The Philadelphia and New York accents presumably descended from a common ancestor dialect in the nineteenth century, since both accents in the twentieth century uniquely demonstrated a high/ɔ/ vowel (creating a contrast between words like cot and caught) as well as a phonemic split of the short a vowel, /æ/ (making gas and gap, for example, have different vowels sounds). One important indicator of this is that Philadelphia's short a split appears to be a simplified variant of New York City's split. Unlike New York City English, however, most speakers of Philadelphia English have always used a rhotic accent (meaning that the r sound is never "dropped").
In the very late nineteenth century until the 1950s, Philadelphia accents shifted to have more features in common with the then-emerging (and now-common) regional accents of the American South and Midland, for example in fronting /oʊ/, raising /aʊ/, and even some reported weakening of /aɪ/. Philadelphians also began retreating from their longstanding New York City-like accent features after this point, and even further developed their own entirely unique phonological features.
Furthermore, higher-educated Philadelphians born in or since the last quarter of the twentieth century are now showing a process of dialect levelling towards General American features. This includes a remarkable regularity among this demographic in replacing the traditional Philadelphia /æ/ split with the General American tensing of /æ/ only before nasal consonants; this probably began around the time the first generation of this demographic attended college. As of today, "the most strongly supported generalization is that Philadelphia has moved away from its Southern heritage in favor of a Northern system, avoiding those forms that are most saliently associated with local phonology".
The vowels in Philadelphia speech have shown volatility across the last century, as Labov's research has identified changes affecting over half of the vowel phonemes.
THOUGHT vowel: A feature unique to Middle Atlantic speakers (including Philadelphians and New Yorkers) and southern New Englanders is the raising and diphthongizing of /ɔː/, as in THOUGHT, to [oə] or even higher [ʊə]. The raised variants often appear as diphthongs with a centering glide. As a result, Philadelphia is resistant to the cot–caught merger. Labov's research suggests that this pattern of raising is essentially complete in Philadelphia and seems no longer to be an active change.
Southeastern vowel fronting: One of the features that Philadelphia shares with dialects of the whole Southeastern United States (but absent from most New York accents) is the fronting of a variety of vowels. This includes /oʊ/ and /u/; the resulting allophones are around [ɜʊ] and [ʉu], respectively. Generally, greater degrees of fronting are heard when the vowels appear in "free" positions (i.e., without a following consonant) than in "checked" positions (i.e., with a following consonant). Fronting does not occur in the context of following liquids leading to a significant difference between, e.g., goat and goal. The fronting of /oʊ/ and /u/ is well established in Philadelphia, though cross-generational data show that it remains an active change. Fronted nuclei in /aʊ/ are well established in Philadelphia speech as in New York. More recent research has noted a tendency among the middle-aged and younger generation of Philadelphians to raise the vowel, resulting in [ɛɔ]. /ʊ/, the vowel in foot, is sometimes fronted though not to the degree seen with /oʊ/ and /u/.
Short-a split: As in New York and Baltimore accents, historical "short a" has split into two phonemes: lax /æ/ (as in bat) and tense /eə/ (as in bath). Their distribution in Philadelphia along with Baltimore, however, is different from that of New York City: for instance, the words mad (tense) and sad (lax) do not rhyme in Philadelphia or Baltimore, but do for New York City and most other English dialects. Not all Philadelphians have this feature and some are beginning to favor the more General Americantensing of short a only before nasals (especially under the influence of youth trends and higher education); in fact, as a general rule, native Philadelphians only consistently have this split system if their own parents are native Philadelphians. For more details on the Philadelphia, Baltimore, and New York systems see: /æ/ raising or click "show" below.
The Philadelphia/Baltimore short-a split compared to the New York City short-a split and General American /æ/ tensing
arable, arid, barrel, barren, carry, carrot, charity, clarity, Gary, Harry, Larry, marionette, maritime, marry, marriage, paragon, parent, parish, parody, parrot, etc.; this feature is determined by the presence or absence of the Mary-marry-merry merger
Alexander, answer, ant, band, can (the metal object), can't, clam, dance, family, ham, hamburger, hand, handy, man, manly, pants, plan, planning, ranch, sand, slant, tan, understand, etc.; in Philadelphia, began, ran, and swam alone remain lax
amity, animal, can (the verb), Canada, ceramic, gamut, hammer, janitor, manager, manner, Montana, panel, planet, profanity, salmon, Spanish, etc.
