Siegel zero: Difference between revisions
improve siegel citation |
m Fixed redirect link |
||
| Line 13: | Line 13: | ||
where ''C'' is a function of ε for which the proof provides no explicit [[lower bound]] (see [[effective results in number theory]]). |
where ''C'' is a function of ε for which the proof provides no explicit [[lower bound]] (see [[effective results in number theory]]). |
||
The importance of the possible Siegel zeroes is seen in all known results on the zero-free regions of L-functions: they show a kind of 'indentation' near ''s'' = 1, while otherwise generally resembling that for the [[Riemann zeta function]] — that is, they are to the left of the line ''Re''(''s'') = 1, and asymptotic to it. Because of the [[analytic class number formula]], data on Siegel zeroes have a direct impact on the [[class number problem]], of giving lower bounds for [[class number (number theory)|class number]]s. This question goes back to [[C. F. Gauss]]. What Siegel showed was that such zeroes are of a particular type (namely, that they can occur only for χ a ''real'' character, which must be a [[Jacobi symbol]]); and, that for each modulus ''q'' there can be at most one such. This was by a 'twisting' argument, implicitly about the L-function of [[biquadratic field]]s. This in a sense isolated the Siegel zero as a special case of GRH (which would prove that it did not exist). In subsequent developments, however, detailed information on the Siegel zero has not shown it to be impossible. Work on the class number problem has instead been progressing by methods from [[Kurt Heegner]]'s work, from [[ |
The importance of the possible Siegel zeroes is seen in all known results on the zero-free regions of L-functions: they show a kind of 'indentation' near ''s'' = 1, while otherwise generally resembling that for the [[Riemann zeta function]] — that is, they are to the left of the line ''Re''(''s'') = 1, and asymptotic to it. Because of the [[analytic class number formula]], data on Siegel zeroes have a direct impact on the [[class number problem]], of giving lower bounds for [[class number (number theory)|class number]]s. This question goes back to [[C. F. Gauss]]. What Siegel showed was that such zeroes are of a particular type (namely, that they can occur only for χ a ''real'' character, which must be a [[Jacobi symbol]]); and, that for each modulus ''q'' there can be at most one such. This was by a 'twisting' argument, implicitly about the L-function of [[biquadratic field]]s. This in a sense isolated the Siegel zero as a special case of GRH (which would prove that it did not exist). In subsequent developments, however, detailed information on the Siegel zero has not shown it to be impossible. Work on the class number problem has instead been progressing by methods from [[Kurt Heegner]]'s work, from [[transcendental number theory]], and then [[Dorian Goldfeld]]'s work combined with the [[Gross-Zagier theorem]] on [[Heegner point]]s. |
||
==See also== |
==See also== |
||
Revision as of 14:01, 12 October 2015
In mathematics, more specifically in the field of analytic number theory, a Siegel zero, named after Carl Ludwig Siegel, is a type of potential counterexample to the generalized Riemann hypothesis, on the zeroes of Dirichlet L-function.
There are hypothetical values s of a complex variable, very near (in a quantifiable sense) to 1, such that
- L(s,χ) = 0
for a Dirichlet character χ of modulus q say. Important results on this type of zero of an L-function were obtained in the 1930s by Carl Ludwig Siegel, from whom they take their name (he was not the first to consider them, and they are sometimes called Landau–Siegel zeroes to acknowledge also the work of Edmund Landau).
The possibility of a Siegel zero in analytic terms leads to an ineffective estimate
- L(1,χ) > C(ε)q−ε
where C is a function of ε for which the proof provides no explicit lower bound (see effective results in number theory).
The importance of the possible Siegel zeroes is seen in all known results on the zero-free regions of L-functions: they show a kind of 'indentation' near s = 1, while otherwise generally resembling that for the Riemann zeta function — that is, they are to the left of the line Re(s) = 1, and asymptotic to it. Because of the analytic class number formula, data on Siegel zeroes have a direct impact on the class number problem, of giving lower bounds for class numbers. This question goes back to C. F. Gauss. What Siegel showed was that such zeroes are of a particular type (namely, that they can occur only for χ a real character, which must be a Jacobi symbol); and, that for each modulus q there can be at most one such. This was by a 'twisting' argument, implicitly about the L-function of biquadratic fields. This in a sense isolated the Siegel zero as a special case of GRH (which would prove that it did not exist). In subsequent developments, however, detailed information on the Siegel zero has not shown it to be impossible. Work on the class number problem has instead been progressing by methods from Kurt Heegner's work, from transcendental number theory, and then Dorian Goldfeld's work combined with the Gross-Zagier theorem on Heegner points.
See also
References
- Siegel, Carl Ludwig (1935). "Über die Classenzahl quadratischer Zahlkörper" [On the class numbers of quadratic fields]. Acta Arithmetica (in German). 1 (1): 83–86.