Your citing standards are not just defined by the standard procedures of your journal or the literacy level of your audience, but also by whether the source you cite argues the point, assumes the point, entertains the point as one of many possibilities, or even whether the source exists solely to address the point you are making.
If you wish to quote Douglas Adams to argue that life is meaningless, very little summary is required. As we see in Douglas Adams's parable of the supercomputer and the meaning of life, "meaning" is not a property of life, and requesting a calculation of "meaning" results in meaningless answers, such as "42".
This is a summary not of a book or a story, but of a paragraph of a book. It is a tiny story within a larger story. Those who have read it will appreciate a chance to chuckle, but everyone will see the connection whether they've read it or not, and there will be no need to summarize further.
But if you need to challenge the straightforward interpretation, for example if you're disagreeing with another author who cited the source, more detail is required.
- "The anti-meaning school of thought emphasizes the results of Douglas Adams's supercomputer", - summarize opposition,
- "which was asked to calculate the meaning of life, and which concluded that life is meaningless." - summarize opposition's summary,
- "But in "Hitchhiker's Guide to the Galaxy" it was understood that meaning can exist even when it cannot be calculated, which is why the story called for an organic invention (earth) to discover the meaning of life" - re-interpret the source,
- "indeed, what is "meaning" if not "value which defies quantification"?" - summarize your interpretation of the source.
Your style and level of detail should walk a delicate balance between risk and rigor.
If you publish a correlation between lung cancer and smoking, hardly anyone will care what the more rigorous points of your argument are; the established topic lacks the risk to be interesting. You won't get published unless your writing breathes new fascination into the old question.
Don't worry about being fascinating or risky if you're presenting a correlation between chocolate consumption and IQ though; a fifth of your audience will label you a clown or classist before they read your work, so everyone will have plenty of energy to read every sentence and citation you write as they scan for technical mistakes. Just focus on accuracy and rigor, that'll be your bottleneck to even being published.
So my final answer is, if you justify too much, you'll be boring. If you justify too little, you'll be stupid.