I am interested in account of languages that had undergone a change in the writing direction somewhere in the history.
We might say, for example, that Greek was used to be written also (not sure if that was the dominant way) leftward some 2500 years ago.
languages that changed their script system(i.e adopted different set of letters) and thus change the direction of writing would also qualify.
Thanks
Really lots of languages have experienced a change from Arabic writing (right to left) to either Latin or Cyrillic writing (left to right) during the 20th century. Notable examples are Turkish, Azeri, Kazakh, Uzbek, Kyrgyz, Tajik, and Haussa.
Mongolian switched from vertical Mongolic writing to left to right Cyrillic at the same time.
Chinese is classically written top-to-bottom from right to left, e.g.,
9 5 1
10 6 2
11 7 3
12 8 4
but it is becoming increasingly common to also see it written left-to-right from top to bottom, e.g.,
1 2 3 4
5 6 7 8
9 10 11 12
Also, Vietnamese was originally written with Chinese-based characters (the Chữ-nôm (𡨸喃)), but has since switched to a Latin-based script (the Quốc Ngữ).
Probably the oldest example — being one of the oldest known examples of writing to begin with — is Sumerian cuneiform writing.
Like Chinese, Sumerian cuneiform was originally written in vertical columns from top down and right to left, but sometime around 2000 BCE the writing rotated 90 degrees counterclockwise and began to be written in horizontal lines from left to right and top down, just like we write English today.
Unlike with modern Chinese, we can't blame this shift on the influence of the Latin script, since that wasn't invented for a few more millennia yet. The only other writing systems in widespread use anywhere nearby were Egyptian hieroglyphs and the related hieratic script, which I guess might have been an influence, since contact between the Sumerian and Egyptian civilizations certainly did exist despite the distance and the inconveniently placed desert in between. However, that's just wild speculation; no definitive evidence on the cause or causes of the cuneiform writing direction change, and whether it was a local innovation or influenced by other writing systems, is currently known.
Left: Cuneiform inscription written in vertical columns on a statue of Gudea, ruler of Lagash, c. 2120 BCE. Detail from photo on Wikimedia Commons taken and released into the public domain by Marie-Lan Nguyen. Right: (Mostly) horizontal cuneiform inscription on a statue of king Idrimi, c. 1600 BCE. Detail from photo by author.
Also unlike Chinese, when the direction of cuneiform writing chanced, the signs rotated with it. A peculiar consequence of this is that the change is unobservable in texts written on clay tablets (which comprise the majority of surviving cuneiform texts) since they could be held equally well in either orientation*. The only place where it can be seen is in inscriptions carved on walls and statues and other objects with a definite up–down orientation. This also makes dating the change more difficult, not only due to the limited evidence, but also due to the fact that monumental inscriptions were often written in a deliberately archaic style, and quite likely preserved the old vertical writing order long after it had otherwise fallen out of favor.
Also, the change was obviously gradual, and both writing directions coexisted for a long time. Indeed the vertical writing style never quite entirely died out (until cuneiform writing as a whole did), especially in archaizing contexts or where vertical text just happened to fit better. Sometimes both could even be found together, as in this illustrated stone tablet (BM 90922) from c. 875–850 BCE, where the main body of text below the image is written horizontally (as usual by then), but the names of the figures above are written vertically next to each of them.
Ps. Having practiced cuneiform writing a bit myself, one relevant observation I've made is that, at least for me, the most natural and convenient way to hold a tablet while writing on it is actually at a roughly 30° to 60° angle between horizontal and vertical, so that one ends up writing diagonally from top left to down and right(!). Also, especially when writing on a small hand-held tablet with an old Sumerian style narrow stylus, where the same edge of the stylus is used for both vertical and horizontal wedges, it's often convenient to rotate not just the stylus but also the tablet when switching between wedge directions (which one generally needs to do at least once per sign, if not more). So, regardless of the orientation in which the tablets were read, I'm pretty sure that the actual direction in which they were written could vary considerably even during the writing process.
