The Conundrum of Conjoiners: an Analysis of Navani's Airship

[+Hoiditthroughthegrapevine]

Interesting new info from Dawnshard on this: 

Spoiler

On 11/6/2020 at 6:24 AM, Pagerunner said:

We've got the "frame of reference" being a perception issue:

This is pretty much how I saw this working, the magic part of the conjoined fabrials is the creation of a shared cognitive framework that allows motion from one relative coordinate system to be translated to another relative coordinate system. The fact that knowledge of motion through space of one of the relative coordinate systems destroys this shared cognitive framework implies that the relative origins of the coordinate systems are established when the conjoiners are paired. 

If user A on a moving chull cart conjoins their spanreed with user B's spanreed which is static, user A is aware that the point in space that their spanreed was initially paired (the origin of their relative coordinate system) is moving away from them at the speed that their chull cart is travelling. It is not a fixed origin for their relative coordinate system so the cognitive reference frame breaks down. 

If user A were in an inertially dampened box with no windows on a moving chull cart (and thus had no sensation of motion) they would believe that that the origin for their conjoined spanreed was fixed and spanreed communication should work. 

On 11/9/2020 at 7:18 AM, Jofwu said:

My primary issue with this came down to the idea that you could make Reversers effectively irrelevant with such flimsy use of perception. 

There's the part in Dawnshard where Rushu states that reversers are created by how the gems are paired, more evidence for reversers just being symmetrically reversed conjoiners, and quite frankly irrelevant:

Quote

“Many people, however, aren’t aware that gemstones can be paired in such a way as to make their movements opposite one another. Traditionally we’ve used amethysts for this, but rubies work as well—and we have an excess of those from the ranches at the Shattered Plains." 

 

On 11/9/2020 at 7:18 AM, Jofwu said:

I mentioned a simple thought experiment: we put two spanreeds in adjacent rooms with their boards pointed opposite directions, start using them, then we lower the partition. Have we created a Reverser? I think this application of perception says not. The fabrials are objectively not Reversers, so I think when the partition is lowered, and SOMEONE perceives what's actually happening, they will no longer work as they did. I could live with that.

I don't agree, I think as long as both spanreed users perceive that their boards are setup in a similar fashion they would function like normal, I think that's the function of the ritual of aligning the spanreed board, it's conditioning the user's perception of their relative framework. I think that the spanreeds also have an innate orientation that has to be aligned. 

Here's another thought experiment. Two spanreed users are on adjoining circular platforms surrounded by curtains. Before the spanreed communication starts the curtains are raised and the operators can see that both boards are orientated in the same direction, and the curtains are lowered again. The circular platform on which the spanreed operator who is writing is slowly and imperceptibly rotated while they write until they've rotated 90 degrees with the spanreed conjoined the whole time. What happens when the curtains are raised with the spanreeds still conjoined, and both spanreed operators see that their orientations are no longer aligned? 

I posit that nothing would happen, because they were unaware of the incremental change in the relation of their relative frameworks, and once they are aware that there has been a change they can see that the orientation of the individual boards, and spanreeds have not changed in relation to their operators. 

On 11/9/2020 at 7:18 AM, Jofwu said:

It does leave some ambiguity though, which I don't love. Spanreeds have an obvious orientation for their usage. What if I put a Conjoiner hanging from the middle of a rope and have two groups play tug of war? Which way is forward and which backward?

Hmm, I think this an interesting one, but I think that this example ignores that conjoiners have an innate alignment that has be taken into account. If you attached a spanreed to the rope, it's easier to imagine how a conjoined spanreed laying on a distant table would move in relation to the rope. Likewise other conjoiners should also have innate orientations, and maybe this will allow conjoined gems to have their orientations (while shielded in aluminum) rotated along a different axiss so that forward motion for the Fourth Bridge could be accomplished by a conjoined weight dropping from Urithiru. 

 

I will agree there are definitely some details about how conjoiner work in Dawnshard that seem to break down. The prime example being Rysn's floating chair. It's x and z translations are decoupled from the hoisted anchor, then it is described as though it were floating (can be easily pushed). But if it really was floating and it's x and z motion were decoupled from the anchor, then it should likewise be decoupled from the ships inertial framework and it shouldn't be experiencing the x and z motion of the ship. Wind resistance should cause it to slow and it should slowly loose the matched velocities that it had prior to conjoining and start to drift in the direction opposite of motion of the Wandersail (plus the direction the wind is pushing it). 