/b/, /d/, /dʒ/, /g/, /ʃ/, /v/, /z/, /ʒ/
add, agriculture, ash, badge, bag, bash, cab, cash, clad, crag, dad, drab, fad, flag, glad, grab, mad, magnet, plaid, rag, sad, sag, smash, tab, tadpole, tag, etc.; in NYC, this environment has a lot of variance and many exceptions to the rule; in Philadelphia, bad, mad, and glad alone become tense
The NYC, Philadelphia, and Baltimore dialects' rule of tensing /æ/ in certain closed-syllable environments also applies to words inflectionally derived from those closed-syllable /æ/ environments that now have an open-syllable /æ/. For example, in addition to pass being tense (according to the general rule), so are its open-syllable derivatives passing and passer-by, but not passive.
Mary–marry–merry three-way distinction: As in New York accents and most native English accents outside North America, there is a three-way distinction between Mary[ˈmeɹi]~[ˈmeəɹi], marry[ˈmæɹi], and merry[ˈmɛɹi]~[ˈmɜɹi]. However, in Philadelphia some older speakers have a merger (or close approximation) of /ɛ/ and /ʌ/ before /r/ (the furry–ferry merger), so that merry is merged instead with Murray (with both pronounced as something like [mʌɹi]). Labov, Ash, and Boberg (2006: 54) report that about one third of Philadelphia speakers have this merger, one third have a near-merger, and one third keep the two distinct. Relatedly, as in New York, many words like orange, Florida, and horrible have /ɑ/ before /r/ rather than the /ɔr/ used in many other American dialects (See: Historic "short o" before intervocalic r).
Canadian raising occurs for /aɪ/ (as in price) but not for /aʊ/ (as in mouth) (Labov, Ash, and Boberg 2006: 114–15, 237–38). Consequently, the diphthong in like may begin with a nucleus of mid or even higher position [lʌik], which distinguishes it from the diphthong in live[laɪv]. Canadian raising in Philadelphia occurs before voiceless consonants, and it is extended to occur before some voiced consonants as well, including intervocalic voiced stops as in tiger and spider. Fruehwald argues that /aɪ/ has actually undergone a phonemic split in Philadelphia as a result of Canadian raising. The raising of /aɪ/ is unusual as the innovators of this change are primarily male speakers while the other changes in progress are led primarily by females. The sociolinguistic evidence suggests this raising is a fairly recent addition to Philadelphia speech.
FLEECE, FACE, and DRESS vowels: Traditional Philadelphia speech shows lowered and/or laxed variants of /i/ were common: [ɪi]. The recent sociolinguistic evidence indicates a reversal of this trend such that the vowel is now commonly raised and fronted. This raising is heard primarily before consonants (e.g., eat). The Linguistic Atlas researchers recorded lax variants of /eɪ/ near [ɛɪ]. As with /i/, recent research suggests this trend is being reversed by raising and fronting of the vowel often to a position well beyond [e]. This raising occurs before consonants (e.g., paid); in word-final position (pay), /eɪ/ remains lowered and lax. Both of these can lead to nonstandard phonemic incidence (see "Phonemic incidence" section).
Labov's research has indicated a tendency toward lowering of the lax vowels /ɪ/ and /ɛ/. This pattern is not yet well established and is labeled by Labov as an "incipient" change.
Many Philadelphians use a rather high, back, and perhaps even rounded vowel for /ɑr/ as in START; something near [ɔ]. The so-called horse–hoarse merger takes place, and the merged vowel is typically mid to high back; it can be as high as [ʊɚ]. As noted in New York, these tendencies toward backing and raising of /ɑr/ and /ɔr/ may constitute a chain shift. The evidence suggests the movement of /ɑr/ began this shift, and this vowel is relatively stable today, while generational differences are heard in the shifting of /ɔr/.