*) Clay tablets could have seal impressions with pictures on them, but the seals were not always applied in a consistent orientation either, limiting their value as evidence of the intended reading direction. Actual drawings on clay tablets are relatively uncommon, and most that I know of tend to be schematic in nature, e.g. maps or geometric figures that also have no obvious intrinsic orientation.
Another example is the Libyco-berber alphabet that was written in many directions (top-down, right-left by imitating Phoenician, boustrophedon).
The (neo-)Tifinagh that is a descendant of this alphabet is written now from left to right.
Note: Which script(s) a language has used is largely historical accident. That is to say, scripts often have very little to connect them to a language except a history of usage. Some scripts are designed specifically for a language at a specific point in time, but languages change over time and scripts change much slower, if at all, causing scripts to gradually become less suitable as time goes on. Modernising a script makes prior writing harder to understand, and not modernising complicates contemporary spelling. Changing a script wholesale makes prior writing inaccessible except by additional education or transcribing from one script to another.
It is generally, but not universally, believed that Brahmi script (from the Indian subcontinent and surrounds) descended from Aramaic script (related to Hebrew and Arabic scripts) although the Indic spoken languages are not related to the Semitic ones. All the modern languages written using a descendant of Brahmi are written left-to-right, but Aramaic, like its Hebrew and Arabic counterparts, is written right-to-left.
If this connection is true, Brahmi was originally written right-to-left and later changed to left-to-right before splitting into its descendants.
From Wikipedia:
Brahmi is usually written from left to right, as in the case of its descendants. However, an early coin found in Eran is inscribed with Brahmi running from right to left, as in Aramaic. Several other instances of variation in the writing direction are known, though directional instability is fairly common in ancient writing systems.
Mongolian script is also descended from Aramaic, so at least one of its ancestors was originally written right-to-left. Under the influence of the neighbouring vertical Chinese script, Mongolian script was rotated, making it the only script to be written top-to-bottom, left-to-right. All other top-to-bottom scripts are written right-to-left.
From Wikipedia:
Mongolian is written vertically. The Uyghur script and its descendants — Mongolian, Oirat Clear, Manchu, and Buryat — are the only vertical scripts written from left to right. This developed because the Uyghurs rotated their Sogdian-derived script, originally written right to left, 90 degrees counterclockwise to emulate Chinese writing, but without changing the relative orientation of the letters.
Turkic languages changed their writing direction from old turkic right-to-left to modern latin/cyrillic left-to-right.
Hungarians used a script similar to runes, it was designed to be carved into wood. It was written/carved right-to-left. It was replaced by the Latin alphabet about 1000 years ago, but it persisted in isolated communities to the 16-17th centuries.
Today, the old "runic" alphabet is no longer in common use, but some enthusiasts learn it and the Scout Movement teaches it, and it might occasionally show up at places of historical or touristic interest.
Old Persian cuneiform was written left-to-right. Pahlavi and Avestan were right-to-left, as is modern Persian. On the web it is reasonably common to see Persian written with latin characters (left-to-right), and modern Tajik (which is another descendant of Old Persian) is nowadays written with Cyrillic characters, i.e. also left-to-right.
Edit: Sogdian, which is another descendant of Old Persion, did use a vertical writing direction at some point, similar to the Mongolian script still in use today.
Modern Malay* (Brunei, Indonesian and Malaysian) used to be written with a modified Arabic script but each separately adopted the Latin alphabet (Brunei and Malaysia from the English and Indonesia from the Dutch). This changes the writing from right-to-left to left-to-right.
The Arabic script is still being taught in schools to Muslim students in Malaysia (Arabic script switches from being necessary to answer exam questions for the subject of Islamic Studies to being optional - students can also answer in standard alphabet - depending on what the current education policy happens to be) but it is not commonly used in everyday life due to the fact that almost half the population cannot read the Arabic script.
The last Malay language newspaper in Malaysia to be printed in Arabic script ceased publication this year.
* Note: I mentioned modern Malay because old and archaic Malay used to have their own scripts before the mass adoption of Arabic script in the region. And yes, scripts - there were several different dialects/languages of Malay in the region that had native writing systems. The earliest evidence of Malay written in Arabic script dates back to the early 1300s.
