Daniel Harris wrote:
Oh, I see, if you have non-plain text characters as substitutions then this will make the reply switch to rich text?
Your use of the terms "rich" and "plain" text is mistaken. Plain text these days means text without any information about formatting in it, such as the font name, font size, bold, italic, etc. When you send plain text, the recipient's machine determines what font gets used, etc. Rich text is text which does contain such formatting information, it is really html underneath, and the recipient should normally see the font which you have chosen.
Curly (smart) quotes and punctuation have nothing to do with that distinction. They don't make a text "rich" and converting "rich" to "plain" (or viewing the "plain" version) will not remove them. If you could confine your message to ASCII encoding, they would not be present, but email apps don't normally have any option for that. As mentioned by others, if they cause a problem, you have to turn off the substitution system which is producing them.