The other thing that is kind of funky is when Rysn uses her chair fabrials in the cave by just putting them up in the air and turning them on. Most other mentions of using conjoiners require the offset of vertical potential energy through mechanical means (weights and pulleys), but this use is a bit like Baron Munchause pulling himself up by his own ponytail. Why not just embed y-axis constrained coinjoiners in solid rock and use those to maintain height and use other coinjoiners to ascend and descend (with the solid rock ones switched off).  You could also have two stair shaped blocks that you could conjoin with solid rock that you could alternately turn on and off and make, like Led Zeppelin said, a Stairway to Heaven. 

 

Edited November 10, 2020 by hoiditthroughthegrapevine

[Jofwu]

4 hours ago, hoiditthroughthegrapevine said:

Here's another thought experiment. Two spanreed users are on adjoining circular platforms surrounded by curtains. Before the spanreed communication starts the curtains are raised and the operators can see that both boards are orientated in the same direction, and the curtains are lowered again. The circular platform on which the spanreed operator who is writing is slowly and imperceptibly rotated while they write until they've rotated 90 degrees with the spanreed conjoined the whole time. What happens when the curtains are raised with the spanreeds still conjoined, and both spanreed operators see that their orientations are no longer aligned? 

I posit that nothing would happen, because they were unaware of the incremental change in the relation of their relative frameworks, and once they are aware that there has been a change they can see that the orientation of the individual boards, and spanreeds have not changed in relation to their operators.

This just doesn't work for me. I suggests the whole curtain thing isn't really even necessary, don't you think? Seems like you're arguing if they can just focus hard enough (?) then they have full mental control over their reference frame. I still feel very strongly that they would stop working like this if you saw them both at the same time because you wouldn't be able to rationalize that they're operating in separate reference frames.

Spoiler

Reverser's aren't totally irrelevant. There's an important point made that there's a specific process to make them, right?

If you just need a bit of mental practice to exercise mental control over your own reference frame then that's totally unnecessary. No need to jump through hoops to create Reversers. Just take your Conjoined gems and say "well, forward for this one is now a different direction" and boom. It's a Reverser.

And honestly, you get into even deeper trouble when you come to this new invention we see in chapter 19. Here they've invented a specific, revolutionary invention that allows them to redirect Conjoined gems. And it requires a somewhat elaborate fabrial. What you're saying seems to suggest to me that this whole thing was unnecessary. That they could have done the same thing by putting up curtains and turning one fabrial slowly (as a simple trick to help an operator mentally turn their reference frame).

That all feels very problematic to me.

[+Oltux72]

19 minutes ago, Jofwu said:

 

  Reveal hidden contents

Reverser's aren't totally irrelevant. There's an important point made that there's a specific process to make them, right?

And honestly, you get into even deeper trouble when you come to this new invention we see in chapter 19. Here they've invented a specific, revolutionary invention that allows them to redirect Conjoined gems. And it requires a somewhat elaborate fabrial. What you're saying seems to suggest to me that this whole thing was unnecessary. That they could have done the same thing by putting up curtains and turning one fabrial slowly (as a simple trick to help an operator mentally turn their reference frame).

That all feels very problematic to me.

It seems to me that you are making the assumption that the frame of reference can be changed after the cojoining is established. Much of your troubles derive from that.

[Jofwu]

2 hours ago, Oltux72 said:

It seems to me that you are making the assumption that the frame of reference can be changed after the cojoining is established. Much of your troubles derive from that.

Not just me. Hoiditthroughthegrapevine's example that I was responding to was explicitly about two spanreeds being conjoined and the frame of reference being changed (without the user being aware) after the fact. I don't think I have any issue with the frame of reference being established and fixed at the time of conjoinment, in and of itself. I don't see how that solves any of the issues I have? Yes, it gives a framework to answer what would happen in these thought experiments. Those are only being introduced to conceptualize what's going on though.

What you seem to be describing is precisely what this new fabrial does. Determine an arbitrary frame of reference, then conjoin them gems to lock that frame of reference. If they could already do this, why do they need a new aluminium-based fabrial for it?

My problem is the fact that the frame of reference is so arbitrary. Adding the notion that it gets "locked" at conjoinment doesn't make it less arbitrary.

[+Oltux72]

26 minutes ago, Jofwu said:

Not just me. Hoiditthroughthegrapevine's example that I was responding to was explicitly about two spanreeds being conjoined and the frame of reference being changed (without the user being aware) after the fact.

