You're correct in pointing out the potential confusion in using an affected son with an unaffected mother to determine if a recessive condition is not sex-linked from a pedigree alone. This method does have limitations and may not provide conclusive evidence regarding the linkage of a recessive allele to the sex chromosomes.
The reason it's challenging to use an affected son and an unaffected mother to rule out sex-linkage is because, as you mentioned, the mother could carry a recessive allele on one of her X chromosomes and still appear unaffected if she has a dominant allele on her other X chromosome. In this scenario, the affected son could inherit the recessive allele from the mother's X chromosome with the recessive allele and still be affected by the condition.
To definitively determine whether a recessive condition is sex-linked, you would ideally need a pedigree that shows the presence of the condition in both males and females, as sex-linked conditions typically show a specific pattern of inheritance. In the case of a recessive sex-linked condition, you would expect to see affected males in each generation (as they only have one X chromosome) and potentially carrier females (who have one affected X and one normal X) who may have unaffected sons.
If a pedigree consistently shows affected males and unaffected females (with affected sons), it suggests that the condition is likely not sex-linked but autosomal recessive. However, if you don't have a pedigree that demonstrates this pattern, it can be challenging to definitively conclude the linkage of a recessive condition to the sex chromosomes based solely on the affected son and unaffected mother scenario.
In biology, it's important to recognize the limitations of the available information and use multiple lines of evidence, such as genetic testing or more comprehensive pedigree analysis, when determining the mode of inheritance and linkage of a specific genetic trait.