 ( the pre follow through fig 8) is still on the working end of the rope, therefore you cinch it down if you weight the rope.

I don't think you left enough slack then. I think they meant use a bowline with a really long tail, and tie a super loose figure 8 in the end.

The section of rope where the figure 8 is, will always be closer to the biner catching the fall, therefore it's the first part of the rope to cinch. Doesn't matter how long you make it. Maybe if you did the figure 8 as the stopper knot inside the bowline's bight?
Edit: that doesn't work either. Still cinches first.

We need a picture of "back it up w/ a rethreaded f8".  My interpretation has the problem Mark describes.  It's possible Andrew meant something else, but I can't (yet) imagine what...