Yes. But the issue goes away if you stop assuming that the frame of reference can be changed. I am afraid the idea that while the frame of reference is arbitrary, the idea of motion would also be arbitrary is false. Given a frame of reference there is an objective reality in distances and motion.

26 minutes ago, Jofwu said:

I don't think I have any issue with the frame of reference being established and fixed at the time of conjoinment, in and of itself. I don't see how that solves any of the issues I have? Yes, it gives a framework to answer what would happen in these thought experiments. Those are only being introduced to conceptualize what's going on though.

What other questions remain? Again, you just need to drop the idea that the cojoined objects would change distance unless a force is used.

26 minutes ago, Jofwu said:

What you seem to be describing is precisely what this new fabrial does. Determine an arbitrary frame of reference, then conjoin them gems to lock that frame of reference. If they could already do this, why do they need a new aluminium-based fabrial for it?

Because they were constrained by gravity. Up is up and down is down.

26 minutes ago, Jofwu said:

My problem is the fact that the frame of reference is so arbitrary. Adding the notion that it gets "locked" at conjoinment doesn't make it less arbitrary.

How else could that be? Why Roshar and not the local star? And it seems to me that there are rules. Gravity is still valid.

[Gilphon]

Come to think it, there's a pretty simple answer for why they needed aluminium: If they ever fully de-coupled the Fourth Bridge from the lattice at Urithiru, gravity is going to do bad things to the ship. The best scenario is 'a whole lot of force starts pushing the lattice at the Shattered Plains into the ground, making it much harder to move and possibly damaging it'. But you can't keep it fully conjoined either, because then it's going to either grind itself into the mountainside or fly away from the tower.

[+Oltux72]

3 hours ago, Gilphon said:

Come to think it, there's a pretty simple answer for why they needed aluminium: If they ever fully de-coupled the Fourth Bridge from the lattice at Urithiru, gravity is going to do bad things to the ship. The best scenario is 'a whole lot of force starts pushing the lattice at the Shattered Plains into the ground, making it much harder to move and possibly damaging it'. But you can't keep it fully conjoined either, because then it's going to either grind itself into the mountainside or fly away from the tower.

You could solve that by a careful cycling, provided the chull cart can take the full weight of Fourth Bridge.
That is, you do the following cycle at the end of a trip of the chulls: stop - switch on Urithiru - switch of Chull cart - turn the Chulls - switch on Chull cart - switch off Urithiru - restart

Even if the directions were cardinal, you could solve this by having a second team of chulls.

(Dawnshard)

Spoiler

What we learned there is that you can switch off the chull cart and Fourth Bridge will glide and could even, in theory, if they could signal Urithiru, rise or sink. Nor do the chulls ever have to bear Fourth Bridge's weight.

The open question is the frame of reference. Spanreeds are very instructive in that regard. They do tell us that cojoiners can use a rotating frame of reference. In fact, it is likely that that is unavoidable on a planetary surface.
The other question is rotation.

[Jofwu]

@Oltux72 I think you're missing the point of what's bothering me here.

We see in chapter 19 that Navani is very excited about an invention which does precisely what you're saying they could do all along. They've invented a device (which needs a new application involving aluminum) for her to change directions on the fly.

Point north, conjoin, get pulled north. Disjoin. Point east, conjoin, get pulled east. All with the pair being pulled the same way each time.

I think you're saying they could do that all along, assuming they have the mental control to assert whatever frame of reference they want. Unless you just think that level of control over the frame of reference just isn't possible [for a typical user]? I'm skeptical that's the case because its too easy to make things fuzzy. You could blindfold yourself so you don't know which way the pair is oriented, point whatever direction you want, be told "the pair is pointed the same way" even if it isn't, then conjoin your gem and get pulled that way. So what, Navani's new invention just skips the need to be blindfolded and told a lie? Where does one draw the line on how arbitrary the reference frames are and how much mental control one can exert on them?

Your comments about vertical direction being different because of gravity are irrelevant to this? Same for the idea of conjoinment locking the frames. My idea that the reference frames could "collapse" (after being joined) when the user realizes the pair is oriented differently was just an idea to solve the my issues, to a small degree. Because it implies mental control over the reference frames is limited and it means Navani's new device was helpful after all.

