What defines a Bowline? - structure, characteristics, topology

[Stalker]

I have to go with what DerekSMith said. One of the fundamental aspects of a bowline is the inclusion of a Sheet bend. 

[agent_smith]

I have to go with what DerekSMith said. One of the fundamental aspects of a bowline is the inclusion of a Sheet bend.

That makes absolutely no sense.

A sheet bend is an 'end-to-end joining knot'.  It is not an 'eye knot'.

By definition, a Bowline is an 'eye knot' - because it has a connective interface (the 'eye') - which enables the Bowline to attach to something (eg a carabiner, or to a climbing harness).

I think what you meant to say is that Bowlines have a SB core (meaning that there is a 'bight' segment and a 'nipping loop' segment...although in the case of a Sheet Bend, the nipping loop is 'partially formed'). This is the basis of Derek Smith's position.
I have difficulties with the SB core theory...because some Bowline structures dont quite fit this definition. An example is the 'Lee Zep Bowline' and #1033 'Carrick Loop' (per Ashleys naming). Both of these structures have a 'nipping loop' as a key mechanism and they are both 'eye knots' that are jam resistant. I view both these structures as 'Bowlines'. The central nipping loop as a key mechanism that all Bowlines share is a theory that does seem to work - while the SB core theory is too narrow and does not apply to some structures.

In my personal view, there is no functional nipping 'loop' in a Sheet Bend. And the Sheet Bend is not an eye knot.

And this comes back to what is the definition of a 'loop' (and indeed...what is a 'nipping loop'). I had required that the nipping loop must be loaded at both ends in order to qualify. In a Sheet Bend - the nipping structure is not loaded at both ends...and so it is not a 'nipping loop'.

In my paper - I have advanced the theory that all Bowlines have a 'nipping loop' as a key mechanism. This of course was a concept originally advanced by Dan Lehman - which I concur with.

Other features which all Bowlines share include; a connective 'eye' (giving rise to the term 'eye knot' - because the 'eye' enables connections to be made), resistance to jamming, and 'PET'.

Many of these concepts were independently advanced by Dan Lehman, Xarax, and a host of other IGKT members over a very long period of time.


Mark Gommers

[knot rigger]

In a Sheet Bend - the nipping structure is not loaded at both ends...and so it is not a 'nipping loop'.

How about a double sheet bend?  Does it have a nipping "loop"?

How about a sheet bend as a netting knot (aka weaver's bend)?  The nipping "loop" is loaded on both ends in that case?

cheers
andy

[knotsaver]

In a Sheet Bend - the nipping structure is not loaded at both ends...and so it is not a 'nipping loop'.

How about a double sheet bend?  Does it have a nipping "loop"?

Hi Andy,
to my mind it does not! the end isn't loaded, we could say that it is locked.
Please notice that another difference  in the Sheet Bend is that the Standing Part of the U turn is 100% loaded,  whilst in a Bowline it is 50% loaded (Edit: roughly speaking, I think it is not so simple to understand how the load is shared between the two eye-legs in a real case, it depends on the angle between the eye-legs too: the extreme case is the ring loading)

How about a sheet bend as a netting knot (aka weaver's bend)?  The nipping "loop" is loaded on both ends in that case?

In that case I think we shouldn't call it a Sheet Bend , it isn't a Sheet Bend...

Ciao,
s.
p.s. @Mark happy to "read" you again 

[Dan_Lehman]

In a Sheet Bend - the nipping structure is not loaded at both ends...and so it is not a 'nipping loop'.

How about a double sheet bend?  Does it have a nipping "loop"?

Hi Andy,
to my mind it does not! the end isn't loaded, we could say that it is locked.
Please notice that another difference  in the Sheet Bend is that the Standing Part of the U turn is 100% loaded,  whilst in a Bowline it is 50% loaded (Edit: roughly speaking, I think it is not so simple to understand how the load is shared between the two eye-legs in a real case, it depends on the angle between the eye-legs too: the extreme case is the ring loading)

It gets tricky to base definitions upon physical characteristics,
as those can be "YMMV" per materials (and loading force)!!

E.g., I've fiddled what I regard as *bowines* where the U-part
doesn't lead to the eye, or where the continuation of the turNip
runs not into the eye but into a collar around the eye legs
(and so might be seen to not really be loaded).

