No, the graph for child was never, AFAIK, used to represent a man – I’ve seen no examples of this in the oracle bone record or early bronzes; and there’s no evidence I can think of to support an original reading of or even resembling “love” either – not until the Shijing, at least. Still, given that earliest attested usage of 好 hao3 (as a personal name in the Shang dynasty archives) is clearly by phonetic loan, and that Shijing is fairly early, an original meaning of love is not impossible. But it sounds like random speculation to me.
The chief modern meaning of 好, ‘good’, btw, is thought to be a later semantic extension from that given in Shuōwén, where 好 is defined as 媄 mĕi, which 許慎 Xŭ Shèn defines as 色好也 sè hao3 yĕ, ‘of good appearance, beautiful’. This meaning might have been acquired by extension or phonetic loan – we have no evidence that it is the original meaning, and the structure does not necessarily support such a presumption.
段育裁 Duàn Yùcái, in his Shuōwén commentary, noted “好, 本謂女子, 引申為凡美之偁” (=稱), i.e., this originally meant ‘girl’ and was extended to mean ‘beautiful’. This is possible, given that 1) the most notable personal name in the OB involving this graph was that of a royal concubine or queen. and 2) an extension from a 女-related graph to ‘beautiful’ is at least plausible. However, again, I’m not sure there’s actually any evidence the graph originally meant ‘woman’; Duan appears to have been speculating, based on the 女 element in the graph. Remember, he didn’t have access to the oracle bones. We have better evidence now than he did then.
In sum, no one knows what 好 originally meant, and none of the theories put forth by scholars are compelling. There is currently no scholarly consensus on this one. It is most likely a picto-phonetic compound, IMO, but possibly a huiyi compound. Either way, it’s likely that only its extended or phonetic loan meanings now survive.
