The story of Titus Andronicus is fictional, not historical, unlike Shakespeare's other Roman plays, Julius Caesar, Antony and Cleopatra and Coriolanus, all of which are based on real historical events and people. Even the time in which Titus is set may not be based on a real historical period. According to the prose version of the play (see below), the events are "set in the time of Theodosius", who ruled from 379 to 395. On the other hand, the general setting appears to be what Clifford Huffman describes as "late-Imperial Christian Rome," possibly during the reign of Justinian I (527–565). Also favouring a later date, Grace Starry West argues, "the Rome of Titus Andronicus is Rome after Brutus, after Caesar, and after Ovid. We know it is a later Rome because the emperor is routinely called Caesar; because the characters are constantly alluding to Tarquin, Lucretia, and Brutus, suggesting that they learned about Brutus' new founding of Rome from the same literary sources we do, Livy and Plutarch." Others are less certain of a specific setting however. For example, Jonathan Bate has pointed out that the play begins with Titus returning from a successful ten-year campaign against the Goths, as if at the height of the Roman Empire, but ends with Goths invading Rome, as if at its death. Similarly, T.J.B. Spencer argues that "the play does not assume a political situation known to Roman history; it is, rather a summary of Roman politics. It is not so much that any particular set of political institutions is assumed in Titus, but rather that it includes all the political institutions that Rome ever had."
In his efforts to fashion general history into a specific fictional story, Shakespeare may have consulted the Gesta Romanorum, a well known thirteenth-century collection of tales, legends, myths and anecdotes in Latin, which took figures and events from history and spun fictional tales around them. In Shakespeare's own lifetime, a writer known for doing likewise was Matteo Bandello, who based his work on that of writers such as Giovanni Boccaccio and Geoffrey Chaucer, and who could have served as an indirect source for Shakespeare. So too could the first major English author to write in this style, William Painter, who borrowed from, amongst others, Herodotus, Plutarch, Aulus Gellius, Claudius Aelianus, Livy, Tacitus, Giovanni Battista Giraldi, and Bandello himself. The possibility that Shakespeare may have used the Gesta, Bandello and/or Painter is strengthened by the fact that he definitely consulted all three sources later in his career; the Gesta provided some of the details for King Lear, Bandello was the primary source for Twelfth Night, and Painter was used during the composition of All's Well That Ends Well.
However, it is also possible to determine more specific sources for the play. The primary source for the rape and mutilation of Lavinia, as well as Titus' subsequent revenge, is Ovid's Metamorphoses (c.AD 8), which is featured in the play itself when Lavinia uses it to help explain to Titus and Marcus what happened to her during the attack. In the sixth book of Metamorphoses, Ovid tells the story of the rape of Philomela, daughter of Pandion I, King of Athens. Despite ill omens, Philomela's sister, Procne, marries Tereus of Thrace and has a son for him, Itys. After five years in Thrace, Procne yearns to see her sister again, so she persuades Tereus to travel to Athens and accompany Philomela back to Thrace. Tereus does so, but he soon begins to lust after Philomela. When she refuses his advances, he drags her into a forest and rapes her. He then cuts out her tongue to prevent her telling anyone of the incident and returns to Procne, telling her that Philomela is dead. However, Philomela weaves a tapestry in which she names Tereus as her assailant, and has it sent to Procne. The sisters meet in the forest and together they plot their revenge. They kill Itys and cook his body in a pie, which Procne then serves to Tereus. During the meal, Philomela reveals herself, showing Itys' head to Tereus and telling him what they have done.
For the scene where Lavinia reveals her rapists by writing in the sand, Shakespeare may have used a story from the first book of Metamorphoses; the tale of the rape of Io by Zeus, where, to prevent her divulging the story, he turns her into a cow. Upon encountering her father, she attempts to tell him who she is, but is unable to do so until she thinks to scratch her name in the dirt using her hoof.