And what of the multi-eye bowlines?  If load is distributed evenly
over alllll of those eye legs, that one leading back to the "Is it a
nipping loop?" part must have only a fraction of the load, right?!
How much is enuff?

Knotty nuances!

 

[agent_smith]

It gets tricky to base definitions upon physical characteristics,
as those can be "YMMV" per materials (and loading force)!!

Tricky yes - but we can say the same thing with many areas of science and physics (particularly quantum physics). But this by itself should not deter humans from seeking to understand and describe and measure the physical world.

I am surprised that you did not speculate about ABoK #1117 (The running Bowline noose).
This is a structure that is worth examining to see if it deserving of having the title 'Bowline' in its name.

What is #1117?
Certainly, it is a 'noose'.

Can it be described as a 'Bowline'?

Strictly speaking - it is not an eye knot (it is a noose). Bowlines are understood to be 'eye knots'. In a classical sense, Bowlines have a fixed 'eye'...however some have a communicating segment which contradicts the notion of 'fixed'.
#1117 does not have a 'communicating' segment as does #1083 and #1087 (see below).
Perhaps it is a composite Bowline noose?
I would be interested to learn of your thoughts here...

I used #1117 as a follow up to your contention re 'nipping loops' (for which you still seem to prefer to use the term 'turNip').
In #1117, the 'nipping loop' would still be functional when the noose structure is loaded and cinches up tight against an object.

And what of the multi-eye bowlines?  If load is distributed evenly
over alllll of those eye legs, that one leading back to the "Is it a
nipping loop?" part must have only a fraction of the load, right?!
How much is enuff?

And some examples include: (list is not exhaustive)
[ ] #1083 (Double Bowline on the bight) - this structure has a functional 'nipping loop'
[ ] #1087 (Spanish Bowline) - this structure has 2 functional 'nipping loops'
[ ] #1088 (Sheepshank knot with Half hitches) - interesting structure - does it have a functioning 'nipping loop'? Yes - it has 2. But, the mere presence of a 'nipping loop' does not by itself qualify the structure as being deserving of the title 'Bowline'.

In #1088, there are 2 nipping loops - but, it has no collar. The collar is a key element of a Bowline.
And all Bowlines are 'eye knots' - but it gets murky here because in all three cases - the eyes have a 'communicating' segment...and so the 'eyes' are not of a fixed dimension.

Ashley correctly identified #1083 and #1087 as 'Bowlines' - because both structures have a collar and a nipping loop (but the eyes are not fixed).

#1088 fails to be a 'Bowline' on account that it has no collar .

As for your posit -

"How much is enuff?"

(per Dan Lehman in relation to the definition of a 'nipping loop')
The answer would seem to be any load - provided that there is load at both ends of the 'nipping loop'.
So this means zero load at one or both ends would disqualify a structure from being a 'nipping loop'.
Some/partial load still qualifies as 'load'.
And this is why a Sheet Bend has no 'nipping loop' - because only one end is loaded.

Summary:

Bowlines must have the following elements present:
1. A collar
2. A nipping loop (there may also be nipping loops)
3. An 'eye' (or more than one eye) - and the eye structure may have a communicating segment, in that it does not have to be of a fixed dimension.
A further quality of Bowlines is that they are jam resistant.
And a further quality is 'PET' (post eye tiable) - although not all Bowlines are PET for attachment to an object such as a climbing harness - so PET is not an absolute requirement.

Mark Gommers

EDITED for clarity and to properly respond to Dan Lehman's questions...

[Dan_Lehman]

I am surprised that you did not speculate about ABoK #1117 (The running Bowline noose).
This is a structure that is worth examining to see if it deserving of having the title 'Bowline' in its name.
What is #1117?

Deserving that, yes, but it's not a *knot* IMO,
but a knotted structure (the knot part being #1010).

Certainly, it is a 'noose'.

Okay, but a noose isn't a *knot*, but a structure.
(And put 2 half-hitches in this group --where the knot
is a clove hitch in this *noose* structure.)

