Rigidity of highly twisted plat diagrams
Nir Lazarovich, Yoav Moriah, Tali Pinsky, Jessica S. Purcell · 2025
Read Rigidity of highly twisted plat diagrams by Nir Lazarovich, Yoav Moriah, Tali Pinsky, Jessica S. Purcell (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-07303.txt.
Introduction
Nir Lazarovich, Yoav Moriah, Tali Pinsky, and Jessica Purcell wrote this math paper in two thousand twenty-five. It proves that a knot or link with a rich enough plat picture has only one such picture, up to plain turns.
A plat picture lays a knot out in a grid of twist bands. The claim needs the plat to be wide and tall enough, with each twist band holding at least a set count of crossings. Then no second plat picture of the same knot can differ in a real way. That makes the picture a canonical form for those knots. As such it sorts and names them. The result gives knot theory a firm grip on a broad class.
Catalog record
| CREATOR | Nir Lazarovich, Yoav Moriah, Tali Pinsky, Jessica S. Purcell |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 96 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-07303
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Rigidity of highly twisted plat diagrams Thanks: The authors would like to thank the Technion, Haifa for its support, and the Institute for Advanced Study (IAS), Princeton. Lazarovich was partially supported by the Israel Science Foundation (grant no. 1576/23). Purcell was partially supported by the Australian Research Council. Nir Lazarovich, Yoav Moriah, Tali Pinsky, Jessica S. Purcell Abstract . In this paper we prove that if a knot or link has a sufficiently complicated plat projection, then that plat projection is unique. More precisely, if a knot or link has a 2 m 2m -plat projection, where m m is at least four, and height at least two, and each twist region of the plat contains at least four crossings, then such a projection is unique up to obvious rotations. In particular, this projection gives a canonical form for such knots and links, and thus provides a classification of these links. 2020 Mathematics Subject Classification Primary 57K45 1. Introduction One way of studying knots in S^{3} is via their regular projections on 2 -spheres in S^{3} . Such projections are called knot diagrams . Deciding when two diagrams correspond to the same knot is a difficult problem, going all the way back to work of Tait in the 1870s. In 1926, K. Reidemeister proved that any two regular projections are equivalent by a sequence of Reidemeister moves ; see [ 26 ] . However, determining when two diagrams are equivalent by Reidemeister moves is also a very difficult problem, and remains an area of active research, for example see [ 16 , 5 ] . The earliest attempts to classify knots began in the 1870’s by Tait, using the crossing number of a knot as the classifying parameter. Recently, knots with up to 20 crossings have been classified, by Burton [ 4 ] and Thistlethwaite [ 30 ] . That is, some 150 years later, distinguishing diagrams using crossing number is possible only for knots with a small number of crossings. Another way around the problem of deciding when diagrams are equivalent would be to obtain “canonical” projections for knots. A first attempt to do so was due to Schubert in 1956 see [ 28 ] . All knots have 2m -plat projections for some m\in\mathbb{N} . Schubert successfully classified all 4 -plats, which are more commonly known as 2 -bridge knots and links. He showed that these knots are classified by a pair of rational numbers, and that any continued fraction expansion of these number correspond to a 4 -plat projection. For a discussion of this see [ 2 ] . In a somewhat different flavor Menasco and Thistlethwaite proved in 1993 that any two alternating projections of a knot \mathcal{K}\subset~S^{3} are equivalent by flype moves [ 20 ] . Flypes involve the existence of 4-punctured spheres in the diagram, and they can be detected in a finite number of steps. However, the sequence of such steps can be arbitrarily long. In this paper, we prove a uniqueness statement for diagrams of an infinite class of plats. We show that such plats have a canonical form which is unique, and can be read off the diagram immediately, without any need to consider equivalence relations such as Reidemeister moves or flypes. Theorem 1.1 . Let K^{\prime} and K be two 4 -highly twisted plats representing the same knot or link \mathcal{K}\subset S^{3} , so that each diagram has width greater than or equal to 4 , and odd height greater than or equal to 3 . Then K=K^{\prime} up to rotation in a vertical and/or a horizontal axis of the plats. Here, when we say two diagrams K^{\prime} and K represent the same knot or link, we mean that there exists an ambient isotopy \varphi\thinspace\colon(S^{3},K^{\prime})\to(S^{3},K) . For precise definitions of plats , and 4 -highly twisted , see Section 2 and references [ 3 ] and [ 13 ] . Theorem 1.1 shows that for knots and links satisfying the required conditions there is a canonical form, namely the 2m -plat projection. This canonical form has other nice features, as it gives information about incompressible surfaces in S^{3}\smallsetminus\mathcal{N}(\mathcal{K}) ; see [ 7 ] , [ 6 ] and [ 31 ] . In many cases, it gives information about the fundamental group of the knot space \pi_{1}(S^{3}\smallsetminus\mathcal{N}(\mathcal{K})) and its rank, that is the minimal number of generators of \pi_{1}(S^{3}\smallsetminus\mathcal{N}(\mathcal{K})) ; see [ 18 ] . It also gives information about the bridge number for \mathcal{K} , about Heegaard splittings of S^{3}\smallsetminus\mathcal{N}(\mathcal{K}) and about manifolds obtained by Dehn surgery on K ; see [ 18 ] and [ 31 ] . In addition, a knot or link \mathcal{K} with at least C crossings in each twist region, for appropriate C , is known to satisfy several nice geometric properties, regardless of having the plat projection required by Theorem 1.1 . For example, if C=3 , then K is hyperbolic [ 17 , 9 ] . If C=6 , then all Dehn fillings of K are hyperbolic [ 9 ] . Closed embedded essential surfaces are known to be high genus [ 1 ] . If C=7 , there are known explicit upper and lower bounds on the hyperbolic volume of K [ 8 ] . And if C=116 , then the shape of the cusp of \mathcal{K} is bounded [ 24 ] . A 2m -plat diagram comes with a family of horizontal bridge spheres , defined in Section 2 . Such a sphere separates the diagram into two trivial tangles. The braids associated with such trivial tangles were investigated by Hilden in 1975 [ 12 ] . Our results have applications to Hilden double cosets as in the following corollary. Corollary 8.5 . If b and b^{\prime} are two 4 -highly twisted words in \mathcal{X}_{2m} with the same Hilden double coset \mathcal{H}b\mathcal{H}=\mathcal{H}b^{\prime}\mathcal{H} , then b and b^{\prime} are the same up to a vertical and/or a horizontal rotation of angle \pi . We suspect that Theorem 1.1 can be improved, as follows. Conjecture 1.2 . The statement of Theorem 1.1 holds for knots and links which have plat presentations of width greater than or equal to 3 and are 3 -highly twisted. It is also likely that the requirement that all the coefficients are 3 -highly twisted can be weakened as well. For example, the proof of Theorem 1.1 uses a result on essential surfaces originally proved for highly twisted plats by Finkelstein and Moriah [ 7 , 6 ] , which was later shown by Wu [ 31 ] to remain true with fewer crossings in all but the outermost twist regions. We suspect Theorem 1.1 may generalise in this way as well. Going back to the question of crossing number, we ask the following: Question 1.3 . Do knot and link diagrams satisfying the conditions of Theorem 1.1 realize the crossing number of the associated links? For 2 -bridge knots the answer is negative since a 4-highly twisted diagram is not necessarily alternating and the crossing number is achieved by the alternating diagram [ 14 , 23 , 29 ] . 1.1. Organisation In Section 2 , we review definitions of plats. In Section 3 , we