/ɔɪ/, as in CHOICE may be more raised than in other dialects; sometimes it is as high as [ʊɪ].
/ʌ/, as in STRUT, may show raised and back variants. In some cases, the vowel is in the high, back corner of the vowel space near /u/. This is reportedly a recent development and is one more common among male speakers.
Philadelphia forms the core of the one fully rhotic major region of the traditional American East Coast. This area runs from Pennsylvania and southern New Jersey down to Delaware and northern Maryland, and remains fully r-pronouncing today.
Non-rhoticity (R-dropping) can be found in some areas of Philadelphia, however (presumably as a recent innovation after the nineteenth century) such as among working-class male speakers specifically from South Philadelphia, especially those born in the first half of the twentieth century and of Italian or Jewish descent. On the other side of the socioeconomic spectrum, non-rhoticity in speakers from the Philadelphia Main Line may be a result of wealthy families sending their children to expensive boarding schools in the United Kingdom up until the 1960s and thus acquiring a "Transatlantic" accent. Non-rhoticity is most prevalent among black Philadelphians, who largely do not demonstrate the regional speech features of Philadelphia English; instead, many black Philadelphians speak African American Vernacular English.
Consonant changes, especially reductions and lenitions, are very common in informal conversational speech, so that:
The sibilant /s/ is palatalized to [ʃ] (as in she) before /tr/. Thus, the word streets might be pronounced "shtreets" [ˈʃtɹits].
L-vocalization is quite pervasive in Philadelphia speech. Phonetically it may be realized as something like [o] or a velar or labio-velarglide, [ɰ] or [w], or the consonant may be deleted altogether. Among Philadelphians, as in other dialects, vocalization occurs quite frequently in word-final and pre-consonantal contexts (e.g., mill, milk). In a more unusual development, vocalization may also occur inter-vocalically in Philadelphia. This tendency is more common when /l/ appears following low vowels bearing primary word stress (e.g., hollow). This variable also shows some lexical conditioning, appearing, for example, with exceptionally high frequency in the pronunciation of the name of the city (Ash 1997). This, in part, leads to the stereotype of Philadelphia being pronounced as "Fluffya" or "Filelfia."
As in other areas, the interdental fricatives /θ/ and /ð/ are often realized as stops, [t] and [d] or affricates [tθ] and [dð] in Philadelphia speech. This variation appears to be a stable class-stratified feature with the non-fricative forms appearing more commonly in working class speech.
The yew–hew merger can be found, as in New York City, in which words like human and huge, which begin with an /hj/ cluster, the /h/ is commonly deleted giving [ˈjumən] and [judʒ].
Consonant clusterreductions, such as removing the "t" sound from consonant clusters, so that "mustard" sounds more like "mussard," or "soft" like "sawff."
On is traditionally pronounced /ɔːn/[oən], phonemically matching the South and Midland varieties of American English (and unlike most New York accents), thus rhyming with dawn rather than don. However, the Northern /ɒn/[ɑn] has also been reported.
The word water is commonly pronounced /ˈwʊtər/[ˈwʊɾəɹ] (with the first syllable rhyming with the word put, so that it sounds like "wooter" or "wooder"), rather than the more standard English /ˈwɔːtər/. This is considered by many to be the defining characteristic of a Philadelphia dialect, even among young Philadelphians.
The word towel is commonly pronounced /tæl/, like tal in the word tally.
Both long-e and long-a sounds may be shortened before /ɡ/. Eagle rhymes with giggle/ˈɪɡəl/ (as in "the Iggles"); league/lɪɡ/ rhymes with big ; vague and plague rhyme with peg (pronounced /vɛɡ/ and /plɛɡ/, respectively). For some Philadelphians, colleague and fatigue also have /ɪ/ (pronounced /ˈkɒlɪɡ/ and /fəˈtɪɡ/, respectively). However, these are words learned later, so many speakers use the more standard American /ˈkɒliːɡ/ and /fəˈtiːɡ/.
In words like gratitude, beautiful, attitude, Baltimore, and prostitute, the i may be pronounced with a long ee sound /i/, as in bee.
