Muslim scholars grade hadiths into a number of categories, the most popular of these categories are called sahīh (meaning reliable) and daʿīf (meaning weak). Obviously this does not exhaust the grading of categories for there are many more, but we just want to get an idea across at this point.  

So a hadith is composed of a chain of transmitters. A chain of transmission in a hadith works like this. Imagine person A hears that the Prophet said X from person B. Person B says he heard the hadith from person C, and person D says he heard it from a companion of the Prophet (s) who was present when the Messenger of Allah (s) stated X.  

 So now if we go back to our question, how do we know these people are truthful? Early in Islam’s history, Muslim scholars painstakingly researched about narrators. Often enough, they were contemporaries of these narrators. Through in depth study of these individuals, Muslim scholars would compile biographical dictionaries outlining their assessment of hadith narrators.  

 If all the narrators were reliable (both in their moral integrity and their memory) and the chain of transmission was unbroken, then the hadith would be labeled as sahīh, meaning authentic or reliable. If the chain of transmission contained people of dubious character, known to be dishonest, or at the very least, someone with a bad memory, then the hadith would be considered daʿīf or weak.

 By no means did this mean that the hadith was reliable from a historical perspective. All it meant was that Muslims were giving the hadith the benefit of the doubt and a probable chance of being an authentic utterance of the Prophet (s) or his Ahl al-Bayt (as).