3
$\begingroup$

Let $R$ be a ring (commutative, with unit), $M$ an $R$-module, and $x_1, \dotsc, x_n \in R$. Consider the following two conditions:

  1. For all $i$, the homomorphism $$\frac{M}{(x_1, \dotsc, x_{i-1})M} \stackrel{x_i}{\longrightarrow} \frac{M}{(x_1, \dotsc, x_{i-1})M}$$ is injective.
  2. $$\frac{M}{(x_1, \dotsc, x_n) M} \neq 0.$$

Taken together, these conditions give the definition for the $x_i$ to form an $M$-regular sequence. However, it is sometimes useful to consider Condition 1 by itself. For instance, this comes up in Eisenbud, Commutative Algebra, Exercise 6.7 (page 174 in my copy). Eisenbud sort of, but not really, calls such a sequence an "(almost) regular sequence."

Is there a standard term (or, for that matter, any reasonable term with a not-too-obscure reference) for a sequence of elements of $R$ (possibly contained in some fixed ideal of $R$, especially if $R$ is local) that satisfy Condition 1, but not necessarily Condition 2?

(Other related notions of "not-quite-regular sequence" would also be of interest.)

$\endgroup$
1

1 Answer 1

7
$\begingroup$

A reasonable term with a not-too-obscure reference: that is called a weak $M$-sequence by Bruns and Herzog in Cohen-Macaulay Rings (p. 3 in the edition I have).

$\endgroup$
1
  • 1
    $\begingroup$ This is certainly a valid reference (+1). The trouble is, based on google searches, that "weak M-sequence" is more commonly used to refer to a sequence in which condition (i) is replaced by the condition that everything annihilating $x_i$ annihilates your entire ideal (e.g., your maximal ideal, if $R$ is local). $\endgroup$ Jul 31, 2011 at 2:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.