[+Hoiditthroughthegrapevine]

On 11/10/2020 at 11:40 AM, Jofwu said:

This just doesn't work for me. I suggests the whole curtain thing isn't really even necessary, don't you think? Seems like you're arguing if they can just focus hard enough (?) then they have full mental control over their reference frame.

Nope, I'm arguing that the magic portion of conjoiners handles the specifics of translation between the two relative coordinate systems, and that conscious knowledge that one of the relative coordinate systems is moving can break the cognitive framework that accomplishes this translation. 

The absolute motions of each relative framework have to be screened out as part of the magical translation of motion between relative coordinate systems. Like in this example where you have a spanreed setup on the top floor of Urithiru that is communicating with a spanreed at sea level in Thaylen city (this has happened in the OB so we know this works), if the y axis for both spanreeds is normal to gravity, the entire spanreed setup on top of Urithiru moves substantially more along the x and z axes due to the rotation of Roshar than the sea level Thaylen City spanreed.

Let's say for the purposes of simplifying the calculation that both the Urithiru spanreed (A) and the Thaylen city spanreed (B) are setup so that their x axes are aligned with the rotation of Roshar. The plateau on which Urithiru is located is above the highstorm and everstorm storm walls so let's use, as a rough estimate, a height for the plateau of 6 km (which is slightly lower than the height of Denali, the 3rd tallest mountain on earth), and the highest floor in Urithiru is 818.4 m (2,685 ft according to the coppermind) higher than the base of the tower. Roshar has a radius of 5,663 km and a rotational period of 20 rosharan hours, so spanreed B in Thaylen city at sea level would travel an arc of 1,780 km in the x direction in one rosharan hour. Spanreed A, at height of 6.8 km above sea level would travel an arc of 1782 km in the x direction in one rosharan hour. In one Rosharan hour spanreed A moves 2 more km in the x direction than spanreed B!

Using your calculation from your blog, 1 Rosharan hour =. 964 earth hours, so spanreed A would move 1.92 km more in an earth hour than spanreed B. That would mean that something has to be correcting for a displacement difference of. 54 m/s in the x axis for spanreed communication between the tallest point in Roshar to the lowest point on Roshar to work. It would definitely be noticeable if the spanreed pen was drifting 1.8 ft along the x axis every second. 

The answer is, I think, that each relative reference frame is like a box, and a cognitive framework is necessary to link the two relative reference frames together. The cognitive framework is made of the two conjoined spren (responsible for movement within each respective relative reference frame) the spanreed operators (responsible for defining the relative reference frame (this is done largely subconsciously)) , and the stormlight investiture which combines all of these together. 

There is this telling bit from Dawnshard, which I think directly confirms this view,

Spoiler

where Huio is describing the effects of aluminum on the spren trapped within a conjoiner and its ability to sense his thoughts as well:

Quote

“Lopen,” Huio said in Herdazian. “This aluminum has fascinating properties; I believe the captive spren are reacting to its presence, almost like prey react to a predator. When I touch this foil to the stone, they push to the other side of their confines. I hypothesize that the aluminum interferes with their ability to sense not only my thoughts of them, but the thoughts of their conjoined half.”

This quote implies that there is some necessary feedback between the spren and the agent using the fabrial housing the spren. I believe this feedback is the part of the cognitive framework that allows the spren to understand their relative reference frame in terms of their innate orientation. 

 

On 11/10/2020 at 11:40 AM, Jofwu said:

And honestly, you get into even deeper trouble when you come to this new invention we see in chapter 19. Here they've invented a specific, revolutionary invention that allows them to redirect Conjoined gems. And it requires a somewhat elaborate fabrial. What you're saying seems to suggest to me that this whole thing was unnecessary. That they could have done the same thing by putting up curtains and turning one fabrial slowly (as a simple trick to help an operator mentally turn their reference frame).

No, this is different, the curtain method relies on the operator to be unaware that their relative reference frame has been shifted. I believe all conjoiners have an innate orientation, the aluminum in Navani's hand pulling fabrial is shielding the operator's portion of the cognitive framework that defines the relative reference frame in respect to all axes other than the one that has been isolated to translate force across. Then the 1 dimensional translation would be based solely on the innate orientation of the fabrials. Positive force applied along the innate x axis of one conjoiners translates to positive force along the innate x axis of the other. So in a lot of ways, as you would expect with aluminum's ability to block certain aspects of the magic, the system is much simpler. 

Edited November 11, 2020 by hoiditthroughthegrapevine

[Jofwu]

5 hours ago, hoiditthroughthegrapevine said:

The answer is, I think, that each relative reference frame is like a box, and a cognitive framework is necessary to link the two relative reference frames together.