Titus' revenge may also have been influenced by Seneca's play Thyestes, written in the first century AD. In the mythology of Thyestes, which is the basis for Seneca's play, Thyestes, son of Pelops, King of Pisa, who, along with his brother Atreus, was exiled by Pelops for the murder of their half-brother, Chrysippus. They take up refuge in Mycenae, and soon ascend to co-inhabit the throne. However, each becomes jealous of the other, and Thyestes tricks Atreus into electing him as the sole king. Determined to re-attain the throne, Atreus enlists the aid of Zeus and Hermes, and has Thyestes banished from Mycenae. Atreus subsequently discovers that his wife, Aerope, had been having an affair with Thyestes, and he vows revenge. He asks Thyestes to return to Mycenae with his family, telling him that all past animosities are forgotten. However, when Thyestes returns, Atreus secretly kills Thyestes' sons, Pelopia and Aegisthus. He cuts off their hands and heads and cooks the rest of their bodies in a pie. At a reconciliatory feast, Atreus serves Thyestes the pie in which his sons have been baked. As Thyestes finishes his meal, Atreus produces the hands and heads, revealing to the horrified Thyestes what he has done.
Another specific source for the final scene is discernible when Titus asks Saturninus if a father should kill his daughter when she has been raped. This is a reference to the story of Verginia from Livy's Ab urbe condita (c.26 B.C.). Around 451 B.C., a decemvir of the Roman Republic, Appius Claudius Crassus begins to lust after Verginia, a plebeian girl betrothed to a former tribune, Lucius Icilius. She rejects Claudius' advances, enraging him, and he has her abducted. However, both Icilius and Verginia's father, famed centurion Lucius Verginius, are respected figures and Claudius is forced to legally defend his right to hold Verginia. At the Forum, Claudius threatens the assembly with violence and Verginius' supporters flee. Seeing that defeat is immanent, Verginius asks Claudius if he may speak to his daughter alone, to which Claudius agrees. However, Verginius stabs Verginia, determining that her death is the only way he can secure her freedom.
For the scene where Aaron tricks Titus into cutting off one of his hands, the primary source was probably an unnamed popular tale about a Moor's vengeance, published in various languages throughout the sixteenth century (an English version entered into the Stationers' Register in 1569 has not survived). In the story, a married noble man with two children chastises his Moorish servant, who vows revenge. The servant goes to the moated tower where the man's wife and children live, and rapes the wife. Her screams bring her husband, but the Moor pulls up the drawbridge before he can gain entry. The Moor then kills both children on the battlements in full view of the man. He pleads with the Moor that he will do anything to save his wife, and the Moor demands he cut off his nose. The man does so, but the Moor kills the wife anyway and the man dies of shock. The Moor then flings himself from the battlements to avoid punishment.
Shakespeare also drew on various sources for the names of many of his characters. For example, Titus could have been named after the Emperor Titus Flavius Vespasianus, who ruled Rome from 79 to 81. Jonathan Bate speculates that the name Andronicus could have come from Andronicus V Palaeologus, co-emperor of Byzantium from 1403 to 1407, but, since there is no reason to suppose that Shakespeare might have come across these emperors, it is more likely that he took the name from the story "Andronicus and the lion" in Antonio de Guevara's Epistolas familiares. That story involves a sadistic emperor named Titus who amused himself by throwing slaves to wild animals and watching them be slaughtered. However, when a slave called Andronicus is thrown to a lion, the lion lies down and embraces the man. The emperor demands to know what has happened, and Andronicus explains that he had once helped the lion by removing a thorn from its foot. Bate speculates that this story, with one character called Titus and another called Andronicus, could be why several contemporary references to the play are in the form Titus & ondronicus.
Geoffrey Bullough argues that Lucius' character arc (estrangement from his father followed by banishment followed by a glorious return to avenge his family honour) was probably based on Plutarch's Life of Coriolanus. As for Lucius' name, Frances Yates speculates that he may be named after Saint Lucius, who introduced Christianity into Britain. On the other hand, Jonathan Bate hypothesises that Lucius could be named after Lucius Junius Brutus, founder of the Roman Republic, arguing that "the man who led the people in their uprising was Lucius Junius Brutus. This is the role that Lucius fulfils in the play."
The name of Lavinia was probably taken from the mythological figure of Lavinia, daughter of Latinus, King of Latium, who, in Virgil's Aeneid, courts Aeneas as he attempts to settle his people in Latium. A.C. Hamilton speculates that the name of Tamora could have been based upon the historical figure of Tomyris, a violent and uncompromising Massagetae queen. Eugene M. Waith suggests that the name of Tamora's son, Alarbus, could have come from George Puttenham's The Arte of English Poesie (1589), which contains the line "the Roman prince did daunt/Wild Africans and the lawless Alarbes." G.K. Hunter has suggested Shakespeare may have taken Saturninus' name from Herodian's History of the Empire from the Death of Marcus, which features a jealous and violent tribune named Saturninus. On the other hand, Waith speculates that Shakespeare may have been thinking of an astrological theory which he could have seen in Guy Marchant's The Kalendayr of the shyppars (1503), which states that Saturnine men (i.e. men born under the influence of Saturn) are "false, envious and malicious."