Many Philadelphians are known to use the expression "youse" both as second person plural and (rarely) second person singular pronoun, much like the mostly Southern / Western expression "y'all" or the Pittsburgh term, "yinz". "Youse" or "youse guys" is common in many working classNortheastern U.S. areas, though it is often associated with Philadelphia especially. However, unlike in other Northeastern U.S. areas, the Philadelphian pronunciation of "youse" reflects vowel reduction more often than not, frequently yielding /jəz/ and /jɪz/ ("yiz") rather than the stereotypical /juːz/ ("youse"). (ex: "Yiz want anything at the store?" "Yiz guys alright over there?"). Second person singular forms commonly are heard as /jə/ and /jɪ/.
A sandwich consisting of a long bread filled with lunch meat, cheese, and lettuce, onion and tomato, variously called a "sub" or "submarine sandwich" in other parts of the United States, is called a hoagie. Olive oil, rather than mayonnaise, is used as a topping, and "hot" or "sweet" peppers are used for spice. The term 'hoagie' originated in Philadelphia. A similar sandwich toasted in an oven or broiler is called a grinder.
Small chocolate or multi-colored confections sprinkled on ice cream and cake icing, elsewhere called sprinkles, are known as jimmies in the Philadelphia area, as well as in the Boston and Pittsburgh areas. (In Boston, only chocolate sprinkles are called jimmies.)
Another distinctively Philadelphian word is jawn. According to Dan Nosowitz, jawn "...is an all-purpose noun, a stand-in for inanimate objects, abstract concepts, events, places, individual people, and groups of people." 
Notable examples of native speakers
The following well-known Philadelphians represent a sampling of those who have exhibited a rhotic Philadelphia accent:
These speakers retain slight traces or elements of a rhotic Philadelphia accent:
Gloria Allred — "slightly nasal, Philadelphia-accented voice that can drip with sarcasm"
Kevin Bacon and Bruce Willis — "two native [Philadelphia] sons, Bruce Willis (Salem County, N.J.) and Kevin Bacon (Center City Philadelphia), who, at least in interviews early in their career, before accent reduction training kicked in, let their diphthong freak flags fly."
Jill Biden — "She exaggerates her Philadelphia suburbs accent, which is already pretty strong."
Movies and television shows set in the Philadelphia region generally make the mistake of giving the characters a working class New York dialect (specifically heard in films set in Philadelphia such as the Rocky series, Invincible, and A History of Violence). A contrary example is the character of Lynn Sear (played by Toni Collette) in The Sixth Sense, who speaks with an accurate Philadelphia dialect. In the film Sleepers, Kevin Bacon, a Philadelphia native, uses an exaggerated Philadelphia accent for the character of Sean Nokes.
The use of geographically inaccurate dialects is also true in movies and television programs set in Atlantic City or any other region of South Jersey; the characters often use a supposed "Joisey" dialect, when in reality that New York-influenced dialect for New Jersey natives is almost always exclusive to the extreme northeastern region of the state nearest New York City. An important factor here is that in the real world, "local" TV, political, and sports personalities in South Jersey and part of Central Jersey are culturally associated with Philadelphia, not New York City.
^Fruehwald, Josef (2013). "The Phonological Influence on Phonetic Change". Publicly Accessible University of Pennsylvania Dissertations. p. 48. "...the White Philadelphian dialect is spoken now by a numerical minority of all Philadelphians...."
^Labov et al., 2006, p. 173: "In NYC and the Mid-Atlantic region, short-a is split into a tense and lax class. There is reason to believe that the tense class /æh/ descends from the British /ah/ or 'broad-a' class."
^Labov, W. & Rosenfelder, I. & Fruehwald, J.(2013). One Hundred Years of Sound Change in Philadelphia: Linear Incrementation, Reversal, and Reanalysis. Language, 89(1), pp. 31, 49.
^ abLabov, W. & Rosenfelder, I. & Fruehwald, J.(2013). One Hundred Years of Sound Change in Philadelphia: Linear Incrementation, Reversal, and Reanalysis. Language, 89(1), p. 61.
