In this short article, Millard discusses the phenomenon of divergence in the meaning of cognate words in kindred languages. He begins with a modern example from British and American English, warning American visitors to London to always walk on the pavement. In British English, ‘pavement’ refers to the part of the street called the ‘sidewalk’ in American English, while ‘pavement’ in American English refers to the part where cars drive. Millard discusses cases where Aramaic shares cognates with Hebrew and/or Akkadian, but diverges in meaning from one or both languages.
Often Aramaic agrees with Hebrew, against Akkadian. For instance, Hebrew and Aramaic אמר, “to say”, but Akkadian amāru, “to see”. Another example is Hebrew and Aramaic ספר, “writing, book”, but Akkadian šipru, “task, message”. It is interesting to note that Ugaritic sometimes goes with Akkadian, ‘mr, “to see”, but sometimes with Hebrew, spr, “document”.
On the other hand, there are also cases where Aramaic shares meaning with the Akkadian cognate against Hebrew. Note that these are not loan-words, but true cognates (as best as we can tell). For instance, Aramaic סחר and Akkadian saḫāru, “to go around”. Hebrew, however, uses סבב more commonly to express “to go around”, reserving סחר for the more specialized use “to go around on business, to trade”.
There are many cases where Aramaic has its own word while Hebrew and Akkadian share cognates for the similar concept. For instance, Aramaic אתא, “to come”, in contrast to Hebrew בוא and Akkadian bā’u; Aramaic נחת, “to go down”, in contrast to ירד and arādu; Aramaic נפק, “to go out”, in contrast to יצא and aṣû, etc. Some of these Aramaic words have cognates elsewhere, such as nḥt in Ugaritic.
Lastly, there are words where each language goes its own way. For instance, Aramaic מלל, “to speak”, in contrast to Hebrew דבר and Akkadian qabû. In conclusion, Millard points out the importance not only of studying common terms and features between languages, but also the individual peculiarities of each language.