[ ] #1088 (Sheepshank knot with Half hitches) - interesting structure -

QUITE !!  Where I have mused about ignoring the eye
of a bowline --and focused on a "cookie-cutter" knotted
part, solely, and of what of the 4 ends (parts exiting this cutter)
are loaded--, how then ... the sheepshank, for IT's center
strands run w/o definite length between knotted parts (two, as
you note), and aren't really a part thereof (knotted, i.e.).
QUITE a challenge.  (And the would-be "collar" bight is just
out in the air, indefinitely sized, too!?)

While we're at it, the twin bowlines structure asks YOU if
it is a *bowline* (if it has an "eye"), and us if it is a *knot*;
its knotted parts are more engaged than above, but are still
plural, separated, and ... !?  (And in neither the "twin..."
end-2-end structures nor the sheepshanks need the two
knotted parts (oh, yes, there can even be more than two
--center stuff for decoration/fascination!) be the same.

The collar is a key element of a Bowline.

Aha, Xarax lives!

Many of MY *bowlines* do w/o a collar and all seem happy at that. 

And all Bowlines are 'eye knots'

Unless I go on with the cookie-cutter *knot* idea,
which would focus solely on the entangled/knotted part.
Maybe "solely" is too strong?  --but it seemed so pure!

--dl*
====

[knot rigger]

As for your posit -
Quote
"How much is enuff?"
(per Dan Lehman in relation to the definition of a 'nipping loop')
The answer would seem to be any load - provided that there is load at both ends of the 'nipping loop'.
So this means zero load at one or both ends would disqualify a structure from being a 'nipping loop'.
Some/partial load still qualifies as 'load'.

I find this notion of definition based on loading pattern to be problematic.

Imagine a Round Turn Bowline (https://tinyurl.com/lrujzxs) tied around a sizable tree branch to suspend a child's swing.  Assuming sufficient friction between the tree bark and round turn, when the swing pendulums, a leg of the eye will slack.  Given the "loading" definition paradigm proposed, when the child swings the knot will transform from Sheet Bend, to Bowline, to Lapp Bend.

I propose that this result is at odds with common sense.

Another example would be that of a Long Tail Bowline, where the working end of a standard Bowline (ABoK#1010) or Outside Bowline (ABoK #1034 1/2) is intentionally left extremely long, for the purpose of being loaded (usually in a rope rescue application).  If their is load solely on the long tail of the bowline (ie working end) and not the loop eye, does the knot cease to be a bowline?  And if the loop were loaded again it would again become a bowline?

Taking this logic an absurd step further, would a standard (#1010) bowline tyed, dressed and set, but unloaded (ie standing end, and eye both slack) would it not be a bowline? Does it become a Bowline only when taut?

In general I mostly agree with Agent_Smith's summary:

Bowlines must have the following elements present:
1. A collar
2. A nipping loop (there may also be nipping loops)
3. An 'eye' (or more than one eye)

I'm not quite sold on some aspects of it, particularly that a nipping loop must be loaded on each end.
My own (current) conclusion is an unloaded Bowline is still a Bowline.

Quote from: knot rigger
How about a sheet bend as a netting knot (aka weaver's bend)?  The nipping "loop" is loaded on both ends in that case?

In that case I think we shouldn't call it a Sheet Bend , it isn't a Sheet Bend...

knotsaver; names of knots are based on common usage IMO, (see ABoK #952)  If a certain knot, in a certain application, that has been called by a certain name for hundreds of years, doesn't fit a proposed definition, I find it more reasonable to modify the definition rather than changing the name of the knot!

One last note,  I came across this topical article:

How Many Bowlines? by Glen A. Dickey in Knotting Matters issue 87

[agent_smith]

I'm not quite sold on some aspects of it, particularly that a nipping loop must be loaded on each end.
My own (current) conclusion is an unloaded Bowline is still a Bowline.

Per knot rigger...

I think there is a fundamental misunderstanding here...possibly caused by language.
Obviously, an unloaded Bowline is still a Bowline!

The absence of load is not the issue.

Going back to #1431 (Sheet bend) as an example - I posit that there is no functional 'nipping loop' in this structure.
The unloaded state does not change the fact that it is a Sheet bend
When load is applied, we can see that the nipping structure will have load at one end only (and so it can't be defined as a 'nipping loop').
The point is, a loaded or unloaded state does not change its identity as a Sheet bend.

If I tie a common #1010 Bowline and hold it loose in my hand - obviously there is no load.
Loaded or unloaded... its identity as a Bowline remains unchanged.
However, in the loaded state - we can see that the nipping structure experiences load at both ends - and therefore it can be described as a 'nipping loop'.

