As I studied the origins of grammar in the 18c, I came to understand that most English grammars originated with classical (Greek and Latin) scholars, who attempted to shoehorn English into similar grammatical patterns (declensions, conjugations, mood, tense, voice, number, gender, etc.). One sticking point was to define what is a consonant, what a vowel. Their definition: a consonant sound is made using the lips and teeth and tongue; a vowel sound is made primarily with the breath. So, what to do with the aspirant "h"? The sound made when "h" begins a word is not made with the lips/tongue/teeth, but rather by a modification of the shaping of the breath when pronouncing the following vowel. So, "history" is primarily the sound that you make for the vowel "i", but slightly constricted to indicate the preceding "h" (that is, if you sound the "h" at all). Since "h" sometimes seems more like a vowel than a consonant, the choice of indefinite article must conform to that uncertainty: "a" before a stressed first syllable word beginning with "h" (an hisTORic occasion), but "an" before one unstressed (a HIStory), based on the belief that the "h" (breathy) sound is more pronounced when the syllable is stressed.

However, over time (as an ngram shows), "a historic" and "an historic" have gone in and out of usage for 300 years, so either one is now acceptable. Otherwise, in Standard English, "a" before consonant sounds, "an" before vowel sounds

Such choices were borne home to me when I had a student named "Yvonne". When I called on her, I pronounced her name "E-von"; she quickly corrected me: "My name is pronounced 'WY-von'." Which is, I take it, a Southern thing. Of course, I adhered to her preference. And I note: You might encounter an "E-von" or you might meet a "WY-von" in certain places, so choose your articles carefully.

Useful examination of this question in the field of linguistics: https://specgram.com/PsQ.XVI.4/06.pulju.indefinite.html

I'm proud and pleased to be obsessed with the minutiae of language.

As I studied the origins of grammar in the 18c, I came to understand that most English grammars originated with classical (Greek and Latin) scholars, who attempted to shoehorn English into similar grammatical patterns (declensions, conjugations, mood, tense, voice, number, gender, etc.). One sticking point was to define what is a consonant, what a vowel. Their definition: a consonant sound is made using the lips and teeth and tongue; a vowel sound is made primarily with the breath. So, what to do with the aspirant "h"? The sound made when "h" begins a word is not made with the lips/tongue/teeth, but rather by a modification of the shaping of the breath when pronouncing the following vowel. So, "history" is sounded in the same way as you sound the following vowel "i", but slightly constricted to indicate the preceding "h" (that is, if you sound the "h" at all). Since "h" sometimes seems more like a vowel than a consonant, the choice of indefinite article must conform to that uncertainty: "a" before a stressed first syllable word beginning with "h" (an hisTORic occasion), but "a" before one unstressed (a HIStory), based on the belief that the "h" (breathy) sound is more pronounced when the syllable is stressed.

However, over time (as an ngram shows), "a historic" and "an historic" have gone in and out of usage for 300 years, so either one is now acceptable. Otherwise, in Standard English, "a" before consonant sounds, "an" before vowel sounds

Such choices were borne home to me when I had a student named "Yvonne". When I called on her, I pronounced her name "E-von"; she quickly corrected me: "My name is pronounced 'WY-von'." Which is, I take it, a Southern thing. Of course, I adhered to her preference. And I note: You might encounter an "E-von" or you might meet a "WY-von" in certain places, so choose your articles carefully.

Useful examination of this question in the field of linguistics: https://specgram.com/PsQ.XVI.4/06.pulju.indefinite.html

I'm proud and pleased to be obsessed with the minutiae of language.