I was taking a Classical Japanese course where the professor mentioned that many considered the Tale of Genji to be the world's first novel, and vague google search results seem to agree with this statement. But I was also taking an Ancient Novels course and most of the novels covered in that course predates Murasaki Shikibu (i.e., works by Longus, Heliodorus, Chariton, etc.). In Collected Ancient Greek Novels edited by B.P. Reardon, Reardon states, in his introduction to Callirhoe, that the work, being from the 2nd century A.D., is considered "the first European novel." So I was wondering why the Tale of Genji from the early 11th century is often touted as the first novel ever?
First, it's important to note that the definition of the word "novel" changes depending on who you talk to, even among the experts. Please note that most academics would cite The Tale of Genji as the first—and there is a great deal of scholarly consensus worldwide on this point, for reasons I'll get to in a moment. To better explain your question, I'm going to start with broader definitions of the term, and then narrow it down until we get to the reasoning behind Genji.
The broadest definitions of the word novel all include the following:
Before discussing verse, I want to note that between the 17th through the 19th centuries, many scholars considered "verse epics" to be novels, eg. The Canterbury Tales by Chaucer. This isn't a thing anymore, though verse epics are considered a novel precursor.
Verse was, for much of human history, the preferred form of entertainment writing as far as we know. Most of what was written didn't survive, so all we can do is speculate. In China, for instance, written records data back to the Shang Dynasty, which was 1800-1050 BC, but Chinese literary tradition didn't really kick off until the Tang Dynasty, which was roughly between 600 and 900 AD. I feel like I should note the distinction between Chinese literary writing and other influential writings—for instance, philosophies like Daoism and Confucianism, which emerged hundreds of years prior to the Tang, would not be relevant here.
In the West, prose was somewhat more popular than in China—or perhaps simply more prose works survived. In the second century BC, a man named Aristeides wrote The Milesian Tale, which was essentially a collection of short stories regarding a central protagonist. Note that if a person was using only the above, broad-strokes definition of a novel, Aristeides' stories, assembled into one collection, would apply.
However, the literary world has long distinguished between short collections and a single, longer story, which makes the assumption that The Milesian Tale might be a novel to be short-sighted. For comparison, One Thousand and One Nights emerged as a loose collection of stories shortly after 700 AD, but it has never been considered a novel. Collections of fairy tales or regional folklore are also not considered a cohesive narrative.
It is worth noting that The Milesian Tale was written with a central protagonist in mind, but because having multiple protagonists doesn't disqualify a book from being considered a novel (think Martin's A Song of Ice and Fire series), then having multiple protagonists shouldn't disqualify One Thousand and One Nights, either. Which means to be considered a novel, the narrative must remain unbroken. In other words, episodic or picaresque works don't count. And when scholars do approach writings done in antiquity, they are forced to focus more on plot and historical context.
Tomas Hägg is noted for tracking Greek romances in his book The Novel in Antiquity. He notes that The Milesian Tale, like other surviving texts of that time period, is sensationalist entertainment told using heightened language, and he alludes to the distance between the reader and the text itself. What he essentially does in his book is redefines what a novel is in order to secure a definition that fits writings in antiquity, and then discusses old texts from that vantage point. This is an important point, because critics who claim that the original novel comes from that time period are using those broader—or outright altered—definitions than what is commonly accepted.
You mentioned Callirhoe in your question. That writing is the oldest surviving prose romance, and features an unbroken narrative that appears to meet the criteria we've set so far. So how is that different?
The answer, which also provides the next filter layer, is that Callirhoe (and other ancient prose fiction works) lacks two other things:
Note that the above two bullet points are related. The difference between epic romance and a novel is time spent in the headspace of the protagonist. Any time you have vast, sweeping adventures that lacks proper character growth and introspection, you are likely reading an epic romance. Martin might write about dragons, but no one can deny the character growth of people like Jaime Lannister and Daenerys Targaryen, nor the skill with which he illustrates Cersei's mental deterioration through her POV chapters. Epic romance isn't concerned with humanizing the protagonist and letting readers step into their shoes. Instead, epic romance gives readers a long view of a heroic figure who is larger than life.
What differs Callirhoe from The Tale of Genji is just that—Murasaki Shikibu works tirelessly to investigate familial ties, civic duty, and feudal happenings as its characters age, change, adapt to new situations, and mature. The story is incredibly long (~1300 pages), the prose is consistent, and the narrative—while jumping forward in time—is highly consistent. Finally, and while many (including myself) wouldn't consider chapters an additional disqualifier, The Tale of Genji has 'em.
It is important to note that even when the idea behind what defines a "novel" is challenged, scholars tend to wind up defining Genji as a novel anyway. Scholar and East Asian literature and languages expert Dennis Washburn argues in his personal translation that classifying The Tale of Genji as a novel means applying a set of Western-specific standards, and Shikibu's work tends to transcend genre. Washburn describes the period in which the author wrote as fluid in terms of genre (indeed, Shikibu weaves in poetry and other forms into the text), and notes that crossing literary boundaries was accepted in that point in Japanese history. He follows up these points by noting that while perhaps the definition of the word novel is too Westernized, the book still meets contemporary expectations.
This makes sense, if you think about comparisons. Slaughterhouse-Five uses an unreliable narrator and time jumps to tell a story that is, at the end of the day, cohesive. Likewise, Chuck Palahniuk's Haunted begins every chapter with a poem. Both of those books would be considered novels. In other words, a novel can incorporate other genres and techniques as long as the narrative, and characterization, is uniform and distinct.
tl;dr
The Tale of Genji meets all the standards for what constitutes a novel. Earlier works, despite being long form prose, do not.
Sources:
Hägg, Tomas (1991). The Novel in Antiquity. University of California Press
Shikibu, Murasaki (2015). The Tale of Genji. (D. Washburn, Trans.). New York: W. W. Norton & Co. (Original work published before 1071 AD)