Mark

[knotsaver]

I'm not quite sold on some aspects of it, particularly that a nipping loop must be loaded on each end.
My own (current) conclusion is an unloaded Bowline is still a Bowline.

Per knot rigger...

I think there is a fundamental misunderstanding here...possibly caused by language.
Obviously, an unloaded Bowline is still a Bowline!

The absence of load is not the issue.

We are mixing two "layers": the static and the dynamic one and this often creates a misunderstanding, but we  have to be able to distinguish the nub of a Sheet Bend from the nub of a Bowline. I like the concept of a functional "nipping loop" and I think that it is fundamental in the  characterization of a Bowline.
Why do we have to say

One of the fundamental aspects of a bowline is the inclusion of a Sheet bend.

?
and why don't we say: one of the fundamental aspect of a Sheet Bend is the inclusion of a broken Bowline?
I think because they work differently...
Perhaps the Sheet Bend was tied before the Bowline, perhaps the first one was the Becket Hitch (who knows? does anyone know it?) but I think we have to distinguish one from another.

.
And a further quality is 'PET' (post eye tiable) - although not all Bowlines are PET for attachment to an object such as a climbing harness - so PET is not an absolute requirement.

About PETness (I agree with Xarax) I think it should be an absolute requirement...

knotsaver; names of knots are based on common usage IMO, (see ABoK #952)  If a certain knot, in a certain application, that has been called by a certain name for hundreds of years, doesn't fit a proposed definition, I find it more reasonable to modify the definition rather than changing the name of the knot!

hm? I exaggerated but we shouldn't name it a Sheet Bend.
ABoK #952??? 
for the name? ah ok so we can't name a netting knot  "Sheet Bend" (Ashley named it Mesh Knot #402 but he said...(he said: "is the ordinary way of tying the Sheet Bend when it is made with a netting needle."))

One last note,  I came across this topical article:

How Many Bowlines? by Glen A. Dickey in Knotting Matters issue 87

I think it does not help us in our attempt to define a Bowline...however it is interesting.

Ciao,
s.

[agent_smith]

Quote

    The collar is a key element of a Bowline.

Aha, Xarax lives!

Many of MY *bowlines* do w/o a collar and all seem happy at that.

Yes...Xarax lives!
I agree with him that a key element of all 'Bowlines' is the collar.
And I'd like to see some clear photos of these alleged 'Bowlines'!

The collar is not the only element though...because a 'nipping loop' also must be present.
And the 'nipping loop' must be in the form of a loop - the definition of a loop was defined elsewhere on this forum (in contrast to a turn or some other maneuver that doesn't qualify as a loop).
And this loop must be loaded at both ends (the amount of load is not defined, but it does not have to be in equal proportions).

And Bowlines are 'eye knots' (not bends or hitches)....although the 'eye' does not need to be of a fixed dimension.
Bowlines are also jam resistant and 'PET' (although 'PET' may be conditional).
NOTE: With regard to 'PET' - not all Bowlines are PET when attached to objects such as a climbing harness. For example, #1080 (Bowline on a bight) cannot be attached to a climbers harness via the TIB method (although in theory, a gigantic bight of rope could be passed over ones head and body to step through...but that would not be practicable and in fact would be silly to attempt). Mind you, a work-around would be to avoid the TIB method altogether and re-thread the entire knot to recreate it via single strand - but this is cumbersome. So the requirement for 'PET' can be problematic.
I am not convinced that PET is an absolute requirement for an eye knot to be awarded the title of 'Bowline' (although it certainly is an attractive proposition).

...

Curiously, Ashley referred to #1057 (single Bowline on the bight) and #1058 (single Bowline on the bight) as 'Bowlines'
I disagree, and the term 'Bowline' should not have been used in the title.
I see no functional 'nipping loop' in either of these structures - which automatically disqualifies them from being Bowlines.

Mark G

[Dan_Lehman]

And I'd like to see some clear photos of these alleged 'Bowlines'!

The Myrtle is just such a knot,
and coming in from the opposite side of the
nipping loop with the returning eye leg one
should re-insert (closing this eye-leg loop)
on the eye-side (vs. SPart-side as done w/Myrtle)
for surer holding.  (Make a double turn, and it matters
less!)