I don't disagree with any of this and don't see how it is releavant to my frustration, so I must not be communicating my point well.

I agree with your interpretation of how the new device works, though then we just fall back on the oddity of what this "innate orientation" is. It still must be a somewhat arbitrary reference frame, defined by the collective cognition of everyone on Roshar I guess? (because they still work relative to the rotating planet and all that jazz.)

I'm curious what you think would happen if you did the curtain experiment and then pull the curtain down? You'd agree with Oltux that the get locked at conjoinmemt and persist even after the operator is aware that their frame of reference turned?

[+Oltux72]

10 hours ago, Jofwu said:

@Oltux72 I think you're missing the point of what's bothering me here.

OK, that makes it clearer.

10 hours ago, Jofwu said:

We see in chapter 19 that Navani is very excited about an invention which does precisely what you're saying they could do all along. They've invented a device (which needs a new application involving aluminum) for her to change directions on the fly.

Point north, conjoin, get pulled north. Disjoin. Point east, conjoin, get pulled east. All with the pair being pulled the same way each time.

Yes. I think the break through is that she can now invert up and down at will and, most important, use downward and upward motion to power lateral movement.

Yet spanreeds make it pretty clear that lateral movements at least can be exchanged.

10 hours ago, Jofwu said:

I think you're saying they could do that all along, assuming they have the mental control to assert whatever frame of reference they want.

Not necessarily. By the data we have we can only tell that the mental frame of reference matters, not that every frame is (meta)physically possible. In particular, it seems that gravity is a force that cannot be mentally overcome.

10 hours ago, Jofwu said:

Unless you just think that level of control over the frame of reference just isn't possible [for a typical user]? I'm skeptical that's the case because its too easy to make things fuzzy. You could blindfold yourself so you don't know which way the pair is oriented, point whatever direction you want, be told "the pair is pointed the same way" even if it isn't, then conjoin your gem and get pulled that way. So what, Navani's new invention just skips the need to be blindfolded and told a lie? Where does one draw the line on how arbitrary the reference frames are and how much mental control one can exert on them?

Good questions I have no answer for. Yet, spanreeds. They care only about levelling them.

10 hours ago, Jofwu said:

Your comments about vertical direction being different because of gravity are irrelevant to this? Same for the idea of conjoinment locking the frames. My idea that the reference frames could "collapse" (after being joined) when the user realizes the pair is oriented differently was just an idea to solve the my issues, to a small degree.

It is possible that that can happen. Yet, it opens another can of worms. Whose perception would break it? If I looked at it, while you do not and you paired them, would
 it break the lock?

10 hours ago, Jofwu said:

Because it implies mental control over the reference frames is limited and it means Navani's new device was helpful after all.

Yes, it does.

10 hours ago, hoiditthroughthegrapevine said:

Nope, I'm arguing that the magic portion of conjoiners handles the specifics of translation between the two relative coordinate systems, and that conscious knowledge that one of the relative coordinate systems is moving can break the cognitive framework that accomplishes this translation.

Well, no. A frame of reference with both cojoiners at rest just need not exist. If you are in a carriage going towards me, there is no frame of reference where you are at rest to me. While we are standing on Roshar there is (an observer at the center of the planet rotating with it). But to an observer on a moon, the one of us further south is slower.

10 hours ago, hoiditthroughthegrapevine said:

The absolute motions of each relative framework have to be screened out as part of the magical translation of motion between relative coordinate systems. Like in this example where you have a spanreed setup on the top floor of Urithiru that is communicating with a spanreed at sea level in Thaylen city (this has happened in the OB so we know this works), if the y axis for both spanreeds is normal to gravity, the entire spanreed setup on top of Urithiru moves substantially more along the x and z axes due to the rotation of Roshar than the sea level Thaylen City spanreed.

If you chose a frame of reference not rotating with Roshar.

5 hours ago, Jofwu said:

I don't disagree with any of this and don't see how it is releavant to my frustration, so I must not be communicating my point well.

I must suggest that we are mixing up things.

1. There is a frame of reference issue
   (Dawnshard)
   Spoiler

   If you are on a ship you can use a frame centered on the ship then the bow and stern swaying up and down at different times, will not mean movement. To an observer outside the ship, they are moving.

   This frame is chosen at least partially on perception. Details are unclear.