^Labov, W. & Rosenfelder, I. & Fruehwald, J.(2013). One Hundred Years of Sound Change in Philadelphia: Linear Incrementation, Reversal, and Reanalysis. Language, 89(1), p. 55.
^ abHenderson, Anita (1996). "The Short 'a' Pattern of Philadelphia among African-American Speakers". University of Pennsylvania Working Papers in Linguistics: Vol. 3: Iss. 1, Article 10. p. 137-139.
^Fruehwald, Josef (2007). "The Spread of Raising". College Undergraduate Research Electronic Journal, University of Pennsylvania
^Gordon, Matthew (2004) "New York, Philadelphia and other Northern Cities" in Kortmann, Bernd & Schneider, Edgar W. (Eds.) A Handbook of Varieties of English: Volume 1: Phonology Walter de Gruyter ISBN3-11-017532-0 p. 290
^Labov, William (2008). "Mysteries of the substrate". In Miriam Meyerhoff and Naomi Nagy (eds.). Social Lives in Language Sociolinguistics and multilingual speech communities. Amsterdam/Philadelphia: John Benjamins. pp. 320.
^Verma, Mahendra K. (1998). Sociolinguistics, Language and Society. New Delhi: Sage. p. 94.
Hindle, Donald. (1980). The social and structural conditioning of phonetic variation. (Doctoral dissertation, University of Pennsylvania).
Kroch, Anthony. (1996). Dialect and style in the speech of upper class Philadelphia. In G. R. Guy, C. Feagin, D. Schiffrin, & J. Baugh (Eds.), Towards a social science of language: Papers in honor of William Labov (pp. 23–45). Amsterdam studies in the theory and history of linguistic science (Series 4). Amsterdam: John Benjamins.
Kurath, Hans; & McDavid, Raven I., Jr. (1961). The pronunciation of English in the Atlantic states. Ann Arbor: University of Michigan Press.
Labov, William. (1980). The social origins of sound change. In W. Labov (Ed.), Locating language in time and space (pp. 251–265). Qualitative analyses of linguistic structure (No. 1). New York: Academic.
Labov, William. (1989). Exact description of the speech community: Short a in Philadelphia. In R. W. Fasold & D. Schiffrin (Eds.), Language change and variation (pp. 1–57). Amsterdam studies in the theory and history of linguistic science (Series 4), Current issues in linguistic theory (No. 52). Amsterdam: John Bengamins.
Labov, William. (1994). Principles of linguistic change: Internal factors (Vol. 1). Language in society (no. 20). Oxford: Blackwell.
Labov, William. (2001). Principles of linguistic change: Social factors (Vol. 2). Language in society (no. 29). Oxford: Blackwell.
Labov, William; Karen, Mark; & Miller, Corey. (1991). Near-mergers and the suspension of phonemic contrast. Language Variation and Change, 3, 33–74.
Labov, William; & Ash, Sharon. (1997). Understanding Birmingham. In C. Bernstein, T. Nunnally, & R. Sabino (Eds.), Language variety in the South revisited (pp. 508–573). Tuscaloosa: University of Alabama Press.
Labov, William, Sharon Ash, and Charles Boberg (2006). The Atlas of North American English. Berlin: Mouton de Gruyter. ISBN3-11-016746-8.CS1 maint: Multiple names: authors list (link)
Payne, Arvilla. (1980). Factors controlling the acquisition of the Philadelphia dialect by out-of-state children. In W. Labov (Ed.), Locating language in time and space (pp. 143–178). Orlando: Academic.
Roberts, Julie. (1997). Hitting a moving target: Acquisition of sound change in progress by Philadelphia children. Language Variation and Change, 9, 249–266.
Thomas, Erik R. (2001). An acoustic analysis of vowel variation in New World English. Publication of the American Dialect Society (No. 85). Duke University Press for the American Dialect Society.
Tucker, Whitney R. (1944). Notes on the Philadelphia dialect. American Speech, 19, 39–42.
Walt Wolfram and Ben Ward, editors (2006). American Voices: How Dialects Differ from Coast to Coast. Malden, MA: Blackwell Publishing.