Bowlines are also jam resistant and 'PET' (although 'PET' may be conditional).
NOTE: With regard to 'PET' - not all Bowlines are PET when attached to objects such as a climbing harness.

Worse misspelling than 'prussic' even, this 'PET' vice 'TIB' !!

Not all of my *bowlines* are PET or TIB (some are),
but the PET ones by definition are thus when attached
--to ANYthing!

For example, #1080 (Bowline on a bight) cannot be attached to a climbers harness via the TIB method (although in theory, a gigantic bight of rope could be passed over ones head and body to step through...but that would not be practicable and in fact would be silly to attempt).

Ha, somewhere I just saw this very process promoted,
for a middle-man tie-in!  (In the case of some accident,
fall, tension ..., it makes getting free of the rope a real
challenge, alas.)

Mind you, a work-around would be to avoid the TIB method altogether and re-thread the entire knot to recreate it via single strand - but this is cumbersome.

It is also recommended, by DAV (German climbing) and
maybe some others.  In general, tie-in knots that entail
making a 2nd pass for a 2nd eye will prove to be more
overall-secure vs. complete loosening than others.

As for the assertion
"a <whatever eye knot or end-2-end knot or hitch> UNloaded
is still that knot,
I beg to differ.  Certainly, my direction points to awkward
*speaking*, but at least for the purpose of some formal
knotting discussion, it could well be that one would consider
things.  And there is some practical side in cases where the
particular loaded_*knot* just falls apart absent tension
(and so would be hard to posit as "STILL being..." --just *being*!).

My e.g. of "bowline" (and other eye knots) considered at the
"cookie cutter" view (all *ends* exit the cutter perimeter w/o
specific connection)
is of a barge pulled with a tow line tied to starboard cleat
and the barge's shorter like line tied to port cleat and then
joined to the tow line in a --well, what do you call it(?),
"bowline" :: any knotter given the cookie-cutter view
of just the entangled/knotted part and its ends' tensions
or not would immediately recognize it as such,
irrespective of the barge *interrupting --or completing?--*
the "eye" bight.
(Take this example but change angles to 120deg all 'round,
and then what?  --bit more change to put >120deg on both
sides of the short line (barge port cleat) and you then see
a "becket hitch"/"sheet bend", per loading!)

.:.  It comes down to how you want to speak, how you
want to regard entities, what purpose you address.  In
common parlance, yes, you don't want to have knots
coming & going per tension; though ring-loading might
acceptably be openly spoken of as transforming an eye knot
into an end-2-end knot (wrong bowline into right Lapp bend).

--dl*
====

[knot rigger]

"It is hardly necessary to name a knot, but it assists materially in finding it a second time if the occasion arises."

From ABoK #952

Part of what I read from Ashley's remark is that a name of a knot is only a convenient way to talk about it with another not tyer. "Bowline" and "Sheet bend" knots both derive their names from the the particular line on a ship with which one would tye these knots.  Saying "tye the knot that I taught you to attach the line that runs from the front of the boat up to the sail" isn't quite as efficient as saying "tye the knot used in the bow line", or even more simply "tye a bowline".  To state the obvious: the name (or classification) is just a convenient description of an idea.

Agent Smith:

Perhaps we are two not tyers separated by a common language.  The point I was trying to make in my earlier post wasn't simply a semantic discussion of "load" or "taut" and "slack".  It is, of course, absurd (as I mentioned) to propose that an unloaded bowline isn't a "bowline".  I will attempt to state my point more directly (while trying to be brief).

I assert that the conclusions of the proposed bowline definition scheme should be judged against common sense.  Should the implication of the definition paradigm be non-nonsensical, then the definition isn't clear and ought to be modified. (or perhaps it's the best we got and we should all stop discussing it, but what's the fun in that)

How should one judge "common sense?" I'm quite certain I don't have the definitive answer for that! But I propose that handing a "bowline" to be judged to a fellow worker of knots (sailor, arborist, climber, rigger etc) and ask him what it is:  If he says something like "that looks like a bowline" I would suggest it passes the "common sense" test.

If a knot can be tyed that passes the "I know it when I see it" bowline test, but fails to meet the (proposed) definition, it would mean (to me) that the definition is inadequate, incomplete, or ill-conceived.