Shakespeare most likely took the names of Bassianus, Marcus, Martius, Quintus, Cauis, Æmilius, Demetrius and Sempronius from Plutarch's Life of Scipio Africanus. Bassianus' name probably came from Lucius Septimius Bassianus, better known as Caracalla, a major figure in Africanus, who, like Bassianus in the play, fights with his brother over succession, one appealing to primogeniture and the other to popularity.
The ballad, the prose history, and the source debate
Any discussion of the sources of Titus Andronicus is complicated by the existence of two other versions of the story; a prose history and a ballad (both of which are anonymous and undated).
The first definite reference to the ballad "Titus Andronicus' Complaint" is an entry in the Stationers' Register by the printer John Danter on 6 February 1594, where the entry "A booke intitled a Noble Roman Historye of Tytus Andronicus" is immediately followed by "Entred also vnto him, the ballad thereof". The earliest surviving copy of the ballad is in Richard Johnson's The Golden Garland of Princely Pleasures and Delicate Delights (1620), but the date of its composition is unknown.
The prose was first published in chapbook form some time between 1736 and 1764 by Cluer Dicey under the title The History of Titus Andronicus, the Renowned Roman General (the ballad was also included in the chapbook), however it is believed to be much older than that. The copyright records from the Stationers' Register in Shakespeare's own lifetime provide some tenuous evidence regarding the dating of the prose. On 19 April 1602, the publisher Thomas Millington sold his share in the copyright of "A booke intitled a Noble Roman Historye of Tytus Andronicus" (which Danter had initially entered into the Register in 1594) to Thomas Pavier. The orthodox belief is that this entry refers to the play. However, the next version of the play to be published was for Edward White, in 1611, printed by Edward Allde, thus prompting the question of why Pavier never published the play despite owning the copyright for nine years. Joseph Quincy Adams, Jr. believes that the original Danter entry in 1594 is not a reference to the play but to the prose, and the subsequent transferrals of copyright relate to the prose, not the play, thus explaining why Pavier never published the play. Similarly, W.W. Greg believes that all copyright to the play lapsed upon Danter's death in 1600, hence the 1602 transferral from Millington to Pavier was illegitimate unless it refers to something other than the play; i.e. the prose. Both scholars conclude that the evidence seems to imply the prose existed by early 1594 at the latest.
However, even if the prose was in existence by 1594, there is no solid evidence to suggest the order in which the play, ballad and prose were written and which served as source for which. Traditionally, the prose has been seen as the original, with the play derived from it, and the ballad derived from both play and prose. Adams Jr., for example, firmly believed in this order (prose-play-ballad) as did John Dover Wilson and Geoffrey Bullough. This theory is by no means universally accepted however. For example, Ralph M. Sargent agrees with Adams and Bullough that the prose was the source of the play, but he argues that the poem was also a source of the play (prose-ballad-play). On the other hand, Marco Mincoff rejects both theories, arguing instead that the play came first, and served as a source for both the ballad and the prose (play-ballad-prose). G. Harold Metz felt that Mincoff was incorrect and reasserted the primacy of the prose-play-ballad sequence. G.K. Hunter however, believes that Adams, Dover Wilson, Bullough, Sargent, Mincoff and Metz were all wrong, and the play was the source for the prose, with both serving as sources for the ballad (play-prose-ballad). In his 1984 edition of the play for The Oxford Shakespeare, Eugene M. Waith rejects Hunter's theory and supports the original prose-play-ballad sequence. On the other hand, in his 1995 edition for the Arden Shakespeare 3rd Series, Jonathan Bate favours Mincoff's theory of play-ballad-prose. In the introduction to the 2001 edition of the play for the Penguin Shakespeare (edited by Sonia Massai), Jacques Berthoud agrees with Waith and settles on the initial prose-play-ballad sequence. In his 2006 revised edition for the New Cambridge Shakespeare, Alan Hughes also argues for the original prose-play-ballad theory, but hypothesizes that the source for the ballad was exclusively the prose, not the play.
Ultimately, there is no overriding critical consensus on the issue of the order in which the play, prose and ballad were written, with the only tentative agreement being that all three were probably in existence by 1594 at the latest.