This is a variant on a well-known problem in mathematical logic, discussed in terms of "recursion", "self-referentiation", and "non-terminating processes". It is discussed by Peter Suber in "The Paradox of Self-Amendment in American Constitutional Law", Stanford Literature Review, 7, 1-2 (Spring-Fall 1990) 53-78.
But to understand it, we first have to ask, "What is a constitution"? One way to approach the answer is to ask, "If a constitution of government is accurately translated into another language, so that a person adhering to it would make the same decisions, is the translation the same constitution, or a different one?"
The most productive way to answer the second question is to say that it is the same constitution. That is, it is the meanings, not the language-bound text, that is the real constitution. The text is only evidence of that constitution, just as a copy of it would be.
If it is meanings, then how might we represent them that is rigorous and not bound to any particular natural language? The answer, with some simplification, is that they translate into a system of deontic statements, in a normative extension of the first order predicate calculus (FOPC), that express a set of constraints on human behavior in the polity for which it is the constitution.
We will establish it as a premise that to be a constitution, it has to constrain human behavior in a way that is nontrivial, and that enables people to cooperate in protecting their fundamental rights.
If there are no constraints on how a constitution may be amended, then a commitment to a constitution in all the ways it could be amended would contradict this premise, because it would be a commitment not to a set of constraints, but to an unconstrained process that could terminate only if the amendment clause were effectively deleted, and that could morph into something that would not be a constitution at all.
Therefore, the amendment process must be constrained, if not by its own terms, then by principles of sound constitutional design that are superior to any written constitution of government. One way to conceive of these constraints is to posit that there are superior constitutions of nature, society, and the state, that form a hierarchy of authority, superior to the written constitution of government. These concepts are discussed in "Social Contract and Constitutional Republics".
This leads to the question, "Is the Constitution for the United States constitutional?" That is, is it compliant with these superior constitutions of nature, society, and the state? The answer is yes, if we examine the language of the document and look for ways it can be reasonably construed that are compliant. Such a construction exists, so an oath to it understood in that way would be an honorable act. We can recognize that the language contains ambiguities that temporized on violations of fundamental constitutional principles for a time after ratification. Tolerating violative practices is not a flaw in the Constitution itself, only in the application of it. We can say that the Constitution used to be unconstitutional as applied, in certain ways, but may no longer be, in those ways. However, it may be in other ways. One takes an oath, however, to the constitution and not to practices.
So to answer the original question, one can only honorably take an oath to a constitution and its amendments that remain constrained by the superior unwritten constitutions of nature, society, and the state, which contain the underlying principles of sound constitutional design.
Interestingly, with the practice of binding stare decisis we have a practice that is fundamentally inconsistent with the Constitution, and any constitution. A commitment to the present Constitution, properly construed, is not dishonorable. But a commitment to binding stare decisis is, in all the ways that have given rise to the initial question.
As for being "democratic", if that is understood as a political order in which majorities can always decide in support of any positive public action, then that is in essential conflict with our constraint that the constitution protect fundamental rights. It is one thing to prevent positive action without majority consent, but quite another thing to enable positive action with only majority support. In a well-designed democratic system, majority support should be necessary, but not sufficient, for positive public action, and structures and procedures that enable minorities or even individuals to block action are not only sound design, but necessary.
Interestingly, with the practice of binding stare decisis we have a practice that is fundamentally inconsistent with the Constitution, and any constitution. A commitment to the present Constitution, properly construed, is not dishonorable. But a commitment to binding stare decisis is, in all the ways that have given rise to the initial question.
As for being "democratic", if that is understood as a political order in which majorities can always decide in support of any positive public action, then that is in essential conflict with our constraint that the constitution protect fundamental rights. It is one thing to prevent positive action without majority consent, but quite another thing to enable positive action with only majority support. In a well-designed democratic system, majority support should be necessary, but not sufficient, for positive public action, and structures and procedures that enable minorities or even individuals to block action are not only sound design, but necessary.