Bowlines must have the following elements present:
1. A collar
2. A nipping loop which requires both ends to be loaded
3. An 'eye'

as per Agent_Smith, edited for clarity with my words in italics

In my earlier example of a Long Tail Bowline (ABoK #1034 1/2) which has the (admittedly unusual, but not unrealistic) loading profile that the standing and working ends be loaded, but the loop not taking any load.  By the common sense test, this is clearly a "Bowline", but it doesn't meet the "nipping loop must have both ends loaded" rule of the definition.

Now if you were to load the loop, then it would fit the definition.  As with the child's swing example I cited, the knot changing classification, or "what it is," based solely on how it's used and loaded is a concept that I intuitively find illogical.  I propose the following maxim:

"The dang-ol' knot I tyed can't change what it is after I tied the f*cker"

I do hope that my comments be taken as constructive, as I've intended them, and not merely contrary.

cheers
andy

[agent_smith]

Dan, I think you have split a sentence of mine into 2 parts - which alters the meaning...and then you critiqued the broken meaning.
The second part of my sentence had to be read with the first - otherwise the meaning is lost.

#1080 (Bowline on a Bight) is most easily tied via the 'TIB' method (tied-in-the-bight...or.. tying in the bight)...at least that is the ordinary intent.
My point was that trying to form an attachment with #1080 to a climbing harness via the TIB method is problematic. I stated that is COULD be done - but, it would be a ludicrous process. In that sense, #1080 is not readily 'PET' to a climbers harness via the TIB method. It was in that context that I said the requirement of 'PET' might be problematic. I am not 100% sure if 'PET' is an absolute requirement for all Bowlines! I think this needs to be given more thought...
There is a lot of jargon here and acronyms abound...its getting to be a bit of a mouthful.

Dan, the 'Myrtle' (in my view) does have a collar. And what of #1033 (Carrick loop)? Does this have a collar? Again I say yes!
I thought this matter had been debated endlessly and resolved?
Based on this - it appears that the definition of a 'collar' and its importance in a Bowline needs to re-opened?
What are your comments re the role of a 'collar' in a Bowline?
NOTE: And yes, you may well jokingly make remarks about Xarax... but, I think he deserves a lot of respect on account of the many contributions he made in this IGKT forum. He broke a lot of new ground and introduced many interesting new ideas and ways of thinking about knots. He did provide sound reasoning in support of the collar and its role in a Bowline. Please don't summarily dismiss Xarax!

I concur with you re the role of the 'nipping loop' in all Bowlines. But I also concur with Xarax with the role of the collar!
Can both elements co-exist in a nice harmony?

...

knot rigger: I concur with most of what you write.
However, the commonsense test may not be entirely valid, nor indeed a foolproof test. I prefer the 'scientific method' - although it is tough to apply in the field of knotting. There must be a way to quantify this body of knowledge into a theory of everything!!

Look at #1033 (Carrick Loop) and Alan Lee's 'Lee Zep Bowline', and indeed the base ''Myrtle'.
These structures might not pass the 'commonsense' test - and this may be due to the perceived difficulty with identifying a 'collar'. For instance, are parallel legs an absolute requirement for a structure to be deemed a 'collar'?

Dan Lehman has posited that the fundamental essence of a Bowline is a 'nipping turn'.
Ashley never stated this - and as far as I can see - no other knot author (even Budworth) has stated such.
I agree with Dan - the 'nipping loop' is a key element - the absence of which automatically rules out a knot from being a Bowline.

What I have tried to do is take this one step further - and to fully define what we mean by 'nipping loop'.
I think we both agree that loaded or unloaded - the identity of a Bowline remains unchanged.
So its just a case of getting the language correct in defining what we mean by 'nipping loop'.

A nipping loop which requires both ends to be loaded

per knot rigger
This is a definition worthy of consideration...
Or
[ ] 'Nipping loop' - tension force will be present at both ends in the loaded state. The tension force is not required to be in equal proportion.
If load is not capable of producing a tension force at both ends - this disqualifies the structure from being a 'nipping loop'. An example of such is #1431 (Sheet Bend).

Mark G

[knotsaver]

let's consider an easy (?!) case...
please look at the attached picture and think about it...
(to be continued...)