2. There is a question of what direction is what direction. That is basically a different question. If I pull my spanreed towards an edge of the board, into which direction does that pull your spanreed? Is that determined by the frame of reference? That seems unlikely, in fact basically disproved by spanreeds.

 

[+Hoiditthroughthegrapevine]

15 hours ago, Jofwu said:

I agree with your interpretation of how the new device works, though then we just fall back on the oddity of what this "innate orientation" is. It still must be a somewhat arbitrary reference frame, defined by the collective cognition of everyone on Roshar I guess? (because they still work relative to the rotating planet and all that jazz.)

Not necessarily, I believe that we really have a case of nested reference frames.

The spren is a cognitive entity in and of itself. I believe that the innate orientation of the conjoiner ruby is set at the time of it's creation, it has a front and a back and a top and a bottom. This is because the piece of spren trapped in the ruby has it's own local understanding of 3 dimensional space, along the lines of this is in front of me, this is above me, this is to left of me, etc. When the conjoiner is used, the operator of the spanreed likewise has a cognitive level understanding of their local 3 dimensional space and further knows where the ruby housing the spren is in relation to their local frame of reference (even though this is most likely a subconscious and intuitive understanding). The magic I believe comes in bridging these two local coordinate systems and translating the relative motion of the ruby to the paired conjoiner's local coordinate system.

 

15 hours ago, Jofwu said:

I'm curious what you think would happen if you did the curtain experiment and then pull the curtain down? You'd agree with Oltux that the get locked at conjoinmemt and persist even after the operator is aware that their frame of reference turned?

I think @Oltux72 brings up some good points, but I think I have a subtley different view of how this is working. 

I think in both instances of the curtain experiment (Case 1: the one where the spanreed operators are facing each other so that the apparent motions of each spanreed would seem to be reversed to an observer watching from the sidelines, and Case 2: the one where the spanreed writer is on an platform that rotates imperceptibly 90 degrees as they are writing while the spanreed that is receiving the communication is on a stationary platform) that there would be no jumping of spanreeds or any attempted after the fact correction when the curtain is raised (or collapse as you called it) . I think in both cases the spanreeds would work while the curtain is down, and I think both would work as well after the curtain was raised in both cases, as long as the spanreed operators see that the conjoined spanreed's relative orientation matches their own spanreed's relative orientation and that their is no motion that affects only 1 reference frame.  

If in Case 2 though, if the curtains were raised while the platform was still rotating and the spanreed operator on the rotating platform was still writing, I think that this would cause a breakdown of the cognitive reference frame and that the receiving conjoined spanreed would behave erratically (as it's stated in text happens when someone attempts to use spanreeds on a moving vessel). 

I think the explanation for this is simple, conjoining spanreeds can only accommodate 2 levels of nested reference frames, because there are only two cognitive entities involved in the operation of the spanreed, each with their own subjective reference frame. 

I think this explanation for how conjoiners work is (Dawnshard Spoiler)

Spoiler

Pretty close to explicitly stated by Rushu when she is telling Rysn about conjoiner fabrials. 

Quote

“But spanreeds don’t work on ships that are sailing,” Rysn said, moving her hoop around and watching its other half respond. “Why does this?” “Well, the problem with spanreeds is that a ship is always rocking and moving,” Rushu explained. “If you’re holding one in your lap and writing, you might feel that you’re being steady—but since the entire ship is moving so much, the reed on the other side will be wobbling around and surging up and down. We’ve found that there is simply too much motion to properly use spanreeds this way. However, right now both of these hoops are on the same ship. They rock together, move together.” “But when the ship goes down,” Rysn said, pointing at the other ring, “shouldn’t it go up?” “Yes, theoretically,” Rushu said. “But it doesn’t. Only your movements affect it. We believe this has to do with the frame of reference, as applied to the person moving the hoop. Spren, it should be noted, have a curious relationship to our perception of them and their motions. You see both of these hoops in the same frame of reference, so they act together. It’s why the motion and curve of the planet don’t influence spanreeds. “It’s proven impossible for someone on a ship with a spanreed to see themselves in the same frame of reference as the person receiving the communication. Perhaps there is a way to train ourselves, but no one has discovered it. Indeed, even the size of the ship can influence these things. If you tried this experiment on a rowboat, for example, the results could be different.”

 

 

Edited November 12, 2020 by hoiditthroughthegrapevine
Doh, had the Dawnshard Spoiler in a quote not a Spoiler tag, this has been fixed.